Tom, I’m happy to receive your constructive criticism, but I’m disappointed you didn’t fix any of the factual errors we alerted you to via email before you launched this video. Examples: 23:42 You cherry-picked this quote to make it seem like the NTSB blamed automation for the crash, when the report focuses squarely on human error: “The National Transportation Safety Board (NTSB) determined that the probable cause of this accident was the flight crew’s mismanagement of the airplane’s descent during the visual approach, the Pilot Flying’s unintended deactivation of automatic airspeed control, the flight crew’s inadequate monitoring of airspeed, and the flight crew’s delayed execution of a go-around after they became aware that the airplane was below acceptable glidepath and airspeed tolerances.” 32:37 Self-driving cars have maps including traffic control so they would know where stop signs are meant to be even if road markings aren’t there or stop signs are obscured. Plus they have better obstacle detection and avoidance than human drivers. 39:16 I’m not saying rare accidents don’t happen, I’m saying they happen less often than common accidents, many of which could be prevented by self-driving cars. I sent you an academic paper that recreated in simulation 72 real-world fatal accidents that occurred in the area where Waymo operates. In almost all cases the accident was avoided or mitigated by the Waymo driver. Why did you omit this study? 47:03 It’s well understood that autonomous cars properly coordinated could reduce traffic because they don’t have the same reaction time delays as humans. For example all cars at an intersection could start moving together instead of one at a time as we currently do. 47:10 We don’t have to increase the car utilization rate to 100% to reclaim significant value. If cars were parked 90% of the time instead of 95%+ we would only need half as many vehicles. Isn’t it ironic that a video purporting to call out misinformation itself contains so many distortions and factual errors? (Which we pointed out in advance but you didn’t feel compelled to fix) On the issues themselves, I like public transport. I also ride a bike, and enjoy walking to get around when it’s practical. But cars will be a part of the transport mix for the foreseeable future. And it’s my opinion, based on the evidence, that roads will be safer the more cars are driven by computers than humans. No one has to pay me to tell you that.
Hiya, As I said in my reply to the email from your team, my decision to make this video came from a place of disappointment rather than animosity. As such, it’s a real shame to see you again coming out swinging in response to my points rather than the video giving you any pause for thought surrounding your relationships with sponsors. I’m glad to see that you've taken on board my critiques of your initial rebuttals to this video and attempted to strengthen them. I imagine you know the flaws in these responses to what you mistakenly call "factual errors" but I'll do you the courtesy of replying to them anyway. 23:42 You accuse me of cherry-picking here. However, the very next sentence from that which you quote states that 'Contributing to the accident were (1) the complexities of the autothrottle and autopilot flight director systems that were inadequately described in Boeing’s documentation and Asiana’s pilot training, which increased the likelihood of mode error [...]'. The reality is that, as in your original video, you're working on the false presumption that humans and automated/autonomous technologies are in conflict with one another. However, as is the case here, such technologies are always going to involve human interaction and they therefore need to be designed in ways that ensure that those interactions don't lead to accidents (as the Acting Chairman of the NTSB states in the quote included in my video). You use the crash of Flight 214 as an example of human failure with the implication that automated systems could have avoided the crash; it is disingenuous and distasteful to have eliminated all mention of autothrottle from your version of events just because it didn't suit the argument you wanted to make. 32:37 It is, again, completely disingenuous to refuse to mention the maps which Waymo vehicles rely on when it suits your argument during your initial video but to now hold them up as a vital part of the technology. You spoke as though those maps didn’t exist in the video and so it was only fair, in critiquing your video, to respond in kind. A further point for consideration here is the extent to which creating maps/scans for the entirety of even the United States (let alone other countries too) is practical. That seems like a highly intensive task which would be very costly, possibly to the point it’s unworkable. 39:16 Two points here: Firstly, you do not say in your video at all that the kinds of accidents considered when people discuss the ethical implications of driverless cars “could be prevented by self-driving cars”-you say they’re rare and quickly move on. Secondly, the “academic paper” your team sent over was a write up of a study produced for Waymo and written by Waymo staff. I can’t seem to find any instance of it being published in a reputable journal (or a journal at all for that matter) or of it being subject to peer review. It is also, as one might expect, based entirely upon accidents which occured in the few suburbs of Phoenix where Waymo cars presently operate (and have had years of “training” in) and so it would be improper to suggest that its findings would be representative of a wider roll out to locations that Waymo’s cars have not been fine-tuned for or where conditions differ. 47:03 While they might have an impact, the idea that these tiny bits of time saved here and there would eliminate congestion is optimistic. The more extreme version of this (included in the CGP Grey video you’re referencing) in which all autonomous vehicles “talk” to one another (or even just to the traffic lights) would require such a level of implementation and standardisation that it is unlikely even in the medium term to the point where it’s a fun thought experiment but little more. 47:10 I make no such claim that we would need to “increase the car utilization rate to 100% to reclaim significant value” in my video. I merely state that the use of the 95% figure is misleading as it ignores several important caveats (like the existence of the nighttime). Even a figure which related to the percentage of the daytime that cars go unused would have been more useful to use here-it may not have given you such an impressive figure but it would have been a fairer representation of the reality. To reiterate, suggesting that my video contains “distortions” and “factual errors” is deeply disingenuous. Nevertheless, I know that, whatever the circumstances, someone suggesting that you’ve acted improperly is a horrible experience. I hope that, with the passage of time, you’ll take on board some of the thoughts provided here about your relationships with sponsors. Best wishes, Tom
@@RazorbackPT UA-cam only allows you to have one pin on a video. Tom did make this a community post, and tweeted it out. That's more than enough due diligence imo. It's not like Veritasium is gonna link this video in a pinned comment on his original. Pinning responses not a reasonable expectation.
I would like to think that my channel is one that promotes the solutions you were getting at near the end of this video: public transportation, cycling, and walkable cities. I worked my whole career in tech, but I'm a firm skeptic of self-driving cars. One of my biggest issues with them is that a lot of the problems they aim to solve are "American" issues, that stem from decades of car-dependent cities. Yes, if you design a city such that everyone needs to drive everywhere all the time, you have problems of crippling traffic, high transportation costs, and dangerous roads. None of these issues really exist in walkable cities with high quality public transport, however. So I see driverless cars as a highly speculative and expensive solution to a problem that really shouldn't exist in the first place. Not to mention the potentially seriously negative implications of having city criss-crosses with high-speed, high-traffic roads full of autonomous vehicles. To me, self-driving cars really seem like a desperate attempt to maintain car-dependent places and car-centric suburbia with a flashy technical gadget, rather than an actual solution to the problems facing cities.
I'm guessing you're not sponsored by people who also promote public transportation, cycling and walkable cities, which I think was Tom Nicholas' point here. It's a good solution that people would and should support, but they don't because the private sector isn't going to throw their weight behind something that doesn't make them money. Edit: This came off as a little antagonistic. For the record, I'm not trying to dispute anything you've said in the slightest.
I think the point was less that nobody is willing to make videos on it, but that there is much less money and sponsorship opportunities in making them. As can be seen in comparing your channel to the likes of Wendover or Real engineering that rake in millions from sponsorship deals for praising car infrastructure and plane travel.
The whole vid I was thinking "pls comment on how cars (of any kind, be it electric and/or self-driving) is not the ONLY or even best option for our future" and he did :)
The single biggest difference between pilots and the average car driver is training. It is a LOT harder to get a pilot's license and is fairly expensive, compared to getting a driver's license.
I’m in an English class right now where a big part of it is finding and scrutinizing sources. Usually you want to look at both sides of an issue and then argue your point with the one that wins you over. That’s impossible when there is only one source, and it’s paying you to say things.
We shouldn't stop at "both" sides. There's often more than 2 options and stopping at that will make you lose sight of better solutions nobody is platforming, making the "middle ground" between those two options unfairly appealing. Plus it leaves any point where both sides agree basically uncontested, limiting everyone's decision making. In the case of self-driving cars, thinking of "yes vs no on self-driving cars" (which is an interesting question by itself) often loses sight of alternatives like public transport, getting rid of car-centric city planning, de-urbanization, etc. Even the really interesting point of accountability (who is accountable after the AI makes a choice in a trolley problem situation?) can be made a disservice if all sides just assume our justice systems don't need corrections. Lastly, even the pro self-driving car movement tends to make itself a disservice by limiting what the AI can do. Calling them self-driving cars makes most people lose sight of all of the non-car stuff that can be made more useful by moving itself autonomously on a 2D plane. Every one of those cases can, and should, have its own "yes vs no" crowd.
This is great and needed to be said. I worked with Gates Foundation. Don’t think it quite fits into this narrative, but I would say that I wouldn’t work them again. Full integrations are a signal you shouldn’t trust a channel. I see some multi millionaire creators use the excuse of “this helps us fund future videos” as an excuse. That’s bullshit. Do sponsorships fund content. Yes. Do you need to sacrifice your integrity to work with a sponsor? No
Gates foundation 100% fits into this narrative, it's one of the champions of sponsored propaganda. And I know "propaganda" has a very harsh conotation, but it is, at the end of the day, what the videos sponsored by Gates Foundation are, whether you agree with it or not, propaganda is propaganda... I was throughly dissapointed in Kurzgesagt...
Thank you, that means a lot. As I said to a few people in the chat during the premiere, making these videos comes from a place of disappointment rather than animosity. I continue to believe that there's real potential (much of it already realised) in educational UA-cam (and the broader free dissemination of education online, however the costs involved in dethroning UA-cam means that it's unlikely to be topped anytime soon). To make their points sufficiently, these videos do kind of have to focus on a case study, however I genuinely wish the team at Veritasium well where they're not engaging in these practices. With the full disclosure that I've been looking at making a video about Gates' use of sponsorships to shape the discourse on YT surrounding responses to climate change, that's interesting to hear about the Gates Foundation. I know Our Changing Climate has said similarly.
I love your videos and respect you greatly (even in admitting you wouldn't work with the Gates foundation again.) Do you think it might be a good idea to even give this information to your viewers in maybe a pinned comment under the video? I think a shift in perspective like this on sponsored content is definitely worth informing viewers about.
@@Tom_Nicholas my complaint about Gates Foundation is primarily that there are some shady things the foundation has done that I was not aware of. Secondary problems were primarily about how unresponsive and inflexible they were. The video I made was already written when they approached me and I didn’t change the script to fit with them. It was a natural fit for a sponsorship, and yet they still somehow made things difficult. I put that down to them not understanding how influencer marketing is supposed to work.
"people are using youtube for education" this is true, and a large part of the problem with removing the dislike count. removing the dislike count is anti-information/education, pro-propaganda.
There is a really good reason why wikipedia has been so defensive about it's policy to never take corporate sponsorships. The day that changes is the day we are all truly doomed.
Tom, thanks for this video. I agree with many of the points you raised (I’m bearish on autonomy as a whole (and highlighted as much in my own video)) and while I very much respect Derek and his team, I took issue with the way in which he rolled out his video. You postulated at around 25:45 my shot at Tesla was drawn from the briefing-it was not-I’m a Tesla owner with extensive Autopilot miles and frustrations from Musk’s promises. Regardless, I was actually surprised Muller’s video stated nearly exactly the same. My video was published a number of weeks before his but I’m not saying he plagiarized or anything. Likely a weird coincidence. I’m certainly not without blame for my own video and if given to travel back in time, I wouldn’t have done the spot with Waymo. I was frustrated with the amount of editorial correction the brand attempted to make to my original script (I think we went through four edits) and it was frankly neutered of my opinions and dissent. Despite much push-back (and a fair number of changes they requested I outright refused to make), the script *was* largely edited and in retrospect I wish I had been more firm. The large majority (90%) of ads I do on my channel are integrations like yours on this video. The remaining 10% are dedicated ads and I do my absolute best to retain editorial control so that the videos can still be interesting and informative (one example I’m very proud of is our video tour of a fiber internet company’s infrastructure) while being very explicit and clear with viewers in the first few seconds that the video is, ultimately, and ad and thus inherently made with the brand’s input (this is an attempt for viewers to stop watching if they’re not interested in such content (I wouldn’t be myself)). Where I erred with Waymo is doing this type of video on such an ultimately controversial topic and one for which I have many of my own opinions I was unable to voice due to more heavy-handed editorial control than I was led to believe based on our contract. I should have fought harder to keep my dissenting opinions in the script. Frankly, I just got really sick of their requests, I have an incredibly small team (just 2 of us) and it was taking too much of our time and I was ready to write the video off as a loss and just rip it off like a Band-Aid but I regret it. I should have done better and I will in the future. This type of content is always tricky so I have (and will continue) to discourage it whenever possible. Thanks for your insight and wise words. Keep on keeping on!
Amazing to see your thoughtful response, in stark contrast to Vertasium doubling-down, acting defensive, attacking small details of the video meanwhile completely ignoring the main argument. Much respect Snazzy!
By you having said that, I in no way think less of your channel, which I love, or your person. I think the way a humans mind is made up, I now might even like you more because of admitting to a mistake. This makes it even harder to understand the aftergoing actions of the Veritasium team, which try their hardest to intimidate, just like a corporate lawyer would.
17:50 i love how the poll itself is slanted like "i can't wait!" and is not mutually exclusive with "they're not ready now". also, it (along with "they're terrifying") makes the poll about your personal excitement and not about your honest assessment of the tech. i'd love to see a poll where the options were neutral and more specific, ie "i think they're ready to be used now", "they will never be ready", and "they will be ready in >10 years". i imagine they may not be as optimistic.
As you know Tom, I really appreciate your efforts to draw attention to this growing trend of educational UA-camrs allowing their trust and audience to be leveraged by corporate agendas, perhaps even unaware themselves that they are being used in this way. Many of us esp in the sciencey space have discussed this privately and feel full integration is not a good idea, but I’m not convinced the audience really sees the problem, suggesting they simply believed their favourite UA-camr’s educational video about Bill Gates, the WEF or Waymo
The nefarious Bill Gates has been bankrolling some of the creators on the platform, and that’s not cool considering how much influence he already has bought himself in mainstream outlets. More needs to be done to counterweight his heavy spending.
Popping up here after your appearance with Tom Scott. Who shall be your next UA-cam Tom? More seriously though, it's kind of heartening (apologies for the pun) to know that influencers are discussing this behind the scenes if nothing else. I know the dance to keep food on the table as a content creator is always hard, and so a lot of the potentially sponsor scaring talk has to happen in private, but knowing it's happening at all is a relief.
Really appreciate how you put this, Tom. Creators have discussed this a bit in private, and I think you handled it really delicately and professionally in bringing it to the public eye.
I love how cute and sinister “discussed this… …in private” sounds. Almost as something from “what we do in the shadows”. Council of high youtubers in special robes summoned for discussion of highest matters. On serious note - every creative community should have healthy and reasonable discussion of limits and representation of commercial interests for reasons of clarity and objectivity.
I agree Tim! This corporate influence really worries me. It's time to set some harder boundaries, especially for those who consider themselves 'educators'.
@The Program _Someone_ is always responsible for the finances needed to make a video, even if it is just the person in front of the camera and their personal wealth. The problem with corporate sponsorship is that it is both pervasive and without care for the media it co-opts. You are correct that other groups can have a similar destructive influence on the reliability of the medium, but each type of organisation has a different way of instilling bias. A non-profit or a university (and some governements) will often have the interest of the general public in mind whereas activist groups rarely have the power to move over media creators that don't already align with them.
The methods Tom Nicholas uses here to draw up scepticism and doubt reminds me a lot of Tucker Carlson on FOX. Injecting their own interpretation of state of mind and intent to something somebody said/did, and then criticising the morality of that intent. Classic straw man argument. Tom's dive into Derek's comment about planes landing was an especially egregious example. Automation reduces opportunities for human error. That's all. That's the message of that segment in Derek's video. Derek didn't say the pilots were "too egotistical", or "autonomy is good, and humans bad", as Tom puts it. It's just rhetoric Tom uses to put words into Derek's mouth in order to invoke emotions of disgust in order to solicit agreement from viewers when he rejects moral outrage he created himself. You see this technique used a lot by politicians. "So-and-so did ABC! The implication is he thinks you're idiots! Well, I don't think you're idiots! You're all the best!". This video is pretty much just Tom tearing down his own straw men. Then, peppering the viewer with suggestive questions, along the lines of "Can we trust ABC if they didn't do XYZ? I don't think I could if they didn't! Why did they not do XYZ? Was it because they're trying to hide something dubious? We will never know". I can see how Tucker Carlson has such a following with these techniques. With that said, Tom is very *VERY* good at rhetoric. His essay-writing skills must be off the charts.
@@benc5528 yeah, it takes real mental gymnastics to go from "guy is paid to advertise automation technology" to "guy wants to make automation technology sound good"
@@InsecureCreator People need to understand this. Gather enough correlations and it becomes trivial to fill in the blanks. And even if you deliberately exclude certain types of data, those can be reconstructed through those correlations - and that is exactly what happens, automatically, in many machine learning algorithms. This problem has been documented already in the field of automated human resources (hiring and CV screening).
@@jimcrelm9478 “no no, we don’t want to only hire men, stop that computer”. “Okay, I’ll just teach myself how to infer gender from names if I can’t use the gender field directly! I’m so clever huh?”
Derek was not talking much in that at all, it was the other people most people will find it odd and the more important thing was if your relative giving DNA to use for investigation purposes will infer on your privacy
As Tom Scott once put it on his own channel: "You cannot trust me". Not because he has ever maliciously spread incorrect information but because he cannot ever guarantee that the information he provided is really correct and that he didn't make a mistake. He's trying his best to avoid it but he realized that there is no way he can avoid it all the time. It's novel approach to tell your audience: Keep in mind, I'm human and I make mistakes, so take everything I say with a grain of salt.
its a somewhat novel aproach, but it shouldn't be. I think its the only way foward, be it from a youtube channel or a big newspaper. Brutal honesty, near-suicidal transparency, and admission of fallibility and bias.
Thank you for this video Tom. As someone who has worked in media through the transition from print to online models and have watched the surprisingly fast creep of sponsored content making its way into what is presented as real content, it’s frankly shocking to me that audiences don’t understand the problem here. Trust is a valuable resource in media and to see it sold so easily is a shame. Anybody who can see the problem with the current distrust people have with media, anybody who gets frustrated trying to convince that relative of theirs that objective reality exists and facts are not a matter of opinion, anyone who wants to be able to agree upon a shared reality so that we can make meaningful, science-based decisions as a society, needs to understand the importance of trusted media sources free from bias. If people haven’t noticed, we are currently in a battle to save the very concept of shared reality and truth, and it’s not clear which side is going to win here. I would have thought @veritasium is among the people who understand the value of truth being backed by integrity. It’s sad to see him discard the decade of trust he’s built up so readily and harm the state of discourse and trust in media in the process.
My cooking hero over here dropping pure wisdom. Love it. Hopefully those of us that value a shared reality won’t lose this one, I am optimistic when I see leaders like you advocating for the good fight.
Thank you, I really appreciate you saying so. It does always shock me how different the perspectives on this are from those with more of a foot in “traditional” media and those with more of a foot in online media. I can’t remember which US website unionised a while back but it was interesting to see that one of their demands (alongside better pay and conditions) was the right to refuse to work on advertorial pieces. There was a real sense that the writers felt that such work made them feel uneasy and diminished their sense of professional self-worth. By contrast, many writers/video creators/etc. who haven’t had experience in that world seem to jump with enthusiasm into working in this way. Also, love The Food Lab. The Ultra Gooey Mac & Cheese is a regular in our house!
On UA-cam propaganda, could you also do a video in the series about VPN providers? They don't really provide protection and the companies paying for these ads are spreading misinformation. Seeing your favorite UA-camrs make an educational claim about VPNs that's wrong can have negative consequences too. For example, they only change an IP address, but they don't actually affect identification. That's done with behavioral and browser profiling. Your traffic still goes through an ISP, even if it's not your own. Commercial network connections can have a lot more tracking and fewer protections than your residential ISP. I appreciate your video, and this isn't really a criticism of it. It just seems to be in line with the problems your video shows.
The thrust of your point is absolutely right. But there is a complication in that some of these services are more than just a VPN alone, their apps also block tracker JS and other things on the web layer too. The marketing spiel glosses over all the complexity, because of course it does lol.
He did put “fluff” in the subtitles when the ad started, so he could(?) be aware? I think it’s understandable to do little ad bits that just tell you to do something like buy a product, but whole videos of uncritical support are the problem.
The juxtaposition of "most experienced driver in the world" with "can't drive in inclement weather" is just too good. Really is an apples to oranges comparison.
"What we call Man's power over Nature turns out to be a power exercised by some men over other men with Nature as its instrument." ― C.S. Lewis, The Abolition of Man
I love veritasium, but I noticed this sort of thing with him lately. I was more concerned with his video on DNA databases, since it was fairly one sided towards "this is a good thing, actually. Catching criminals trumps privacy rights" without much attention paid to the human rights issues of the subject, while the video was clearly meant to be objective and educational.
@Shimmy Shai "Best to let a thousand guilty go free than one innocent be put to death" It rather depends on how guilty and of what. Many wish Herr Schickelgruber had not be set free.
@Shimmy Shai Uh, did you read your comment before posting? "Best to let a thousand murderers go free than to put 1 innocent to death." No, that's why America is such a terrible place with almost nothing but criminals. You're kinda dumb if you think like that imo.
Just watching this video now. I wanted to point out that there was a Veritasium video in the past that was sponsored by a paper towel company, and as part of the advertisement the host creates an "experiment" to measure how much bacteria are deposited when reusing a cloth towel versus a paper towel. The conclusion of this advertisement/experiment was that paper towels are safer because they don't leave behind bacteria. As a proponent of zero waste practices I found the whole thing to be rather disgusting, because there was no discussion of environmental impact, and it was disingenuous because I use cloth rags, but I keep a pile of clean ones and I can always grab a fresh one if needed. This advertisement was worse than just saying "buy these paper towels" because it created a rigged experiment by a trusted science youtuber, giving much more weight to these claims which were not properly discussed. So I am really glad you are digging in to this.
There was another one about dandruffs sponsored by that one shampoo company too. Created quite a stir, though I don't remember feeling the content being too dishonest at the time I watched it
@@moscanaveia yeah but they normally feel fine when you're watching it, especially if you're not normally exposed to the subject. they're well presented and follow a superficially logical path. you can really feel that they reached these conclusions in a good way, which is exactly what makes this such effective and insidious propaganda. its only with further research and fact checking we find out. though for all i know, the dandruff science is legit, buuut I've not looked into it. and i may never look it up, and this might have never come up so I wouldn't even have thought about this again.
yeah the paper towl one was worse than the weymo video.. at least weymo is still developing its product... but the paper towl video sent the messsage that there was no alternative.. and didnt propose any better solutions or further required research
Another point in the "20M miles driven" argument is that comparing the distance traveled by Waymo cars to a human driver is completely pointless. I could write a program to make a car self driving and send it out to drive for 100M miles; if the program is incapable of learning and I don't update it in any way, it's exactly the same as when I sent it out having driven 0 miles. Waymo presumably does update their software regularly, and it may very well be capable of learning to some degree, but that doesn't change the point that saying "we've driven 20M miles, that's more than any human so we're more experienced than any human driver is" is completely ridiculous. The company/program simply does not learn in the same way a human being does.
Learning is a non-sequitor. Humans crash at a fairly predictable rate. And new inexperienced drivers are entering the road all of the time. Even experienced drivers get distracted or tired. So long as this program is consistently less crash-prone than an average human, that's an improvement! In addition self driving cars CAN 'learn' better than humans: The data of a crash can used to make the next version of the program better, compared to a human driver that may not survive the crash and thus won't have a next time to apply their knowledge.
Excellent point. 20M miles means something much different to a computer algorithm than it does to people, but we are expected to value them equally. False equivalency.
Additionally, I could write a program in an afternoon that could drive 100K miles way safer than the average human, so long as those 100K miles were on salt flats. It would be better to compare the accident rates of taxi drivers, since the composition of roads driven on is more similar to these cars.
When one hears that a person has a significant number of years of 'experience' doing something, there's an assumption that there's a not-easily-articulated nous that the person has developed. Using the word 'experience' for these cars is disingenuous.
"Technological fixes are not always undesirable or inadequate, but there is a danger that what is addressed is not the real problem but the problem in as far as it is amendable to technical solutions." (Engineering and the Problem of Moral Overload) "If machines produce everything we need, the outcome will depend on how things are distributed. Everyone can enjoy a life of luxurious leisure if the machine-produced wealth is shared, or most people can end up miserably poor if the machine-owners successfully lobby against wealth redistribution. So far, the trend seems to be toward the second option, with technology driving ever-increasing inequality." Stephen Hawking, 2015 Reddit AMA
@@Pluveus Only in capitalism would automation be a bad thing. It's astounding. There are so many things I just can't get down with until we're rid of capitalism, like anarchism, automation, UBI and a host of other policies.
This was great. Well done, Tom. As someone who grew my channel in the “science” niche before making my switch to political content, I’m very familiar with the full integration problems in that community. I’m glad to see someone taking the time to really discuss it.
Bold youtubers like yourself and Jake Tran really opened my eyes to the flaws of the Corporate-ocracy and the detrimental state of capitalism. I’m a tremendous fan, thank you for sharing such critical information. Thanks to you, I turned Socialist.
YES after watching the video on self driving cars I was shocked to not see a single word of criticism in the comments. I’m so glad you’re addressing this video!
@@internetdumbass It's a shame UA-cam makes it very hard to search the comments. In theory it is possible but only by loading all comments (which can take quite a while). I'm affraid this won't change anytime soon because YT has no real concurrents; as long as their website generates revenue, why would they improve it? I sent feedback to YT asking to make the comments searchable and I wonder if it was ever read by a human.
On the subject of driverless cars: There is a good Kyle Hill video that talks about how an ai beat the world champion at the game Go in 2016. But in 2022, amateur go players were able to beat that ai consistently. He brought up a great point that computers are incredible at recognizing patterns, but ultimately don’t understand what a game is, what a board is, what a piece is. This is the thing that scares me the most about automation in anything involving humans: we assume that computers can make leaps in logic like humans. It comes as naturally to us as breathing. Ai looks like it does on the surface, but it is relying entirely on a predetermined information set. Taking that for granted is extremely dangerous when it comes to the unpredictability of humans.
The fact about Go is nonsense. AlphaGo is not available for ANYBODY to use! Unfortunately, Google has kept it under wraps. AlphaGo is miles above most of the existing AI programs. The same program also claimed to beat Stockfish handily. And Stockfish is already the overlord of Chess. You may not like it, but AI is superior to humans in more ways than you can ever imagine. Its still not perfect in being a general intelligence but that day is not too far off either. AlphaGo is actually meant to be a prototype GAI and its coming. Personally, I am not too happy about what AI would do to our society. But, I am not going to be ignorant about what is coming my way either.
Replacing human work with machines is great, if the work and situation fits it. Repetitive tasks, common safety check etc. all those thins are fields where machines excel because (assuming correct maintenance) they don't miss anything that is presents to them. Can driverless cars make roads safer? Yes. Under one condition: There are only self driving cars and you don't allow humans to ever steer again. Because if you do, all assumptions are out of the window.
Really, if AIs do become self-aware one day, they will be so completely alien to us we might not even recognise it. The way they work is just so insanely different from our minds.
Yeah AI might suck as a driver, but AI cars will allow us to watch titkok for the entire ride, instead of just at red lights and on straighaways and when there isn't too much traffic - ALL the time - and if anything goes wrong, the robotcar is at fault. It's such a perfect promise, that people will overlook any problems that don't interfere with their personal comfort.
I was really hoping someone would make a video like this. As a scientist with some knowledge of the state at which research stands in this field i found Veritasiums video to be nothing but corporate BS.
He's just uncritical period if you look at his defense comment in the comments section. Had an off feeling about him a while ago, glad I stopped watching.
@@LeavingGoose046 I was skeptical of him the since I saw him summon Bill Nye and Neil D.Tyson to act as overseers on an argument about mechanical engineering and fluid dynamics, none of them actually has any real experience in these fields.
This is why I have such respect for Tom Scott - who, when a VPN company tried to change the sponsored video more than he was comfortable with, dropped the sponsorship and released the video he wanted and obscured the name of the VPN company. He chose to release an educational video he believed in instead of releasing an ad that'd have made him a decent amount of money.
I was just going to comment about the irony of this video about distortion by financial interests containing a VPN sponsorship spot without any critique... Tom Scott's video really turned me off VPN sponsorships and I cringe and skip through every time I see them now xD
Fuck yes! That doesn't mean he isn't swiping cash behind the scenes though and feeding you information he gets paid for royally. Distraction by whataboutism is the perfect way to hide things in plain sight. It's used constantly by companies, politicians, etc., even with scandals they create themselves to draw away attention. Not so much by mr. Scott though. You can actually check his integrity by checking the facts. I'm just saying that, while his actions and words are evidence that support his integrity, if it was the only evidence, I would be very cautious and even more cautious after that video. The reason you notice it as evidence and it _is_ evidence is because it fits the pattern of Tom's integrity. There's a long unbroken pattern. Tom's good people.
looking forward to watching this one! i've generally loved most of veritasium's work, but the episode in question (and a few others) really rubbed me the wrong way
@@benjene according to the video description, this uses that godawful puff piece he did for waymo as a jumping off point, although i hope it gets into the bill gates interview he did which might have actually been worse
@@Kfdhjgethfdtgh774rvbjs I don't think it's a reason to lose respect - almost all VPN sponsor spots read the exact same. But I completely agree that they need to add the caveat that "Your traffic is hidden from your ISP, however, the VPN operator can see everything you are doing if they want to". Wasn't too long ago we found out a major "no logs" VPN provider was actually keeping logs and happily handed them to law enforcement. I wanna say it was Nord, but I can't remember for sure. Oh and the idea that a VPN alone makes you completely anonymous is laughable - anyone with a shred of knowledge about how web browsing works knows that's total BS.
The irony is honestly kinda sad, imagine making a video criticising another UA-camr of corporate propaganda and doing said propaganda in the same fucking video, worst part is I kinda agree with most of his points.
94% of car accidents are caused by humans is one of the most useless stats ever when you think about it. There are very few other factors that can cause accidents at the moment. It's like saying 99% of boating accidents involve water.
I know right? People say that only a few lethal accidents have been caused by Autopilot and self-driving cars and I'm like no shit sherlock, self-driving cars have been deployed in the past 5 years and are driven by a super tiny minority of the population. How on earth can you compare those two.
To be fair, the quote is "94% of car accidents are caused by human ERROR", which is to say, a person fucked up, rather than a mechanical failure, a meteor strike, an IED, etc.
This is one of the reasons why Statistics is widely looked down upon in the collective field of Mathematics, given how little grasp of it one requires in order for one to use it to misrepresent, misinterpret, or outright lie about a given data set. You wouldn't be able to pull off the same nonsense with Topology, Number Theory, or Geometry.
Love Veritasium, so this will be a great blow to my current perception of not only that channel, but many "scientifically informative YT channels" (looking at you kurzgesagt, SciShow, and Vsauce). I had forgotten to be critical of the channels I just put in the (I trust you box).
That's a wake up call alright. Often, these science education channels veer too close to a technocratic worldview. Veritassium and Kurzgesagt especially give me that vibe.
I mean, massive, free, AI driven public transport doesn't seem like a terrible idea once it's viable. The problem is the capital to build it is being funded by investors who expect to then capture all the value.
Better public transit is only a good thing, however there are a couple of issues with a public transit only system: 1. Great solution for urban areas but less so for towns and rural regions. 2. If anything causes a "hiccup" an entire city can be shut down. 3. Giving the government complete control of anything can be dangerous, obviously 4. Costs money the government probably doesn't have to build and maintain. 5. Nobody wants to walk to the bus station in cold rainy weather
And we need to price in the externalities, that way public transportation becomes feasible everywhere and not just places with high density. And even if it doesn't, it does once you factor in the greater productivity of people who can move around freely and therefore get to meet more people who will become useful contacts in the future.
I love how they think that adding more cars "that are more aware" would stop rush hour from being a thing. You know how you get less cars on the road? You build a subway system, a bus system, a train system. Any kind of collective traffic where one driver can ferry 30-700 people in one go. I take the car, the train, and the subway to get to work every day. The only reason I no longer take the bus instead of the car is because the bus times did not fit me on my way home. A good collective traffic system will make people go "oh I no longer need the car" and they will stop using it. This is the case for most of my coworkers, who live a lot closer to work.
It... Would, actually. CPG Grey did a great video on how synchronized autonomous vehicles would completely eliminate traffic jams, as the majority of slowdown comes from a compounding delay of cars moving after being stopped. This video is kinda trash for the sheer fact that it ignores legitimate arguments for autonomous cars simply because 'Grrrr.. car bad... Train good... Grrr...' Don't get me wrong, I want robust public transit systems. I would love to be able to take a high speed train from any major city to any other major city. I would love well maintained subways and bus systems. But to immediately discount the fact that, factually, truthfully, synchronized autonomous vehicles would eliminate the majority of rush hour waits is intellectually dishonest.
@@staff4226 Again, that simply does not track with reality and the physics of it. Anti-car advocates jump through hoops to try and disprove the idea of synchronized self-driving cars removing traffic jams solely because, in their heads, there can never be a world in which cars can be improved to solve any of the problems they have.
@@greed0599 I don't disagree that cars can be improved. It's just that if you tape a bunch of cars together with only one big engine pulling them, and place them at low friction rails, you suddenly get much better improvements.
@@staff4226 But that's not the argument for or against automated cars. I am, in fact, in favor of trains and public transit. I don't drive, I take the bus and tram everywhere. But you're never going to 100% eliminate automobiles, so why not improve them all and acknowledge that improvement can be made. Nothing is ever going to be 100% bad or good.
The legendary Pulitzer prize-winning journalist, Chris Hedges, recently made a video on this topic. The corporatization of Science in the US. You will find things like Nutella-funded "scientific" studies that claim that sweets greatly improve your health.🤔
Hah imagine that being a recent thing. Remember that controversial study funded by sugar related companies to make fat look worse than sugar? It’s been happening for a while. Always look at authorship for journals before you make conclusions and especially look into a journal if it’s being mentioned a lot.
Jesus, what a dreadful thought. Science classes brought to you by Nestle. Even worse is that corporations funding faulty research is something that is decades, maybe even a century old if we talk about pharma or tobacco industry.
@@XMysticHerox The problem isn't a lack of Publicly funded research, the problem is that biased corporate-funded research is used to muddy the waters and is then placed on the same pedestal as impartial publically funded research in the media and public discourse.
The irony is that taking the time to explain in detail the additional measures of 3D mapping of the road, real-time monitoring by support staff and other human involvement, Waymo actually WOULD make themselves look good; maybe slightly less cool and futuristic, but way more safe.
'Not quite as cool and futuristic, but definitely Waymo safe' I'd love that slogan, though I know it would never happen and would almost certainly be counterproductive for the company because safety, as foundational is it is in reality, doesn't appeal to human emotions like "cool" and "futuristic" do, and emotions (unfortunately) motivate people to buy things far more than careful conscious consideration does. Most people really do think that the majority of their past and present purchases are made due primarily to logical decisions, but psychological and sociological data prove that that is very, very often not actually the case as much as people want to believe.
Showing that aspect makes people start to think about safety and potentially questioning if it's actually tight and not with flaws. Like if they say "it works by X Y" the viewer might come to ask "but what if Z?"
Another great video, Tom. When watching Veritasium, Vsauce, Kurzgesagt and all the such, I usually turn off my critical eye. It's like watching a 3blue1brown or numberphile video. I just lay back and absorb whatever scientific information is given to me, trusting that it was well-researched and unbiased. This makes them the amazing vehicles for corporations to advertise. When you're sold something without knowing you're sold something. It's a little sad that I now have to be sceptical of the UA-camrs I've learned to trust for the past few years. Good work as always. Keep it up!
When you hear about the social behaviour of ant colonies you can have that sort of trust. But if you look, you will always see people calling them out for mistakes even in that sort of content. But those are harmless, to an extent. Minutiae about hard science facts do not have a solid impact on everyday life as, say, discussions about GMOs or organic foods or information security which these channels are equally likely to approach and which may provide them with even more engagement that "drier" pure science topics
Look up Folding Ideas video about VSauce. The idea that the channel about pictures that look like dicks and boobs reinvented its image enough that people took it for reputable is hilarious
even sabine hossenfelder was just recently sent onto the slippery slope of dis-infotainment. getting her video on quackery sponsored by brian keating, then promoting his channel. he is a grifter who invites far right propagandists and participates in the pseudoscientific pragerU (fake-) university scam, funded by rightwing fracking billionair cash with the goal to further the republican agenda of privatizing everything, cut social spending, increase giant corporations federal subsedies
This reminds me a lot of the early days of TV when you could see your favorite cartoon characters whip out their favorite brand of cigarette and advertise to you directly in the content. Here it is even more insidious, in my opinion, because it is content that purports to be informational/educational. There were regulations that drew bounds between ad and content then, why not now?
I want to jump on your question on why there 'aren't regulations separating advertising from other content by stating that while cable television, it's infrastructure are a lot more centralized and 'public' in nature, therefore easier to regulate, the internet is by contrast much more amorphous, and UA-cam is a private company with a scary monopoly on video sharing. Therefore, 1. UA-cam, being a monopolistic social media of sorts, actually benefits from both ubiquitous advertising and not having the same inherent accountability as governments and centralized infrastructure, and 2. Consequentially, internet activity of this nature is (in my view, for better and worse) barely regulatable, at least not without employing drastic measures that may do more harm than good.
@@willthealmightygamer6368 The government doesn't enforce existing regulations. People need to stop blindly trusting producers. We are WAY better off than we were 30 years ago. You have any idea how much work it used to take to find out the government had DEFINITELY liked to the public in 1990? It took years, it takes days, sometimes only hours now. We've always had garbage information, it's far easier to identify it now.
On this platform - "This is a sponsored video" is spoken at some point, and the words "paid promotion" briefly will appear onscreen. That's it. It's self-regulatory, but individuals can report violators. In the USA, the regulators are in a revolving door relationship with industry, so the distinction between educational material and a sales pitch just isn't in their world view.
The point about "educational" and "news" youtube channels focusing on cars as the default mode of fixing issue is something that has bothered me for some time. "Hmm, cars are producing too many emissions, making them electric will fix our issues." and "People die because humans are bad drivers, lets make computers do it." have been repeated throughout major online media channels, when busses and trains get rid of both of those issues, as well as have other benefits such as encouraging more sustainable and equitable urban design.
Because Americans have been brought up on the idea that owning a car is the embodiment of American individualism, which is a load of sh!te ofc, greater investment into public transport instead of reforming cars is the way to go, but that can't really happen when the automotive industry lobbies the government to stamp out any sort of transport reform that would actually benefit people, it's dystopian, it's gone so far that cities are designed in such a way that owning a car is a necessity, it's a vicious cycle and many people have been brainwashed to such a degree that they're even proud of the fact that they live in a country with barely functioning public transportation,
Without a core understanding of civic design, it's really easy to fall into the trap of "not thinking about it", or thinking that the american way of life is default, that there's nothing wrong with it, which just leads to car-as-default. To be frank, it's very easy to miss civic design as even being a factor here, because there's really very little published on the subject in youtube or in terms of academia relative to other shit. I've seen like 4 civic design channels above 100k, that's about all I've seen to be honest. The only one I've really seen tackle the problem of car-dependent infrastructure in a way that was coherent was "not just bikes".
@@emilycampbell6375 Absolutely. To go a step further, Americans have been taught to think about things only in terms of individualism and their inability to imagine any non-car solutions to public transportation is a symptom of that. Like, for Americans, the solution to global warming is to buy eco-friendly products. The solution to racism is to be nicer. The solution to poverty is to work harder. etc. etc.
Viewers have parasocial relationships with their favorite youtubers. Sponsors know that and want to exploit it. Sponsored content is deeply exploitive. It's greed over integrity when creators cash those cheques.
This must be an Adam Something video in disguise because the answer was "Just build a fuxking train" all along. Thank you for the well thought out, fantastically structured video essay. Very big shout out to the subtitling. It's a small thing for most people, but for me it makes or breaks a good video essay. Thank you so much for including them, and doing them well, too!
That is an answer, but it isn't a solution. I personally have never heard of trains randomly popping out of the ground, ready to be used. If it is more efficient to first have self driving cars to suit areas only covered by train in distant future, so be it. The difference between answer and solution, is that a solution is actually specific, and not just a general statement. In general Adam Something's videos remind me a lot of propaganda, just that I think that some of the content is actually true.
@@jan-lukas I mean, in London, the stations are so close together and the trains come so regularly that they might as well pop out of the ground ready for you. If you wait for more than 6 minutes for a train in London, it's either a bank holiday or something has gone horribly wrong.
@@jan-lukas there's always a smaller train. Metros, trams. And u can walk a last kilometre. I have been living without a vehical for 5-6 years now I do have 2 bikes but they are rarely needed. Bcz I live in India and there's train for everything here where I live there are metros trains suburb trains. Etc. And metro at my place also provide rent bikes. For going to college etc.
It rings a bit hollow when you push your own sponsor too What's really irking me here is the lack of data, you brought up a smattering of cherrypicked anecdotes and sensationalised news headlines, as your counter evidence against driverless cars. Sure if you look hard enough, any technology is going to have failures somewhere. But the absolutely critical thing to get across to the public is that they are still safER. Because human drivers really aren't safe. A human is a squishy, error prone piece of hardware that suffers all kind of health issues, judgement problems, opinions, tiredness and intoxication, that have turned our roads into a constant churn of death for the past 70 years. Car accidents are by far the most common non-disease cause of death, they ydeserve far more intense calls for actions than they currently get, and certainly more than gun crime gets. The fact that the tech isn't perfect, so we should wait until it is, is an endlessly moving goalpost that ignores the realities of thousands of people dying worldwide because of cars every day And your claim that since derek's video is propaganda, that you don't have to "engage with the minutiae" of his arguments, is an intellectually dishonest handwave. So far i'm 8 minutes in and you have repeatedly used opinions and anecdotes in an attempt to fight data. You've not demonstrated that you have any more integrity than those you disparage, and i'm struggling to see a reason to continue
To be 100% fair, VPN sponsors oftentimes also make you say talking points that are either just barely true or just straight up a lie in those sponsored segments. But those are never a part of fully integrated videos so it's different.
It is different, but still not right. Because the main content of the video was not influenced, it is different. But that does not make the misinformation justified. (This should be applied to all UA-cam sponsorships. I don’t know if Tom’s sponsored segment had misinformation.)
The VPN ad just hammers home the point that if the information is sponsored in any capacity, even if it's in the same video that's debunking another video for completely misrepresenting reality, it will be complete and utter bullshit. Be it Gandhi, god himself, an actual saint, they will lie and/or knowingly misrepresent facts through their teeth.
@@TheLolilol321 I sincerely doubt that all sponsorships corrupt content completely. What if the sponsorship was just about getting the recipient/UA-cam to engage with a certain topic on their own terms (and such sponsorship is disclosed.) The main issue stems from the direct control that a UA-camr can give to the sponsor, best exemplified by SnazzyLabs’ comment on this video. That being said, a bias can still form when a UA-camr is given money to simply research a certain product, but if the the tuber provided all the sources to the audience and encourage them to engage in due diligence themselves before making a final decision on the product, then most bias should be eradicated. Honestly, things would be different if Veritassium provided the sources of the information directly in the video/description. And although it is a different scenario, things would also be different if Tom said directly in the ad that he got his information about the vpn , but and vpns in general, directly from the vpn company.
@@TheLolilol321 And what if the sponsorship is just “Hey, this thing exists. Look into it, maybe.” Like the many icons on nascar cars and such. At the very least, there’s no misinformation in those instances. Other issues, certainly, but usually far less impact on the audience’s decision.
@@xzylo451 An exaggeration to be sure, yeah, citing sources in the video right before the claims would indeed be a nice solid step to take. Corporations get their mouthpiece and the UA-camr can indicate that they do not necessarily personally vouch for the information. I did say sponsored "information" though, which a logo with text is not.
"You convinced me, drivers are unreliable and traffic is a dangerous problem beacause of the chaos it produces" "So you gonna support our self driving car start up?" "I'm gonna fund public railways" "wait no"
@@Cybermaul Also superimportant: saving a LOOOOT of energy. People thinks of Hydrogen/Electric cars as "green" but even if we can solve all the problems related to the production/efficiency, they still consume lot of energy (generating and encasing hydrogen is a complete mess and it will never EVER be for free). Cars are just that inefficient to transport people compared to train, buses, bycicles and..feet, and there is nothing we can do about it unless we move to a universe where thermodynamics do not apply, or discover magic.
No thanks I don't like walking miles to wait for a half hour for a dirty bus I have to wear a mask on, which then moves slow as fuck and drops me off a 30 minute walk away from the grocery store, then have to repeat this process to go home when I could have driven there and back by the time i even got there with public transit
I completely forgot how angry Bill Gate's book makes me. It's disgusting that he tried to sell a book of extremely basic climate change information, preaching "personal responsibility" while he has the money, power and influence to directly make the massive industry changes necessary to actually solve the climate crisis but never will.
In his book he does actually say that we do need governmental solutions to the problems, but doesn't really talk about all the subsidies fossil fuels already get, He just talks about subsidies for consumers to close the gap between green energy and fossil fuels and carbon taxes and stuff like that. He's a billionaire so of course he's not really advocating for massive change to a system that worked out for him.
@@arsenalfanatic09 and sidesteps the latter part of OP’s comment by saying “well, hey, capitalism makes self sustaining enterprises, so it’s way better for the environment if I invest in companies to do the changes for me (while I make dividends from them all, whether they actually help the climate or not)”.
Veritasium's content is great in general. but this is a friendly reminder that we should be always thinking for ourselves and be critical of the things we hear from anyone, no matter their qualifications.
I was shocked recently by one of their oh-so-sensational physics videos about speed of light. It cherry-picked outdated methods dating back to Galilei whilst it argued with 1900s papers by Einstein - at the same time, left out methods, empirical evidence, and entire areas of physics that not only came decades after, but also do provide evidence on numerous "ooooooh soooo mysterious" claims in the video. Just one example: speed of light and its (potential) directional dependency... it refers to antiquated methodology, a nowadays unnecessary and frankly ludicrously misleading thought experiment flawed on numerous levels, AND leaves out the Michelson-Morley thought experiment. Not to mention how it leaves out the arrival of Maxwell, atomic clocks, and shockingly precise interferometry. And all the evidence that demolishes the numerous "ooooohhhhh mystery...look at that" nonsense. Seeing something like this I admit: I at least have to treat the entire channel as highly suspicious in its purpose & agenda, because 1. if presenter has erudite background, then cannot make such videos, it would be impossible without a strange agenda, 2. if presenter really understands only what is in the video then must never have made a video about this topic. Not sure which is worse, but to me, Veritasium seems to be an at-best sensationalist "education" channel, at worst tries to raise doubts about known areas of science with some non-scientific (take your pick) agenda.
his content is great? it´s largely a clickbait channel full of half truths and inaccuracies for scientifically illitarate people that want to feel like they learned something while beeing misinformed. it´s just sensationalism, "education" is just its marketing theme while at its core it´s mostly just an advertising vehicle for his many, many, many sponsors. i mean the guy is a multimillionaire a hundred times over, you don´t get there by plaing fair. and it´s not like he´s using his ill gotten gains to give out grants, like real educators like sean carol for example with less than 1% of what he has do. in his pinned reply to tom, he´s seriously citing an "academic study" that´s been bought and paid for by waymo. he knows damn well, that this piece of blatant propaganda isn´t worth the paper it´s printed on. yet he still has the gall to present that as an argument. i´m looking forward to his video advertising the health benefits of smoking, since there were plenty of "acadamic studies" bought and paid for by phillip morris and co. as well that showed just that. i´m stressing this because this isn´t an oversight, that´s not a tiny oversimplification or a little clickbait sensationalism, this is straight up lying to your face. and not only that he´s misusing "science" as a tool to fluff up his lies when calling it an academic study. for me, someone capable of something like that has lost every shred of credibility. i mean i only personally object to your first sentence. i hardly couldn´t agree more with the second. but if you apply that critical eye to veratsium, theres hardly anything left that´s of any substance at all. to the extent that it's true, it´s trivial and to the extent that it´s not trivial, it isn´t true. so in short, you´re wasting your time watching veratasium, when there are so much better creators out there
@@symmetrie_bruch Exactly. I can only use "great" as sarcasm. I was frankly astonished of the content on this channel. Either someone deliberately ignores decades of physics (and/or cherry picks things), or truly doesn't know. In both cases, should not establish a channel with "veritas" in its name. Rule of thumb, if such amazing names are given to something, usually it is overcompensating for other things... and best to avoid.
Hearing that clip at 44:45 where Muller talks about how people can be happier because commuting and being stuck in traffic sucks had me nearly shouting "TRAINS!" at my screen.
@@electrochipvoidsoul1219 Lol, have you seen american roads? If you have more than 3 lanes in a street, you can just replace the two central ones with trains. Well, they would be trams, but the point still stands. You could put a train on every single street that has more than 2 lanes. And some cities are just doing exactly that. And let's say some streets can't do that for whatever reason, putting a bus on that road is just as easy.
@@electrochipvoidsoul1219 You can. You can also put a bus, metro and tram in the middle of a city. In fact, the only places you can't put them is outside of cities in small villages in the middle of nowhere. And that is mostly because in the middle of nowhere in the US often means in the middle of nowhere instead of a 10 minute drive from the next village over.
I love both channels but I love this youtube "peer review" even more. Initially it was a hard pill to swallow - especially the Johnny Harris video, but I'm glad to have come across this because it makes me a more responsible and informed consumer of youtube content.
More ‘education’ channels should do stuff like this. Every other form of education, science and engineering for example face scrutiny and review so why not UA-cam content as well.
If you are more responsible please don't listen to the video on this channel. The bold claim that self driving tecnhology is just "a pipe dream" shows Tom nows nothing about the topic and his just trying really hard to find the clickbait views. Which is ironically the exact same thing he is criticising veratassium of doing. He is spreading missinformation for the money. In this case the youtube views of a clickbait title.
@@paulogaspar8295 Nobody is arguing that self driving tech isn't something we should be pursuing. The main issue is that self driving tech doesn't work well with human drivers involved, or at least not well enough to squash the perception it's unsafe. We need to massively redesign transport for self driving cars to work. Unfortunately that won't be happening anytime soon so self driving tech will most likely be limited to interstate transport where conditions are much more predictable. And lets be honest here, the main issue with the US transport system is that it's designed with one goal in mind: cars. That needs to change before all these technologies can be fully utilized. Until the US begins to change how cities are designed, zoned, etc, none of these problems are going away. Everything is patchwork until the US drops the mindset of cars first.
Clearly, the "94% human error" doesn't say anything about the safety of AI drivers since they haven't even been pooled into the sample. To prove anything about safety either way, there needs to be a direct comparison between humans and AI, and it needs to be expressed in some sort of a ratioed measurement, such as "number of accidents per 100000 miles" or the like.
I agree that a proper comparison needs to be made. AI shouldn't be compared to purely hypothetical 'perfect' driving. Also, when I look at some car crash compilations here on yt, it seems like 99.9999% of accidents are due to human error. Even for example in heavy fog or on icy roads, it is still human error to drive too fast in such conditions. (imo)
Is this sarcasm? NHTSA has been putting out these statistic for well over 25 years, and AI driver reports from various companies (and the government agencies keeping tabs on them) have also been publicly reported with "number of accidents per x miles" since before the pandemic. The standard number of miles is 100 million vehicle miles traveled, so 100 thousand is almost too on the nose to be a random guess on your part. They even have break downs to causes of accidents as reported/discovered by the police reports to help narrow down what or who is at fault. These are USA specific, unfortunately, but that was the target of the comparison to begin with, and will hopefully be updated with other countries as autonomy is rolled out, if possible. That being said, not all governments track these sorts of things.
@@donaldhobson8873 Yep, fair point. Granted, that's just for now. As the technology matures, they will have their use cases expanded too. I believe those details are also in crash logs (reported in public records). I'll have to double check. Still most human accidents happen in easily avoided, but overwhelming situations. Like crowded parking lots. Here at least, the usual bot shines (at avoiding crashes). Keep in mind, I am not claiming they are ready for the prime time. Anyone who says that is trying to sell you something (and in the case of Tesla, they may be lying by omission quite heavily).
What a masterpiece. Honestly in my opinion this is one of the most valuable videos you've made. The criticism of the specific Veritasium video is great but especially pointing out the implications of sponsored educational content in general is something that deserved way more attention than it got so far, and you've made an incredible video to give it the attention it needs.
That's very kind. It's always difficult to avoid these feeling a bit like "hit pieces" in some regards (and, yes, I add a bit of theatrics to make it enjoyable), however my focus is always on the broader trends that these examples throw up and I hope that comes across in the video.
@@Tom_Nicholas To be quite honest, though I haven't finished watching your video, I always end up thinking "if there's one type of content creator people should be especially demanding and harsh with, it's educators". How could someone have the pretense of educating people while also doing sponsored videos and generally responding poorly to good faith criticism? Like, if nazis like PewDiePie did that (and he might have done that, idk) I wouldn't bat an eyelid because I know better than to ascribe educational value (or indeed, any value) to his videos, but the Veritasiums and Kurzgesagts of the world should know better and be more humble when caught with their pants down, especially if they're already doing very well for themselves financially. This is a long-winded way of saying that I support what you do. Keep up the great work!
@The Program if he'd made those two separate videos, both videos would be weaker because the two points reinforce each other. Veritasium is wrong about self-driving cars in a way that specifically benefits Waygo, and it's disingenuous to try and divorce the nature of that wrongness from the corporate relationship between his channel and Waygo
I've seen a fuller version of the "space-saving" argument, it isn't a complete non-starter, it just assumes that with the added safety of driverless cars, we would be able to have cars travelling much closer together, at much higher speeds, thereby having less road space taken up by the same "human transport bandwidth" of cars. This is completely correct, and the configuration of many cars tailgating close together at high speeds has been tested, and named: it's called a train.
I was going to add this as well. Congestion is typically not caused by car volume but by car spacing adjustments made (relatively) slowly by human drivers, as well as lane changes propagating slow downs. I think the best counterargument is the one you and Tom made though: public transport solves this and many other problems, driverless or no.
@@RaunienTheFirst oh god I remember when I was an automated car Stan and my great idea was to have the cars link up together as they go down the road to take advantage of reduced drag and increased travel density and now I realize that that is called a train
@@land_and_air1250 Yeah, it's a funny thing. I used to be a stan for it, too, and I even went into computer engineering and AI to study things like this, but imo the tech developed for self-driving cars is best used for anything but cars; just make trains and bikelanes, for goodness sake, and put the AI to good use on autonomous Mars rovers and stuff.
Gonna say this video has aged like fine wine. The @veritasium response to it has aged like milk. Especially in light of the massive scaling back of autonomous driving programs across the industry, the revelation of cooked numbers by Tesla and Waymo, and the drastic reduction of investment in these projects. Thanks to Tom for the video, and I await Veritasium's meo culpa.
Funnily enough, Veritasium's driverless cars video is what lead me to the Not Just Bikes and Adam Something videos about urbanism and car-centric city planning, which in the end made me realise how bullshit that Veritasium video was
As nice as it would be, you can dream and plan for your future, while still living and acknowledging where you're at now. AI cars would solve a vast majority of our societies issues in the transportation sector in an affordable manner.
@@RebelTvShka More affordable than a literal tram? Edit: I mean I suppose if you're going by the least additional infrastructure required, but like. Mass rollout of autonomous cars is certainly not more affordable than basic public transit.
There's a really nice quote from nuclear historian Joseph Morone which I feel is applicable here: _This is not a story of technology run amok - although that is how many people understand it to be. The history of nuclear power is a history of political and economic and social decisions being made about a technology, and the key decisions weren’t being made by the technologists. They were done in the business room._ _What science and technology gives you is a range of possibilities. And those possibilities can take you in any number of directions. It’s potentially a liberating force. But to get there, society has to stop sleepwalking, and realise it’s not a scientific choice. It’s not an engineering choice. It’s a moral choice._ I think my frustration with the Veritasium video comes from a wider frustration in the belief that *this* form of the technology is the inevitable form it must take. And if that was the form it must be the right form. And to me the blind faith in science is a very wrong headed view of science. The existence of a technology does not automatically guarantee utopia. Without holistic planning and policy, negative unintended consequences are likely, and Science Tube routinely forgetting that will always be a source of frustration to me. Autonomous vehicles will not automatically drive us to a better transportation future. We have to take the wheel ourselves.
You're right. We can dream of a future with all-around self-driving vehicles, but the truth is: We don't know if that's the future for us. Just stating, that it _will_ be the future and now we need to get there asap may leave other options and ideas underdeveloped. PS: I love your videos, and I've watched all 4 of them. :D
I think what often happens is that technology becomes a stand in for a company or institution that is using said technology and that is intentional. People basically always think of technology as an automatically good thing and with good reason seeing how much our lives have improved due to it and that is why companies and the army love to associate themselves with technology and having propaganda to make it seem like they are that technology. It's a great way to pre-emptively derail the conversation, if I criticize Tesla I'll inevitably be hit with the accusation that I hate electric vehicles as if Tesla somehow is the concept of electric vehicles and we never get to discussing the problems with the company Tesla.
yeah mate, absolutely. i think we're coming into an era now where its too easy to see a technology as the future & talk down to people who truely think through if something is right or wrong in nuanced ways. like blockchain.
Having a degree in Artificial Intelligence and working in the AI field for almost a decade now, the Veritasium video really irked me when I watched it for its completely uncritical portrayal of Waymo and driverless cars. Him doubling down makes it even worse. Content creators should be extremely prudent when doing a sponsor integration like this, because as we say in Dutch: "trust comes on foot, and goes by horse."
Tom Nicholas is one of the best youtubers when it comes down to tackling big issues like these. He has criticized a few massive ecational youtubers in the past.
@@victoriap1561 It can definitely be flexible. Train isn’t just limited to heavy duty subway or passenger train it could be a light rail or even a tram. If there are encompassing system in place, utilizing a network of various public transport can be just as flexible as private vehicle. You just never get a chance to see a world outside of this dystopian reality where automotive industry has total control in shaping our infrastructure
@@victoriap1561 If we match the amount of cars, and length of rail to the demand of the city it's being built in, then there wont be an issue of having "enough people". Trains arent just built 1 size. They're flexible because they can be built to suit the city's needs and population density. Cars can be added or taken off. Multiple engine cars can run at once during peak hours. It employs locals, allows cities to become more walkable, reduces emissions/improves air quality, and... so much more. You say trains need a lot of people to run efficiently, ok... There are a lot of people in cities. lmao. That's why they work better than sitting in a traffic jam. I know new futures are hard to imagine, so maybe check out "Not Just Bikes" youtube channel if you're interested in learning more.
I've been feeling iffy about the channel ever since he did an episode on DNA and how it caught the Golden State Killer. The video seemed to gloss over the negatives of mass DNA surveillance and not explore them very fully at all, focusing 95% of the video on a company whose business is the use of said DNA. Not exactly an unbiased or balanced look. It just really rubbed me the wrong way.
worse yet, in my opinion, he alluded to it many times and liked comments that mentioned it. However he never actually addresses that huge issue and really get into the surveillance problem
While I think doing a whole sponsored video is off-brand for someone claiming to be a source of true information, this expose is actual garbage. Just a few points: As for the flight, I can't speak as to what Veritasium's point was about what it showed, it definitely did not show the danger of automation, because the landing was not fully automated, and only automation-aided, and the crash happened because of the human's wrong choice. As for the 94% study, you claim their methodology includes any cases where any blame whatsoever can be laid on the driver, even in an unreasonable way as it would be in your example. Yet you show zero substatiation for that claim about the methodology of that study. Then, of course, you claim that Veritasium tells us knowing about the lidar and cameras is enough to have a complete understanding of how the system works, and, again, show absolutely no evidence of that. Seems to me like you are trying to put ridiculous words in his mouth.
Veritasium/Waymo deliberately leaving out human created maps and other human labor is a great example of Potemkin AI-where a system is presented as "real" AI when in reality regular human workers operate critical parts of the system. When a company claims to have an AI your first question should be "who's behind the curtain pulling the levers?"
Like the "self-sustaining" train in one of Chris Evans movies. Train creator/designer/owner claims the train is self-sustaining, but later, it turns out otherwise. Not gonna spoil it for other ppl.
@@heartache5742 Any term is a real term as soon as people start using it. And as described, it's a fine reapplication of the longstanding term Potemkin.
I really like that this video seems to be careful to not just be "automation bad" or something like that, especially given that automated systems *can* improve safety a lot in a lot of contexts (just not every possible one). Also trains are good, actually. And more easily automated, too, I guess ^^
@@SECONDQUEST You enjoy your Anti-American trains, I'll be watching you from the highway while enjoying my freedom to be stuck in traffic for 6 hours. 😎
Having read Veritasium's response, I am beyond disappointed in them. Its pretty obvious to me, that no matter how good or honest a video is, if the subject of the video is about a sponsor it is unethical to try and pass it off as anything other than an advertisement. Any attempt to say "well the video was actually unbiased" misses the point entirely. If they truly believe that driverless cars are the future and that we need to support the technology and their video needed to made exactly the way it was, they then should not have undercut their argument by taking the money.
Tom, this is a great video. What irks me though, as a programming student, is the fact that you critisize veritasium (rightfully so), but fail to see that VPN ads are very misleading too. I am not saying that the two are equally bad, I agree that the Wayme placement was worse, but VPN ads are also very disingeniuous. Tom Scott has made a video about it (ua-cam.com/video/WVDQEoe6ZWY/v-deo.html) which explains most problems. Besides those, claiming that a VPN protects agains big companies snooping on you is completely false. Big companies nowadays gather data through fingerprinting, account logins and cookies, all of which work regardless of whether you are using a VPN or not. There are ways to stop these things from happening and they are free (like using privacy plugins and using less of the big companies services). This makes me even more against these ads, since they infir that you need to pay for a VPN to be safe, while this is not true at all (it doesn't make you any safer in most circumstances).
I commented this elsewhere, but I think it's 2 different beasts. Is there some level of hypocrisy there? maybe, but it's not like tom nicholas did an entire 10 minute video praising how much safer VPNs are while saying "btw surfshark good, here's an interview with surfshark head of operations, btw im using surfshark now.". It's a very delimited section of the video that is not integrated with the content in any way, and it is the integrated ad that is truly dangerous. VPN ads irks me too though, and they never focus on the only real reason people would actually use them (region-locked content).
Veritasium is in the science and tech sector of youtube, he undoubtedly *knows* he is misleading people willingly. He is trusted to speak about science and tech specifically, so making a full length advert on science and tech that subtly pretends to be an unbiased video is misleading on another level from Tom Nicholas's sponsorship. Tom Nicholas's sponsorship is on something he isn't expected to know about, and although there can be a level of exploitation of this fact, if this fact wasn't respected at all, most sponsorships just wouldn't happen on youtube. Most VPN companies lie about the utility of their services and you can recognize this because you are a programming student, but you might not recognize if another commonly sponsored product such as raycons were bad, so if you were a creator, you would need to do fairly detailed research on the sponsor, which isn't something you can expect all creators to be capable of doing correctly. Just now after writing this have I decided to watch the video you linked, at Tom Scott actually respects the point made in this paragraph. Another point is that when the sponsor and sponsoree are closely linked together, the marketing is much more effective, thus much more valuable. Not only is Nicholas's sponsorship less valuable because it is temporally shorter than veritasiums, but it is also less valuable because it is less effective because nicholas's sponsors aren't chosen to be in line with his content. This raises a further point about future content. If veritasium gets a deal with a company who has the same products/services/interests and vertasiums content, veritasium will have to avoid conflict with the sponsor not only in the video in which the sponsorship is in, but any future videos if he is wishing to get similar sponsorships in the future. He is likely to want to get similar sponsorships in the future for the same reason he got them in the past. On the other hand, Nicholas gets his sponsors from various backgrounds, meaning they are replaceable and that he doesn't need to worry about a conflict of interest in a future video. This means that the non-sponsored parts of Tom Nicholas's videos can always be trusted and aren't at risk of bias because of sponsorships, but veritasium's UNSPONSORED videos can no longer be fully trusted, thus every second of what vertasium is currently outputting in terms of content is somewhat tainted.
@@StupudPerson He isn't 'expected to know' about VPN, the thing he is taking money to push? I mean, if you think that is 'honest' actor who 'couldn't know' he was misleading people or 'VPN is so advanced you can't expect' (any) research to be done. And the point about how VPN is less effective ad, yeah, because they don't care who they pay to put their ad WITHIN a video.. you telling me this guy so dumb he doesn't know how VPN ads work? And can't do research but can make 51 min that I should then believe anything within? ALRIGHT? No.
I really like how Linus Tech Tips does VPN ads on the WAN show. "Want to pirate terabytes of content without getting those annoying cease and desist letters? Try Private Internet Access!" Most of the good uses for a VPN are of somewhat questionable legality and ethicality, so they're kinda hard to advertise honestly, but they are great for gay pirate assassins.
Look, I work in tech and hate VPN ads but these aren't really the same thing. One is a sponsored ad break -- the other is literal editorial control over content being produced. Honestly tho, I am 99.9% sure you and I both know why people are actually getting VPNs. Very few are getting them for privacy. Most are getting them to get around content restrictions and pirate shit because they don't want a seedbox. But you can't explicitly say "this shit is great for piracy"
I spent about half of this video yelling "public transport" at the Veritasium man, so of course I appreciated the conclusion. Wonderful work as always, Tom.
This is like when Elon Musk proposed this new form of transport that is revolutionary, it will fit in multiple people, travel underground at high speeds.. and everyone was like wow what a visionary.. more enlightened people were just “well that is just a shitty version of the subway”.. people need to get a grip of reality
TRAINS ARE THE BEST Even long busses are just trains on wheels. Even better, trolleybuses are busses that are entirely electrical, don’t carry their own fuel and are crazy efficient... hmmm, those soviets were onto something.
@@cezarcatalin1406 If drivers now stopped staring at their phones while driving, everything would be perfect. (Sorry, just saw a tram driver do that a couple of days ago.)
Public transport works well in a metropolis. It does not work in most other areas in the same way. Things are too spread out. And in environments where the climate can make walking to distant bus or train stops dangerous, it is even worse. So a combination of public transport and self driving cars works best. For example, if I live in small city and work a late shift, it would be hard to use public transportation to go to the usual places people go in the daytime since public transport shuts down after dark in small cities due to not enough people to justify running it.
@@NandR Public transport can work wonderfully in for example Swiss versions of rural areas that are not as extremely rural as some rural areas in Australia or the US for example.
it is not though? it's terrible in sparsely populated areas, which is like most of the world when you exclude cities which are the only place where public transport is actually good and *should* be the preferred transport method
@@Capybarrrraaaa And exactly that makes it terrible? Imagine having a bus that arrives once per hour and needs double the time to arrive in the city, further away from your workplace than the parking lot. Thats public transport in rural areas for you. Unless people are EXTREMELY poor (and not even that really counts cause public transport is expensive as hell), you will never choose to waste up to 2h to/from work. Whereas self driving cars are much more flexible and can accomodate those situations much better Hell, i live in the city and need about 50m with public transport (with a tram+bus that both arrive about every 10m), with a bike i need 35m, with a car id need 20m. Public transport is an intermediate solution, but ultimately f sucks for nearly everyone
On the topic of driverless cars - it is the last thing we need. What we need is everything else: Trains, Trams, Busses and Bikes. I was in Freiburg (Breisgau) a short while ago. It is a city with more bikes and public transport than cars in the inner city - it was simply the most beautiful, relaxed city I have ever seen. 🙂
Its already horribly hot in Freiburg without cars, so its good that they only run trams in the most spaces. They need more greenery in the city center for sure
I was so captured by Mullers statements!! I want to be able to quickly step in a cheap, selfdriving vehicle everywhere I go. And I want everyone to be able to do that. To enable this, I propose to put them on rails, to make it safer and easier for the algorithm in all weather, and make them bigger and longer, so they can take many people to their destination at once. Selfdriving vehicles for everyone! That would be an amazing future! Oh yeah. As the name "self driving vehicle" is quite long, I would propose a new name. Maybe "train" or "tram" or something like that. What do you think?
@@madthough2298 no, they don’t but there are many other companies who make high quality public transport, and I as someone from Sweden living in the suburbs can vouch and say that public transport is objectively better than cars, you can watch videos from either “not just bikes” or other educational UA-camrs making content about urban and suburban transport and design
You don’t have to walk longer than 1km to the nearest buss stop and almost always less than 500 meters, these busses go to very 5 minutes to every 15 minutes depending on the popularity of the route, and not to mention but this is in the 11th biggest city in Norway with only 61 k people (or less than a tenth the population of San Francisco) the city is named asker
@@camelopardalis84 damn, guess I better disregard all of his points because he uses harsh language (btw neither Vaush or Adam use "retarded" to describe the things they don't like/agree with).
@@lolusuck386 Of course they do. That's how I found out they use such language. And it's not "harsh language" - it's a slur based on - at best - disrespect for a marginalised group.
On the airplanes, I’m a flight instructor, what you said was pretty correct(aside from the localizer just being one part of the ILS system, in the picture there you can see the second part which is the glide slope) and what veratasium said was technically correct, but both you left out some important details to frame this into a black and white automation good automation bad situation, which it just isn’t. Automation with human watching over the shoulder of the automated system is really the safest way currently. No way would I hop into the back of a driverless vehicle, but that doesn’t mean I wouldn’t want some driverless capability. For example airplanes nowadays have RNAV and in particular LPV approaches, which basically do the same thing as ILS(though they tend to have higher minimum visibility and ceiling requirements), but are done via GPS and satellites, with no ground based navigation aside from ATC direction(though it’s sometimes not used if you’re doing a practice approach or are in VFR conditions, and while most don’t land the plane for you, it sure does get your close to where the only thing left to do is the flare. Though Garmin recently came out with an emergency autoland system, where the plane will talk with ATC, pick an airport, go to the airport, follow an LPV or ILS approach in, and land it at the push of a button, all be it for smaller airplanes. Veratasium missed a bit too, for the same reasons you stated about it being a more involved process, but while it may seem impressive that a plane can land in little visibility, that’s basically what the entire instrument rating is for, so you can fly an airplane when you can’t see out a window just by reference to your instruments. Pilots can easily shut off the CAT III autoland and land it themselves no problem, and most do unless it is legally necessary in like zero visibility conditions(usually because they give notoriously hard landings and what’s the point in being a pilot if you can’t fly the thing every once and a while). Airplanes too are surprisingly much easier to automate than cars. For one, you aren’t dealing with lots of traffic comparatively and where autopilots are mostly going to be used is under ATC guidance(unless you’re landing at an uncontrolled airport). You also have instruments which will pretty much tell you everything you need to know about where the airplane is and what it’s doing, and everything happens in an much more predictable manner. Many airline pilots will tell you that they’re almost like actors following the same script every time when they fly, and while yes you can get things like crazy wind every now and then, you can totally automate corrections in. Even with all this automation, there’s a reason why there are two pilots in the cockpit, and why autonomous planes aren’t a mainstay in aviation. Every pilot whose flown with an autopilot has dealt with an autopilot being stupid, and I’ve had to flip off(both the button and my fingers), the autopilot to actually get the airplane to do what I wanted it to. You need someone to step in because computers get stuff wrong, more often than not. The safety benefit of computers isn’t that they’re smarter than the pilot it’s that they’re consistent. Computers are human inventions after all and if they get bad information, garbage in, garbage out. Even as someone that has used automation to fly, I would never move to a passenger seat if the plane was doing an autoland. Also your Asiana 214 statement is correct. A lot of the time pilot error related accidents happen these days because of what is called mode confusion, or as the ever present line goes in aviation “what’s it doing now?” Pilots can get confused by automated system just like a driver would get confused by someone else doing something stupid. However that doesn’t mean that “automation caused the accident”. It was human error, and the crew made some really poor decisions, though unlike veratasium is implying here going for a non-autoland was not one of them, like I said before it’s quite common because computers don’t make the smoothest landings and again computers can be stupid. The pilot is at the end of the day responsible for the flight, and just because the computer is acting up doesn’t rid them of that responsibility(either that or the pilots are easy scapegoats for glaring automation issues depending on where you stand on the political spectrum). The 737 Max case being a very rare exception where it was the case because Boeing made the decision to not tell their pilots about the MCAS system onboard their airplane. If they had so much as put in a PowerPoint slide about the system, it would have been called pilot error. It was a pretty bad pick of an accident by Veratasium, the better pick for an accident could have been Air France flight 447 where the autopilot came off because of an instrument error, and the relief pilot incompetently put the airplane into a stall for almost 20,000 feet until the Captain came in and saw he was acting stupid. But by then it was too late. That all being said I really enjoyed this video. Although I say all these things, really this is like a lawyer hearing two non-lawyers debate law. Like you know they are both wrong on some things, but you weren’t really expecting perfect information. But imperfect information is different from misinformation, and this is definitely growing problem on UA-cam and you did a good job calling it out.
you know? this is the real problem i see with youtube educators: they try to be jacks of all trades when it comes to their content, but they arent. I wish they would, at least, consult neutral experts -if possible- and have a good peer review when writing their scripts and choosing their evidence. It would be a lot better to have channels dedicated to, and only to the discipline of expertise of the one/s who research and write the content. For this reason I never truly trust youtube content, and less when theres sponsors involved. I would think its quite obvious they wouldnt speak neutrally of the products sponsored, and its just a proposal, but i guess some people take this too seriously.
I'd like to add that they are aircraft out there that can take off and land themselves including managing the guidance, flaps, throttle and landing gears, most notably military unmanned drones like the global hawk and reaper drones. But those also have a human operator present in a control room to take over at anytime.
Thank you for sharing what you know! This was valuable context. I knew automation + human teamwork was best but you provided a lot of learning material that helped me understand why.
I've done enough flight school for a PPL (no IFR yet) and read a ton of air accident reports. As a person who's actually a software engineer this issue of aircraft modes horrifies me. Modal systems in software applications are avoided as much as possible in all modern systems because they're simply too hard to use. Users get confused about which mode they're in all the time, and this is even though there are often very clear mode indicators near the user's centre of attention. From what I've seen of the modal systems in aircraft the output of which mode the aircraft is in is not so easy to determine and takes mental effort to happen -- no wonder pilots get confused about which mode they're in (and the fact that the aircraft is free to switch modes without pilot interaction makes this even worse). It would be far more surprising if mode confusion didn't happen, and I can only assume that the consequences for mode confusion are generally benign or air safety would be much worse.
I get your point. I dislike such obviously paid content too. I think, however, you went overboard and now you produced a very one-sided view on the topic as well. Just for the clicks. And therefore ultimately for money too…
@@christopherreichle6670 That's true without a doubt for programming in the usual sense, but not so much for the machine learning pipelines that are the core of self-driving systems. You don't really tell the machine to look for a stop sign, for instance, you just show a bunch of pictures and tell it where the stop sign is (along with other signs, traffic lights, cones, pedestrians, other vehicles, and so on). You then set up a machine learning model, typical a neural network, that analyses new images and tries to understand what it's seeing. You, as the designer, can improve its performance by messing with the neural net architecture, or the objective function for the training, or by giving it more training data, or by training it longer, etc. Doing so effectively will result in fewer errors. But errors cannot really be eliminated in this approach, so it is fair to say it's the network that's making the errors (though I suppose morally speaking the responsibility still lies with the designers because it can't lie with anybody else). I mentioned the image analysis part of it specifically but really there's likely a whole lot of machine learning bits and pieces for which version of the the above apply to a larger or lesser extent.
@@2adamast machines don't make mistakes. They can't think for themselves. Mistakes are accidents made due to an action made under false pretenses. Machines have no false pretenses they just have instruction lists that are chosen from the logic structure that we gave them.
The content of the video is well made and interesting, but funny how he talks about misinformation promoted by sponsorships and then proceeds to tell us how a VPN will make you anonymous.
sure. But I think even that shows the point. I believe everyone expects ads to be somewhat manipulative. Tom's talking about a situation where the ad is integrated into the entire video.
Yes, a quick ad break is exactly the same as Veritasium making an entire video for his sponsor wherein he seriously downplays the downsides of the sponsored product.
Thank you. Sorry for the long post but I develop machine learning systems for medical devices. When the the Veritasium video came out I was really irked by it. It was such blatant corporate propaganda and the critical thoughts (if you can really call them that) that were expressed in the video were laughable and clearly designed to be straw mans. I am always annoyed when people overstate the capabilities of technology and even more when they lie with statistics. Machine learning is just a fancy term for function fitting. Like you did in high-school where you fit a polynom to some data points. That is basically it. There is nothing more to it. Our models may be a lot more complicated and the data sets a lot larger, but in the end it is function fitting. And the fundamental problem is: You cannot make sensible predictions in regions of the input space where there is no data. And therein lies the problem: It's easy to get data from "normal" situations. Like driving along a highway. And since it is "normal", there is little information in this data according to information theory. But the interesting case is: What happens in difficult situations? You won't get much data there because by their nature they are rare. You can try to simulate it. But the problem is: How do you know whether your simulation is reasonable complete? You don't know what you don't know. And how do you proportion "normal" situations with critical situations in the training data? Too little and they will get overwhelmed by the "normal" data and too much and your algorithm gets a distorted sample of reality. And regarding the evaluation: A car driving thousands of miles means nothing on its own. It is the composition of the test dataset that is important. If most of the driving happens on a highway that is not really impressive. We already had a working system in the early 90s (ALVINN) that could do that fairly reliably. Of course we have made some progress since then. But the idea that you could put one of these cars in a highly dynamic environment like New York City or any European city and expect it to work reliably is ridiculous. There is a reason Waymo is located in a car friendly city. Very limited bike traffic and very limited pedestrian traffic makes for a less dynamic environment.
I mostly agree with your points. But in the "not normal" situation, both human and machine will perform poorly. Human just has the advantage of "may be" that machines don't. i.e. human will be like "I am fucked. But if I do this, then may be it will work or make it a little better". Machines don't have that. So they are like "welp. Gotta sit tight"
@@youtubeusername1489 The problem is significantly more complex than that. humans seem to be better at knowing what they don't know, artificial neural networks clearly do have some similarity to the brain and there have been papers about that, but the amount of similarity is not enough to have all of the benefits yet. humans seem to be much better at understanding a new experience correctly the first time still. artificial neural networks can generalize but they're fragile in ways we don't totally understand yet. ultimately what that means is that small changes that a human wouldn't even pick up on can result in the model not having any idea what's going on. meanwhile it performs at superhuman levels for situations that are imperceptibly different from failure cases. unbeknownst to you, perhaps it's because somewhere in the neural network pink and yellow in particular positions on the image happen to confuse it and you need to slightly move the pink down because no car in the original training data set had pink and yellow in those particular pixels. or something. we can even construct examples that cause a neural network to fail and they are mysteriously too easy to create and don't even seem to change the image. neural networks are amazing, but you're still smarter than them, and artificial ones are not to be trusted yet
@@youtubeusername1489 Humans are vastly better than self-driving software at handling situations they have no training for, because humans have general intelligence. They can use world knowledge and reasoning to navigate situations they've never seen before, and to correctly recognize when there is something unusual going on in the first place.
@@notnullnotvoid sometimes the body language of another driver lets me know when they're about to cut me off or change lanes without signaling. Can a computer do that? Napes.
Fellow engineer working with machine learning here--I think a lot of your specific criticisms could be addressed with GANs. Granted, there's still the question of "how can we know if this is reasonably complete", but I think you'll find that the more important problem with machine learning, especially when it comes to self-driving cars, is an engineering ethics question rather than a technical one. Eventually, we'll develop self-driving cars that are genuinely better drivers than people. How will we know when we've hit that point? Can we get there without experiments that risk the lives of other people? When someone dies because of the AI, who is responsible? These are really important questions that we don't have answers to yet, and we really should *before* we see *any* of these things on the road. No matter how good your neural net is, it's going to make a wrong choice eventually, and we really need to be ready for that eventuality before it happens.
idk these critics seem very negatively minded in my opinion. You make some good points about sponsorships and probably not enough negativism from Veritasium. But there are no really wrong numbers that are hugely misinterpreted. The point to blame is that most concepts at the moment are more niche uses of the driverless concept than one might expect at first. Btw did you really blame the car to be too cautious when someone is nearby? As long as the car does not stop forever and for every person nearby i would consider that braking fine
Man it is hard to watch this, with veritasium being one of my favourite channels, but in the spirit of learning, it pays to have a balanced view by hearing both sides. And after watching the video (and being a pilot myself) I have to agree that the Waymo video was a huge hype piece that left out a lot crucial information (like an overreliance on automation being the cause for accidents in the aviation piece for eg.). Another aspect of automation to think about this is "I know that I dont have to worry too much about the condition of my taxi's critical systems (brakes for eg.) because my taxi driver would have checked on them in the interest of self preservation." Its a new issue that would arise out of complete automation and I wouldn't want to be a part of the corporation's "acceptable risk parameters" statistic when it comes to safety and integrity. I'd rather be the cause for my own injury than be part of a cost cutting statistic. TLDR, I'm happy I gave this video a chance and will keep my head about myself when it comes to completely sponsored videos.
I think the recent Physics Girl's series on hydrogen cars was a pinnacle of quality degradation caused by sponsorship. Instead of making a video about the technical aspect of how hydrogen fuel cells work, she spent about 20 seconds saying that it's "basically magic" and moved on to praising them.
personally i have never watched an ad on youtube and even remotely thought about buying a product. matter of fact , if i see an ad for something on youtube? i will not buy from that company ever
yeah all the recent videos on hydrogen cars often ignore or gloss over a lot of the massive problems with the technology. They're just like "It'll take some new infrastructure being built but we can get there!" completely ignoring the actual sheer scale of that infrastructure, and the fact that one of the primary advantages of Electric cars is that they don't require anywhere near that level of infrastructure.
@@BlazeMakesGames Well electric cars require a gigantic amount of electricity infrastructure on the level which requires a ton of investment. Countries like Germany could not hope to supply even a 20% electrified private car environment when there's hogh charging demand. You're right though, hydrogen infrastructure is expensive and fuel cells are less efficient. I do wonder about hydrogen combustion engines though.
Funny enough that a video proporting about corporate propoganda struggles from the same issue. Granted this channel isn't an education channel nor was the entire video based on a sponsored product, but VPN's are almost a worthless product and here is Tom making a biased ad in support of one so he can get paid. Quite commical if you ask me. If you want to be a solution to the issue, maybe finding a useful product to be sponsored by (though they are quite hard to find nowadays), or not be sponsored at all and rely on Patreon.
I think that just underscores the fact that the real problem isn't that youtubers taking sponsorships, but that companies are willing to spend it and youtubers have to take it.
Yeah, not to disregard the points Tom makes in this and other videos by falling in a tu quoque fallacy, but VPN sponsorship segments are truly questionable in their affirmations. Tom Scot has a good video about that if anyone following this thread is interested.
It's a bit heartbreaking to see that ad in this video, yeah. VPNs are some shady, scammy shit these days and Tom is falling prey to all the things he's accusing Veritasium of doing. Basically, you have to treat any and all advertisements as entirely worthless - not just skippable or dishonest, but actually worthless, empty pieces of noncontent.
Veritasium also disappointed me by doing an interview with Bill Gates where he puffs him up as a visionary for predicting that a pandemic would happen and we wouldn't be ready for it, but doesn't ask or mention anything about how Bill Gates himself has contributed to making covid vaccines far harder to access in poor countries
@@sanders555 not sure, but I went back to the video and it turns out I was wrong, he does address the concerns people have with his interference in preventing the vaccine from being open sourced by completely uncritically allowing Gates to give a very weak defense of his actions. It's actually much worse than I remembered, avoiding the topic would have been preferable, I've gone from feeling disappointed to feeling outright betrayed
@@HamidKarzai It wasn't that long ago that I thought Gates was just another random target of the batshit birther QAnon cult before I realized there was something to it. It's a shame they're so off the rails that they're actually helping to provide cover for actual celebrity villains and their misdeeds by lumping them in with all their fairy tales of Presidential resurrections, pedophilic lizard people, and powerful Satanists with a huge predilection for green energy, universal healthcare, and authoritarian communism.
@@sanders555 if you were a conspiracy, wouldn't you want conspiracy theorists to all be insane? wouldn't you want to do anything you could to cause them to be more likely to go insane, so they can't actually figure out what the conspiracy is because they're too busy assuming everything is a conspiracy :p conspiracy is a real legal charge for a reason and it does occur, and it typically isn't as all encompassing as conspiracy theorists think. It's typically more mundane, people following incentives locally rather than actually working together in conspiracy, and when they do work together it's in groups that don't break the rules of reality. veritasium himself documented an example of an actual conspiracy, and like most actual conspiracies it was much more finite than a conspiracy theorist would propose. in general it should be expected that if there is a lot of conspiracy in the world it would be coming from many small ones rather than one big one. see also zoe bee's video on how to talk to conspiracy theorists.
Those self-driving cars just recently expanded into Downtown Phoenix. I've encountered them more and more on my drives down there, and they're out in full force during big events. The streets in DT Phx are 2 lanes wide, some areone way, and always have cars parked on the side. This is pretty usual for other major cities but in most of Phx its unusual. These Waymo cars regularly get confused, stopping a lane for hours. If a car is parked just slightly over the line, they stop, thinking the car is pulling out. During First Friday I witnessed a passenger prematurely leave a Waymo, which caused the car to panic and stop dead in its tracks. Both lanes of traffic broke down because people stuck behind the Waymo tried to pass it, but they couldnt get around the Northbound land people in the NB lane couldnt back up to let them pass because of how conjested the street was. I was stuck there for about 30 minutes before the technician was able to walk up to the car and get it moving, letting traffic flow again.
Since I can edit on mobile... Additionally, the comparison of Self driving vehicle AI systems to Aircraft control system is comparing two different beasts. All commercial flights are heavily monitored and follow generally the same rules of flight (not all countries have the same regulations, but the physics of flight are restrictive enough that the rules are generally universal). If we were to compare SD cars to aircraft in good faith, there would need to either be way less private cars on the road (or dedicated lanes for SD and private cars) or there would need to be way more private, unmonitored aircraft flying at the same elevations as commercial aircraft. Its just two entirely different beasts that you cant compare those systems at all.
The VPN ad spot is a weird choice for this video, especially all the talk of massive companies wanting your data (true) but then saying a VPN can help you rest assured you're not being tracked is complete bullshit, and just technically inaccurate. Most large tech firms (websites in particular) can build fingerprints of specific devices, so changing your IP does piss all for most people who actually want your data, along with this given the number of breaches coming from VPN companies you're actually better off getting sponsored by linode or something with their 1 click private VPN through openvpn, you don't NEED locations anywhere in the world to have a useful vpn, that's used to bypass content filters or restrictions which i don't have a problem with if they would stop acting like no one uses these services to watch netflix shows not available in their area. It stops websites from being able to track you? this is just completely false in most situations. because a vpn doesn't log you out, nor does it have the ability to do so even if it wanted to, and you're easy to track if you're logged into any website with or without a vpn. marketing vpn's as a catch all solution to internet security in my opinion is on par with the criticisms leveled at derek in this video, that's not even getting into the fact their basically pulling a fallout 76 by saying it's 83% off, and it always is, it's literally a tactic proven to push people into buying a product because "oh, that's a good deal, and it can't possibly be that cheap for long", and the idea that their service is 83% more valuable than they can sell it for making it seem worth more than it is. it would cost me $5/month to host a vpn anywhere a cloud provider has a location which is now most places in the world so they can't possibly expect much more than that for a closed source less transparent version of free software running somewhere in the cloud that you own completely, along with the fact you can pay monthly, often times by the hour for a cloud compute provider which means if you don't need it/want it anymore you can just turn it off, and you only pay for what you used, in contrast to being sold something for two years, which you likely only use for netflix, i doubt anyone actually concerned with internet security would use a public vpn although i could be wrong. also it's a little ironic considering youtube tracks people, they in turn get recommended videos to their taste (possibly this one), at which point they watch the video, some with ads on, you get paid by youtube, then sell them a product with the proposal it will stop them being tracked which if it actually could, would pretty quickly kill any channel recommending it because youtube wouldn't know the first thing about them and we all know the subscribe button means jack shit now anyway, along with the bell for the most part.
Indeed, as someone who works dealing with information security, uses and hosts his own VPNs all the time, knows what they are and how they work and their limitations. His sponsor spot is just as disingenuous as the waymo video. Is there a use for VPNs yes, is there uses for waymo self driving cars, yes. Are they what they are being made out to be in these sponsor pieces. No. Lets just say, if I'm using a VPN in a security that I'm not hosting personally, it is by no means the only measure I'm taking, or the most important and is only being used for specific circumstances.
Finally see this being called out. I noticed it a lot of times, I think primarily on LinusTechTipps, but never never went to comment on that as it's not a relevant part of the video, par of the course and most people probably don't care and it would be lengthy to explain what it does and what it doesn't (like your comment^^). Particularly in this video it's somewhat relevant to point it out though, being after the truth and criticising misinformation from other videos, but actually spread misinformation or half-truths in an ad. Also would be nice to see this comment further up, so more people get a better understanding of the use cases of a VPN.
the surfshark add is very funny in context, considering how vpns dont keep you safe or secure. it just swaps who is snooping on you from your internet service provider to your virtual private network provider. is good for flaunting laws ur government cares about that your vpnp doesnt, but dont expect to never ever get caught
@@raulinurminen7299It's not just an ad, it's actively misleading. Of course an honest person wouldn't bullshit you while exposing someone else, right? That's what makes the ad so valuable.
Really appreciated the video until the sponsored part. VPNs can be a part of a security-strategy, but claiming they protect your information in general on the internet is simply false. Most tracking is done via cookies and fingerprinting which a vpn is not protecting against.
I really appreciate this video but tom is in a really easy position to just make explanation video without any feedback video even after the person said give clarification back. Tom will NEVER wrong in a video like this even when CLEARLY he knows vpn is bollocks and still got that sweet sweet SPONSORED money. He in the position of everything he said is right even after said person clarify, that least it's not cancelation of the person in the video happened. This is the exact same way Magic of Rahat got accused and lose everything until 2 years after revealed he ALWAYS in the right. I just afraid this will be the precursor of it.
I hate to see those vpn misinformation ads, especially from creators that combat other misinformation. However I would also say that there are by far many worse sponsorships out there.
Yeah I also noticed this. I mostly agree with the overall argument that "veritasium did a bad"... but those claims about VPNs are not entirely true either. Having said that, both videos are entirely different beasts. Tom has a small sponsored section about VPNs and that's it, he doesn't make a whole 10 minute video saying "why surfshark good", which is basically what the waymo video is.
Hate it as well, but do we think he has the luxury of picking sponsors? Also fact fiend has shown how shitty these sponsors can be with the contracts where you have to say what they want otherwise you don't get paid. That's why every audible sponsorship has the person say what they're favourite book is and repeat the whole last line twice, cause they have to to get paid.
I think, public transport companies all over the world should do the same waymo did: sponsor educational yt-channels to make (good and well researched!) videos about the high benefits and all the problems, public transport would solve.
Honestly, this was really great as an introduction to the supposition of neutrality (political and ideological) not only in educational content but in the scientific field as whole, going forward i really hope this topic can be further explored
This hits way harder for me than I think it would for a lot of people bc I actually lost my dad in one of those “human error critical moment” accidents and had to face a similar sensationalization of his death and more specifically his blame in it from our local news when they covered it... my dad died after taking a curve too fast and rolling his car. He was also not wearing a seatbelt. When the news reported on this, both in the local paper and on tv, they put a huge emphasis on my dad /speeding/ around the curve and how dangerous he must’ve been driving to crash like that and how reckless it was bc he had a kid (my brother) in the car with him. They also REALLY emphasized that he wasn’t wearing a seatbelt in a way that kind of implied that he deserved to die for not “taking his safety more seriously”. They even speculated on whether my brother was wearing a seatbelt, basically trying to make my dad out as a bad, dangerous parent based on something they completely made up about the scenario. Critically, they didn’t include any ACTUAL information about my family or the accident other that the things I listed above and never asked any questions of my brother who survived and witnessed the accident, the state troopers who found their car, or the paramedics who worked on them. In reality my dad took a curve too fast bc rural back roads where I’m from are notoriously ill maintained, with no guardrails, huge drop offs, loose terrain, and vegetation or other debris blocking the street signs. In this case the curve was supposed to have a speed limit drop with a warning about the curve that had been overgrown by plants, had loose gravel on it that had just been left over from snowfall a few months before and reduced tire control, and overlooked a large embankment with no guardrails or safety measures to stop cars from rolling ALL THE WAY down it. His only failing in the accident was not being able to correct for a bunch of errors on the part of the state in maintaining the road that he didn’t know about until he was actively mid-crash. The reason he wasn’t wearing a seatbelt is less black and white that the cause of the crash but still very nuanced. My dad had previously survived an accident that killed every single other person in it when he just forgot to buckle up and got thrown out of the car while everyone else was crushed when the cage less convertible rolled. After that he was so scared and so traumatized he was genuinely /scared/ of his seatbelt. He knew it wasn’t logical and we all tried to get him to wear it anyway but trauma isn’t that easy to fix. He still insisted everyone else in his car buckle up. Which brings me to the last fallacious idea they threw out when reporting; my brother was in the front seat bc he was over the required weight, height, and age for a child to sit in the front, was wearing his seatbelt, and even told me after the accident that my dad had physically held him in the seat, putting as much of his body in the way of the impact and debris as he could to protect my brother. They basically demonized my dad bc they needed something to blame the accident on and knew they wouldn’t get much from focusing on the bigger issues of the bad roads that politicians in the state didn’t care about fixing and that citizens would support only until they realized that fixing roads and saving lives costs money that’s going to eventually come from them.... needless to say it made it a lot harder for our family to mourn the sudden loss and for my brother to go thru therapy while every news outlet on the tv, radio, internet, and paper in out small town was running lies about my dad to make him at fault.
Condolences. My younger brother also died in a head-on crash. I have often wondered if the oncoming truck had been automated, would it have been “programmed” for proper safe speed. Would the third car near the scene also have avoided the scene by following all “rules of the road?” We may never avoid all deaths, but seems like we have a safer future ahead of us when “on board” systems double check on the signs to keep us safer. Either way, we need honest and full disclosure of all options and have the FREEDOM of choice as to whether to protect ourselves. That said, what freedom should we have to endanger others? Perhaps our vehicles should override our choices when we fall asleep, DUI, or excessively speed?
my condolences, this sounds like an awful, heartbreaking experience. I'm very sorry it happened, the way the media jumped to traumatize a grieving family sounds horrid
Him using flight 214 as an example in this context and with those implications is OUTRAGEOUS! I don't believe in suing people for everything, but this gets me mad to the point I feel there should be some accountability! And to imply pilots choose not to use auto landing because of hubris or ego just shows how he knows NOTHING about aviation! Guess what, pilots need to train manual landing actually, it's pretty import in the case, you know, in the case the need to do it.
it's funny how many commenters here follow the same channels, and how many commenters are actually owners of those channels. And now I find fellow aviation enthusiasts, too? I guess UA-cam is better at connecting like minds than the FBverse.
I recall from the time of the 214 accident that the biggest contributing factor was that the airport had turned off the ILS system but kept the airport open while they moved the runway itself. This would be like running the Waymo service on live roads during an outage at the central map servers. The automation couldn't save them because the primary automation input was turned off. Also that particular runway isn't easy to aim for without computer assistance.
I agree that a better argument for self-driving vehicles, when it comes to weird trolley-problem-like scenarios, is that there's approximately no human drivers I'd trust to make a split second decision about who to run over either.
But as it is a split second decision it is very defensible. Someone programming who to kill in such a scenario without any pressure is less defensible I think
@@user-hl7ic7wc1r I don't agree. By the time someone has to decide it's too late, but they knew the risks when they drove thousands of pounds of metal into a public place at faster speeds than anyone can run. The programmer, not being under pressure, can more easily just instruct the AI to hit the fewest people possible, which should always be zero, and when it isn't that's when planning committees need to get together to find out why it happened and prevent it from happening again.
@@PM-wt3ye Just like this, that's how. Who's responsible or if there's a law isn't as important as how accidents can be prevented, which we're already aware right now usually result from humans not having the reaction time, situational awareness or ability to communicate they'd need to avoid them. It's easier to get humans out of those situations than to turn everyone into super-cyborgs...
False, it is the MOST important and most basic thingof all. If there is no law in case something terrible happens, theres no need to prevent it from happening, like in any other situation in the world. In case of an accident, there has someone to be hold accountable for. Denying the impotance in this situation is ABSOLUTELY rediculous and shows, that we are FAR away from establishing this correctly...@@amymason156
Man it’s really disappointing to learn that some educational channels you thought were trustworthy do stuff like this. I really appreciate these videos, they’re a stark reminder to always remain critical of the information people give you
I have to steer away from everything vlogbrother-related for such reasons. Hyping Gates, war criminals like Obama, telling their audience Biden is wonderful even after the main election ...
They literally state they are being sponsored... How do you miss that? Even this video is sponsored. That should always factor into your mind when watching a free video. Anytime you see a company giving access to an outside person, you should be critical of what you are watching. It does not take away from the point of Veritasiums video, that self driving cars are and will be safer than humans. And I am a skeptical person about that myself because of all the odd roadways and signs and missing road lines, etc.
@@NandR you expect them to be as neutral as possible even when being sponsored, that's what being an educator should be, I haven't seen the waymo video but I'd expect pros and cons of the technology with the least amount of bias possible, but I guess that's asking too much of today's UA-cam
What's worse is in his reply to this video he's doubled down and what's worse is he implies he's just as uncritical to any other cause he deems noble, so yeah steer clear of this man.
Yeah. After watching Adam's videos for a few weeks, I am honestly baffled how a lot of suburban problems can be effectively tackled by improving public infrastructure. Guy single handedly change my mindset
@@felixstenson1654 No, Adam Something on youtube. Hungarian youtuber who makes great arguments for public transport over even more cars to solve the congestion problem.
I like you Tom, but was this really the best video to include a paid promotion about a VPN where you add misinformation about what they're able to do for you?
@@kirklandday The issue is that he, like many creators misrepresent the capabilities of such VPN services. In that regard it's pretty damn close to what he critizises. The only difference is that it's an ad break and not the entire video.
@@shiniri2647 That's kind of the point though. This video isn't being made to say "sponsorships are bad", it's just to encourage larger 'educational' content creators to draw a solid line between what the commercial bit and 'educational' bit is, which Tom does very well with his VPN sponsorship.
If he made an entire video going on about how good VPNs are, that would be the problem but he isn't. I do see the slight irony but I don't think it takes anything away from the point of the video
"Human error causes 94% of traffic incidents, therefore AI should drive intead of humans" There has been not a single traffic accident caused by a body pillow behind the wheel. Therefore, all cars should be driven by body pillows.
The most notorious "educator" in Italy is probably Piero Angela (now 92 and still very smart and lucid) he always worked for the public television and when a private offered him considerable more money he refused to keep his work as independent as possible and said "I eat two times per day, I don't need more butter or more meat"
Yeah, same thing with electric cars. Electrifying many forms of public transport could be hugely beneficial and cost effective (slower speed, central hubs makes charging easier, higher purchase price is amortized over the fleet, etc.) but the focus is almost all on consumer electric cars because that's where the money is and because Americans see it as their flashy new tech.
You Europeans don't understand a thing. A big portion of the world doesn't have the climate for people to walk outside at 9 am and 12 pm for half an hour without showing up to work like an unshowered mess. If y'all don't invent driverless cars to sell to us, the rest of the world will.
@@mmmk6322 shhh let them shot themselves in the foot. Qhen the state mandated public transport gets limited to some arbitrary gvernment mandated standard, they will still support it, but the schadenfreude will be wholesome
That already exists actually. The Copenhagen metro is self driving and most high speed rail systems are basically fully automated and have been since their invention. As it turns out automation is a lot more feasible when you're in a controlled environment like a rail vehicle.
Tom, I’m happy to receive your constructive criticism, but I’m disappointed you didn’t fix any of the factual errors we alerted you to via email before you launched this video. Examples:
23:42 You cherry-picked this quote to make it seem like the NTSB blamed automation for the crash, when the report focuses squarely on human error: “The National Transportation Safety Board (NTSB) determined that the probable cause of this accident was the flight crew’s mismanagement of the airplane’s descent during the visual approach, the Pilot Flying’s unintended deactivation of automatic airspeed control, the flight crew’s inadequate monitoring of airspeed, and the flight crew’s delayed execution of a go-around after they became aware that the airplane was below acceptable glidepath and airspeed tolerances.”
32:37 Self-driving cars have maps including traffic control so they would know where stop signs are meant to be even if road markings aren’t there or stop signs are obscured. Plus they have better obstacle detection and avoidance than human drivers.
39:16 I’m not saying rare accidents don’t happen, I’m saying they happen less often than common accidents, many of which could be prevented by self-driving cars. I sent you an academic paper that recreated in simulation 72 real-world fatal accidents that occurred in the area where Waymo operates. In almost all cases the accident was avoided or mitigated by the Waymo driver. Why did you omit this study?
47:03 It’s well understood that autonomous cars properly coordinated could reduce traffic because they don’t have the same reaction time delays as humans. For example all cars at an intersection could start moving together instead of one at a time as we currently do.
47:10 We don’t have to increase the car utilization rate to 100% to reclaim significant value. If cars were parked 90% of the time instead of 95%+ we would only need half as many vehicles.
Isn’t it ironic that a video purporting to call out misinformation itself contains so many distortions and factual errors? (Which we pointed out in advance but you didn’t feel compelled to fix)
On the issues themselves, I like public transport. I also ride a bike, and enjoy walking to get around when it’s practical. But cars will be a part of the transport mix for the foreseeable future. And it’s my opinion, based on the evidence, that roads will be safer the more cars are driven by computers than humans. No one has to pay me to tell you that.
Hiya,
As I said in my reply to the email from your team, my decision to make this video came from a place of disappointment rather than animosity. As such, it’s a real shame to see you again coming out swinging in response to my points rather than the video giving you any pause for thought surrounding your relationships with sponsors.
I’m glad to see that you've taken on board my critiques of your initial rebuttals to this video and attempted to strengthen them. I imagine you know the flaws in these responses to what you mistakenly call "factual errors" but I'll do you the courtesy of replying to them anyway.
23:42 You accuse me of cherry-picking here. However, the very next sentence from that which you quote states that 'Contributing to the accident were (1) the complexities of the autothrottle and autopilot flight director systems that were inadequately described in Boeing’s documentation and Asiana’s pilot training, which increased the likelihood of mode error [...]'. The reality is that, as in your original video, you're working on the false presumption that humans and automated/autonomous technologies are in conflict with one another. However, as is the case here, such technologies are always going to involve human interaction and they therefore need to be designed in ways that ensure that those interactions don't lead to accidents (as the Acting Chairman of the NTSB states in the quote included in my video). You use the crash of Flight 214 as an example of human failure with the implication that automated systems could have avoided the crash; it is disingenuous and distasteful to have eliminated all mention of autothrottle from your version of events just because it didn't suit the argument you wanted to make.
32:37 It is, again, completely disingenuous to refuse to mention the maps which Waymo vehicles rely on when it suits your argument during your initial video but to now hold them up as a vital part of the technology. You spoke as though those maps didn’t exist in the video and so it was only fair, in critiquing your video, to respond in kind.
A further point for consideration here is the extent to which creating maps/scans for the entirety of even the United States (let alone other countries too) is practical. That seems like a highly intensive task which would be very costly, possibly to the point it’s unworkable.
39:16 Two points here:
Firstly, you do not say in your video at all that the kinds of accidents considered when people discuss the ethical implications of driverless cars “could be prevented by self-driving cars”-you say they’re rare and quickly move on.
Secondly, the “academic paper” your team sent over was a write up of a study produced for Waymo and written by Waymo staff. I can’t seem to find any instance of it being published in a reputable journal (or a journal at all for that matter) or of it being subject to peer review. It is also, as one might expect, based entirely upon accidents which occured in the few suburbs of Phoenix where Waymo cars presently operate (and have had years of “training” in) and so it would be improper to suggest that its findings would be representative of a wider roll out to locations that Waymo’s cars have not been fine-tuned for or where conditions differ.
47:03 While they might have an impact, the idea that these tiny bits of time saved here and there would eliminate congestion is optimistic. The more extreme version of this (included in the CGP Grey video you’re referencing) in which all autonomous vehicles “talk” to one another (or even just to the traffic lights) would require such a level of implementation and standardisation that it is unlikely even in the medium term to the point where it’s a fun thought experiment but little more.
47:10 I make no such claim that we would need to “increase the car utilization rate to 100% to reclaim significant value” in my video. I merely state that the use of the 95% figure is misleading as it ignores several important caveats (like the existence of the nighttime). Even a figure which related to the percentage of the daytime that cars go unused would have been more useful to use here-it may not have given you such an impressive figure but it would have been a fairer representation of the reality.
To reiterate, suggesting that my video contains “distortions” and “factual errors” is deeply disingenuous. Nevertheless, I know that, whatever the circumstances, someone suggesting that you’ve acted improperly is a horrible experience. I hope that, with the passage of time, you’ll take on board some of the thoughts provided here about your relationships with sponsors.
Best wishes,
Tom
Apologies for the delayed response by the way, it was the middle of the night here.
@@Tom_Nicholas hello, I think you should pin this in comments for people to see. Feel free to delete my reply afterwards.
Why isn't this pinned?
@@RazorbackPT UA-cam only allows you to have one pin on a video. Tom did make this a community post, and tweeted it out.
That's more than enough due diligence imo. It's not like Veritasium is gonna link this video in a pinned comment on his original. Pinning responses not a reasonable expectation.
I would like to think that my channel is one that promotes the solutions you were getting at near the end of this video: public transportation, cycling, and walkable cities.
I worked my whole career in tech, but I'm a firm skeptic of self-driving cars.
One of my biggest issues with them is that a lot of the problems they aim to solve are "American" issues, that stem from decades of car-dependent cities. Yes, if you design a city such that everyone needs to drive everywhere all the time, you have problems of crippling traffic, high transportation costs, and dangerous roads. None of these issues really exist in walkable cities with high quality public transport, however.
So I see driverless cars as a highly speculative and expensive solution to a problem that really shouldn't exist in the first place. Not to mention the potentially seriously negative implications of having city criss-crosses with high-speed, high-traffic roads full of autonomous vehicles.
To me, self-driving cars really seem like a desperate attempt to maintain car-dependent places and car-centric suburbia with a flashy technical gadget, rather than an actual solution to the problems facing cities.
I'm guessing you're not sponsored by people who also promote public transportation, cycling and walkable cities, which I think was Tom Nicholas' point here. It's a good solution that people would and should support, but they don't because the private sector isn't going to throw their weight behind something that doesn't make them money.
Edit: This came off as a little antagonistic. For the record, I'm not trying to dispute anything you've said in the slightest.
I think the point was less that nobody is willing to make videos on it, but that there is much less money and sponsorship opportunities in making them. As can be seen in comparing your channel to the likes of Wendover or Real engineering that rake in millions from sponsorship deals for praising car infrastructure and plane travel.
I really love this crossover between two great channels!!
The whole vid I was thinking "pls comment on how cars (of any kind, be it electric and/or self-driving) is not the ONLY or even best option for our future" and he did :)
@@seanrezash7371 yeah REAL engineers DESTROY the environment with FACTS AND LOGIC
The single biggest difference between pilots and the average car driver is training. It is a LOT harder to get a pilot's license and is fairly expensive, compared to getting a driver's license.
I’m in an English class right now where a big part of it is finding and scrutinizing sources. Usually you want to look at both sides of an issue and then argue your point with the one that wins you over. That’s impossible when there is only one source, and it’s paying you to say things.
id lile to add: and even once your won over you should not just suddenly hide away any counterarguments but rather argue them openly
homestuck
We shouldn't stop at "both" sides. There's often more than 2 options and stopping at that will make you lose sight of better solutions nobody is platforming, making the "middle ground" between those two options unfairly appealing. Plus it leaves any point where both sides agree basically uncontested, limiting everyone's decision making.
In the case of self-driving cars, thinking of "yes vs no on self-driving cars" (which is an interesting question by itself) often loses sight of alternatives like public transport, getting rid of car-centric city planning, de-urbanization, etc. Even the really interesting point of accountability (who is accountable after the AI makes a choice in a trolley problem situation?) can be made a disservice if all sides just assume our justice systems don't need corrections.
Lastly, even the pro self-driving car movement tends to make itself a disservice by limiting what the AI can do. Calling them self-driving cars makes most people lose sight of all of the non-car stuff that can be made more useful by moving itself autonomously on a 2D plane. Every one of those cases can, and should, have its own "yes vs no" crowd.
This is great and needed to be said. I worked with Gates Foundation. Don’t think it quite fits into this narrative, but I would say that I wouldn’t work them again.
Full integrations are a signal you shouldn’t trust a channel. I see some multi millionaire creators use the excuse of “this helps us fund future videos” as an excuse. That’s bullshit. Do sponsorships fund content. Yes. Do you need to sacrifice your integrity to work with a sponsor? No
Well put, and glad to see this sentiment from one of my favourite content creators.
Gates foundation 100% fits into this narrative, it's one of the champions of sponsored propaganda. And I know "propaganda" has a very harsh conotation, but it is, at the end of the day, what the videos sponsored by Gates Foundation are, whether you agree with it or not, propaganda is propaganda...
I was throughly dissapointed in Kurzgesagt...
Thank you, that means a lot. As I said to a few people in the chat during the premiere, making these videos comes from a place of disappointment rather than animosity. I continue to believe that there's real potential (much of it already realised) in educational UA-cam (and the broader free dissemination of education online, however the costs involved in dethroning UA-cam means that it's unlikely to be topped anytime soon). To make their points sufficiently, these videos do kind of have to focus on a case study, however I genuinely wish the team at Veritasium well where they're not engaging in these practices.
With the full disclosure that I've been looking at making a video about Gates' use of sponsorships to shape the discourse on YT surrounding responses to climate change, that's interesting to hear about the Gates Foundation. I know Our Changing Climate has said similarly.
I love your videos and respect you greatly (even in admitting you wouldn't work with the Gates foundation again.) Do you think it might be a good idea to even give this information to your viewers in maybe a pinned comment under the video? I think a shift in perspective like this on sponsored content is definitely worth informing viewers about.
@@Tom_Nicholas my complaint about Gates Foundation is primarily that there are some shady things the foundation has done that I was not aware of. Secondary problems were primarily about how unresponsive and inflexible they were. The video I made was already written when they approached me and I didn’t change the script to fit with them. It was a natural fit for a sponsorship, and yet they still somehow made things difficult. I put that down to them not understanding how influencer marketing is supposed to work.
"people are using youtube for education" this is true, and a large part of the problem with removing the dislike count. removing the dislike count is anti-information/education, pro-propaganda.
Corporate interest has always involved keeping us as uninformed as possible in order to sell us lies and unneeded shit
This reminds me of the extremely democratic voting process of the Democratic People's Republic of Korea
@@zucc4764 Omg the most Democratic nation in the world Glorious North Korea!
@@78anurag I love it when they democratically and fairly ensure that everyone votes for the correct candidate!
@@them2545 Yes!!!!
There is a really good reason why wikipedia has been so defensive about it's policy to never take corporate sponsorships.
The day that changes is the day we are all truly doomed.
Spare a moment for the poor dude who saw Veritasium tweet at Tom regarding this vid and commented, “oh boy this collab will be awesome!”
🙏
They deleted it lol. Poor guy
😬
@@stonium69 what did the tweet say? :D
🙏🏾🙏🏾🙏🏾
Tom, thanks for this video. I agree with many of the points you raised (I’m bearish on autonomy as a whole (and highlighted as much in my own video)) and while I very much respect Derek and his team, I took issue with the way in which he rolled out his video. You postulated at around 25:45 my shot at Tesla was drawn from the briefing-it was not-I’m a Tesla owner with extensive Autopilot miles and frustrations from Musk’s promises. Regardless, I was actually surprised Muller’s video stated nearly exactly the same. My video was published a number of weeks before his but I’m not saying he plagiarized or anything. Likely a weird coincidence.
I’m certainly not without blame for my own video and if given to travel back in time, I wouldn’t have done the spot with Waymo. I was frustrated with the amount of editorial correction the brand attempted to make to my original script (I think we went through four edits) and it was frankly neutered of my opinions and dissent. Despite much push-back (and a fair number of changes they requested I outright refused to make), the script *was* largely edited and in retrospect I wish I had been more firm. The large majority (90%) of ads I do on my channel are integrations like yours on this video. The remaining 10% are dedicated ads and I do my absolute best to retain editorial control so that the videos can still be interesting and informative (one example I’m very proud of is our video tour of a fiber internet company’s infrastructure) while being very explicit and clear with viewers in the first few seconds that the video is, ultimately, and ad and thus inherently made with the brand’s input (this is an attempt for viewers to stop watching if they’re not interested in such content (I wouldn’t be myself)). Where I erred with Waymo is doing this type of video on such an ultimately controversial topic and one for which I have many of my own opinions I was unable to voice due to more heavy-handed editorial control than I was led to believe based on our contract.
I should have fought harder to keep my dissenting opinions in the script. Frankly, I just got really sick of their requests, I have an incredibly small team (just 2 of us) and it was taking too much of our time and I was ready to write the video off as a loss and just rip it off like a Band-Aid but I regret it. I should have done better and I will in the future. This type of content is always tricky so I have (and will continue) to discourage it whenever possible.
Thanks for your insight and wise words. Keep on keeping on!
Amazing to see your thoughtful response, in stark contrast to Vertasium doubling-down, acting defensive, attacking small details of the video meanwhile completely ignoring the main argument.
Much respect Snazzy!
This was a very thoughtful and humble response, and it’s good to see that Tom’s critiques didn’t fall on completely deaf ears
Thank you. I really appreciate you taking the time to leave such a thorough and thoughtful comment. Hope you’re having a great week!
Thank you for your measured response, it is appreciated.
By you having said that, I in no way think less of your channel, which I love, or your person. I think the way a humans mind is made up, I now might even like you more because of admitting to a mistake. This makes it even harder to understand the aftergoing actions of the Veritasium team, which try their hardest to intimidate, just like a corporate lawyer would.
"Just build a train"
-Adam Something
And this work
Ah, a fellow enthusiast of Mr.Something😆.
As someone who has never rode a train, I too am leaning on we building them, even if only so I could try it.
@@OLucasZanella then go to west central Europe and see how this can work, in Poland we still sometimes use EN57 from 1991 on local route's
I wouldn't be surprised if I came across him in the comment section
17:50 i love how the poll itself is slanted like "i can't wait!" and is not mutually exclusive with "they're not ready now". also, it (along with "they're terrifying") makes the poll about your personal excitement and not about your honest assessment of the tech.
i'd love to see a poll where the options were neutral and more specific, ie "i think they're ready to be used now", "they will never be ready", and "they will be ready in >10 years". i imagine they may not be as optimistic.
Good point. I suspect the “give it 10 years” category makes huge gains
not only that but the 'positive' option is intentionally placed at the top, meaning people read it first
Congratulations, you've discovered the intention of polls.
Gaging public opinion.
Good job.
As you know Tom, I really appreciate your efforts to draw attention to this growing trend of educational UA-camrs allowing their trust and audience to be leveraged by corporate agendas, perhaps even unaware themselves that they are being used in this way. Many of us esp in the sciencey space have discussed this privately and feel full integration is not a good idea, but I’m not convinced the audience really sees the problem, suggesting they simply believed their favourite UA-camr’s educational video about Bill Gates, the WEF or Waymo
It's really nice to see you here doctor. Glad you also enjoy Tom Nicholas.
The nefarious Bill Gates has been bankrolling some of the creators on the platform, and that’s not cool considering how much influence he already has bought himself in mainstream outlets. More needs to be done to counterweight his heavy spending.
Oh, I went through Veritasium’s back catalogue as I remembered some others along these lines. Oof there have been a few 😬
Popping up here after your appearance with Tom Scott. Who shall be your next UA-cam Tom?
More seriously though, it's kind of heartening (apologies for the pun) to know that influencers are discussing this behind the scenes if nothing else. I know the dance to keep food on the table as a content creator is always hard, and so a lot of the potentially sponsor scaring talk has to happen in private, but knowing it's happening at all is a relief.
+
Really appreciate how you put this, Tom. Creators have discussed this a bit in private, and I think you handled it really delicately and professionally in bringing it to the public eye.
I love how cute and sinister “discussed this… …in private” sounds. Almost as something from “what we do in the shadows”. Council of high youtubers in special robes summoned for discussion of highest matters.
On serious note - every creative community should have healthy and reasonable discussion of limits and representation of commercial interests for reasons of clarity and objectivity.
I agree Tim! This corporate influence really worries me. It's time to set some harder boundaries, especially for those who consider themselves 'educators'.
@The Program _Someone_ is always responsible for the finances needed to make a video, even if it is just the person in front of the camera and their personal wealth. The problem with corporate sponsorship is that it is both pervasive and without care for the media it co-opts. You are correct that other groups can have a similar destructive influence on the reliability of the medium, but each type of organisation has a different way of instilling bias. A non-profit or a university (and some governements) will often have the interest of the general public in mind whereas activist groups rarely have the power to move over media creators that don't already align with them.
The methods Tom Nicholas uses here to draw up scepticism and doubt reminds me a lot of Tucker Carlson on FOX. Injecting their own interpretation of state of mind and intent to something somebody said/did, and then criticising the morality of that intent. Classic straw man argument. Tom's dive into Derek's comment about planes landing was an especially egregious example. Automation reduces opportunities for human error. That's all. That's the message of that segment in Derek's video. Derek didn't say the pilots were "too egotistical", or "autonomy is good, and humans bad", as Tom puts it. It's just rhetoric Tom uses to put words into Derek's mouth in order to invoke emotions of disgust in order to solicit agreement from viewers when he rejects moral outrage he created himself.
You see this technique used a lot by politicians.
"So-and-so did ABC! The implication is he thinks you're idiots! Well, I don't think you're idiots! You're all the best!".
This video is pretty much just Tom tearing down his own straw men.
Then, peppering the viewer with suggestive questions, along the lines of "Can we trust ABC if they didn't do XYZ? I don't think I could if they didn't! Why did they not do XYZ? Was it because they're trying to hide something dubious? We will never know".
I can see how Tucker Carlson has such a following with these techniques.
With that said, Tom is very *VERY* good at rhetoric.
His essay-writing skills must be off the charts.
@@benc5528 yeah, it takes real mental gymnastics to go from "guy is paid to advertise automation technology" to "guy wants to make automation technology sound good"
I personally think the DNA video was more revolting, I'm still hoping someone makes a response video to that.
That was just straight up sacrifice your liberties stuff, really worrying since you just need enough people to get everyone.
@@InsecureCreator People need to understand this. Gather enough correlations and it becomes trivial to fill in the blanks. And even if you deliberately exclude certain types of data, those can be reconstructed through those correlations - and that is exactly what happens, automatically, in many machine learning algorithms.
This problem has been documented already in the field of automated human resources (hiring and CV screening).
@@jimcrelm9478 “no no, we don’t want to only hire men, stop that computer”. “Okay, I’ll just teach myself how to infer gender from names if I can’t use the gender field directly! I’m so clever huh?”
@@kaitlyn__L "you often hire people with other typically male traits, so I'll guess I'll hire based on those"
Derek was not talking much in that at all, it was the other people most people will find it odd and the more important thing was if your relative giving DNA to use for investigation purposes will infer on your privacy
As Tom Scott once put it on his own channel: "You cannot trust me". Not because he has ever maliciously spread incorrect information but because he cannot ever guarantee that the information he provided is really correct and that he didn't make a mistake. He's trying his best to avoid it but he realized that there is no way he can avoid it all the time. It's novel approach to tell your audience: Keep in mind, I'm human and I make mistakes, so take everything I say with a grain of salt.
Nah. Socrates did it, too.
@@Duiker36 You put Tom on the same level as Socrates? Okay, *now* you are flattering him maybe a bit too much ;)
@@xcoder1122 Fair point, Socrates was no Tom Scott
@@WaterYaDune man can only be killed with words which do not exist
its a somewhat novel aproach, but it shouldn't be. I think its the only way foward, be it from a youtube channel or a big newspaper. Brutal honesty, near-suicidal transparency, and admission of fallibility and bias.
Thank you for this video Tom. As someone who has worked in media through the transition from print to online models and have watched the surprisingly fast creep of sponsored content making its way into what is presented as real content, it’s frankly shocking to me that audiences don’t understand the problem here. Trust is a valuable resource in media and to see it sold so easily is a shame. Anybody who can see the problem with the current distrust people have with media, anybody who gets frustrated trying to convince that relative of theirs that objective reality exists and facts are not a matter of opinion, anyone who wants to be able to agree upon a shared reality so that we can make meaningful, science-based decisions as a society, needs to understand the importance of trusted media sources free from bias.
If people haven’t noticed, we are currently in a battle to save the very concept of shared reality and truth, and it’s not clear which side is going to win here.
I would have thought @veritasium is among the people who understand the value of truth being backed by integrity. It’s sad to see him discard the decade of trust he’s built up so readily and harm the state of discourse and trust in media in the process.
My cooking hero over here dropping pure wisdom. Love it. Hopefully those of us that value a shared reality won’t lose this one, I am optimistic when I see leaders like you advocating for the good fight.
Thank you, I really appreciate you saying so. It does always shock me how different the perspectives on this are from those with more of a foot in “traditional” media and those with more of a foot in online media. I can’t remember which US website unionised a while back but it was interesting to see that one of their demands (alongside better pay and conditions) was the right to refuse to work on advertorial pieces. There was a real sense that the writers felt that such work made them feel uneasy and diminished their sense of professional self-worth. By contrast, many writers/video creators/etc. who haven’t had experience in that world seem to jump with enthusiasm into working in this way.
Also, love The Food Lab. The Ultra Gooey Mac & Cheese is a regular in our house!
@@Tom_Nicholas thanks Tom! I’m so glad you enjoyed that recipe brought to you by Real American Cheese™!
Keep up the always excellent work.
I've found mozzarella works as a decent substitute. Still have to ask Siri how many grams a cup is every time though!
@@Tom_Nicholas my next book will have volume and mass in both American and metric units built in!
On UA-cam propaganda, could you also do a video in the series about VPN providers? They don't really provide protection and the companies paying for these ads are spreading misinformation. Seeing your favorite UA-camrs make an educational claim about VPNs that's wrong can have negative consequences too.
For example, they only change an IP address, but they don't actually affect identification. That's done with behavioral and browser profiling. Your traffic still goes through an ISP, even if it's not your own. Commercial network connections can have a lot more tracking and fewer protections than your residential ISP.
I appreciate your video, and this isn't really a criticism of it. It just seems to be in line with the problems your video shows.
Tom Scott had a video about VPNs talking about what you've described. If you haven't already seen it I would recommend checking it out.
The thrust of your point is absolutely right. But there is a complication in that some of these services are more than just a VPN alone, their apps also block tracker JS and other things on the web layer too. The marketing spiel glosses over all the complexity, because of course it does lol.
My man has to get a new source of funding before he can do that haha
That’s why apple’s iCloud + with built in private relay is the answer.
He did put “fluff” in the subtitles when the ad started, so he could(?) be aware? I think it’s understandable to do little ad bits that just tell you to do something like buy a product, but whole videos of uncritical support are the problem.
The juxtaposition of "most experienced driver in the world" with "can't drive in inclement weather" is just too good. Really is an apples to oranges comparison.
"What we call Man's power over Nature turns out to be a power exercised by some men over other men with Nature as its instrument."
― C.S. Lewis, The Abolition of Man
I love veritasium, but I noticed this sort of thing with him lately. I was more concerned with his video on DNA databases, since it was fairly one sided towards "this is a good thing, actually. Catching criminals trumps privacy rights" without much attention paid to the human rights issues of the subject, while the video was clearly meant to be objective and educational.
@Shimmy Shai "Best to let a thousand guilty go free than one innocent be put to death"
It rather depends on how guilty and of what. Many wish Herr Schickelgruber had not be set free.
what's wrong with a DNA database? why would someone feel uncomfortable with it?
I noticed that too, and there were LOTS of people pointing out the bias in the video. It's a dangerous precedent.
@@MrDarren690 Ya, if you're a criminal. Ya, if they start using it for everyone, but then that's just paranoia talk.
@Shimmy Shai Uh, did you read your comment before posting? "Best to let a thousand murderers go free than to put 1 innocent to death." No, that's why America is such a terrible place with almost nothing but criminals. You're kinda dumb if you think like that imo.
Just watching this video now. I wanted to point out that there was a Veritasium video in the past that was sponsored by a paper towel company, and as part of the advertisement the host creates an "experiment" to measure how much bacteria are deposited when reusing a cloth towel versus a paper towel. The conclusion of this advertisement/experiment was that paper towels are safer because they don't leave behind bacteria. As a proponent of zero waste practices I found the whole thing to be rather disgusting, because there was no discussion of environmental impact, and it was disingenuous because I use cloth rags, but I keep a pile of clean ones and I can always grab a fresh one if needed. This advertisement was worse than just saying "buy these paper towels" because it created a rigged experiment by a trusted science youtuber, giving much more weight to these claims which were not properly discussed. So I am really glad you are digging in to this.
There was another one about dandruffs sponsored by that one shampoo company too. Created quite a stir, though I don't remember feeling the content being too dishonest at the time I watched it
@@moscanaveia yeah but they normally feel fine when you're watching it, especially if you're not normally exposed to the subject. they're well presented and follow a superficially logical path. you can really feel that they reached these conclusions in a good way, which is exactly what makes this such effective and insidious propaganda. its only with further research and fact checking we find out. though for all i know, the dandruff science is legit, buuut I've not looked into it. and i may never look it up, and this might have never come up so I wouldn't even have thought about this again.
I've been seeing ads saying that lately
yeah the paper towl one was worse than the weymo video.. at least weymo is still developing its product... but the paper towl video sent the messsage that there was no alternative.. and didnt propose any better solutions or further required research
Yeah that paper towel video pissed me off
Another point in the "20M miles driven" argument is that comparing the distance traveled by Waymo cars to a human driver is completely pointless. I could write a program to make a car self driving and send it out to drive for 100M miles; if the program is incapable of learning and I don't update it in any way, it's exactly the same as when I sent it out having driven 0 miles. Waymo presumably does update their software regularly, and it may very well be capable of learning to some degree, but that doesn't change the point that saying "we've driven 20M miles, that's more than any human so we're more experienced than any human driver is" is completely ridiculous. The company/program simply does not learn in the same way a human being does.
Learning is a non-sequitor. Humans crash at a fairly predictable rate. And new inexperienced drivers are entering the road all of the time. Even experienced drivers get distracted or tired. So long as this program is consistently less crash-prone than an average human, that's an improvement! In addition self driving cars CAN 'learn' better than humans: The data of a crash can used to make the next version of the program better, compared to a human driver that may not survive the crash and thus won't have a next time to apply their knowledge.
Excellent point. 20M miles means something much different to a computer algorithm than it does to people, but we are expected to value them equally. False equivalency.
@@Brickcaster but none of what you said is directly related to miles driven, which was exactly the point.
Additionally, I could write a program in an afternoon that could drive 100K miles way safer than the average human, so long as those 100K miles were on salt flats. It would be better to compare the accident rates of taxi drivers, since the composition of roads driven on is more similar to these cars.
When one hears that a person has a significant number of years of 'experience' doing something, there's an assumption that there's a not-easily-articulated nous that the person has developed. Using the word 'experience' for these cars is disingenuous.
"Technological fixes are not always undesirable or inadequate, but there is a danger that what is addressed is not the real problem but the problem in as far as it is amendable to technical solutions."
(Engineering and the Problem of Moral Overload)
"If machines produce everything we need, the outcome will depend on how things are distributed. Everyone can enjoy a life of luxurious leisure if the machine-produced wealth is shared, or most people can end up miserably poor if the machine-owners successfully lobby against wealth redistribution. So far, the trend seems to be toward the second option, with technology driving ever-increasing inequality."
Stephen Hawking, 2015 Reddit AMA
Exactly
Stephen Hawking do be sounding like Marx, pretty based. (Or even Keynes in this case)
In other words, automation is an undisputed good, but only if we get rid of capitalism and unjust hierarchy. Well said comrade Hawking.
@@Pluveus Only in capitalism would automation be a bad thing. It's astounding. There are so many things I just can't get down with until we're rid of capitalism, like anarchism, automation, UBI and a host of other policies.
Errico Malatesta wrote something similar to what Stephen Hawking said in *1883*
I'm afraid the trend won't change any time soon :(
This was great. Well done, Tom. As someone who grew my channel in the “science” niche before making my switch to political content, I’m very familiar with the full integration problems in that community. I’m glad to see someone taking the time to really discuss it.
Love your videos, keep it up. Much love from Jamaica
nice to see you here man.
Based as usual, second thought
Bold youtubers like yourself and Jake Tran really opened my eyes to the flaws of the Corporate-ocracy and the detrimental state of capitalism. I’m a tremendous fan, thank you for sharing such critical information. Thanks to you, I turned Socialist.
Thank you JT! It’s certainly not something that’s entirely confined to “Science UA-cam” but it does seem slightly more prominent there.
YES after watching the video on self driving cars I was shocked to not see a single word of criticism in the comments. I’m so glad you’re addressing this video!
i tried :( super buried
You were shocked not to find criticism in the UA-cam comments, well known as being the bastion of rationality and critical thinking?
@@サンゴ礁Scleractinian It's also a bastion of shitposting and trashtalking, so yeah, even then it is suspicious XD
@@サンゴ礁Scleractinian LOL I guess I’ve been expecting too much of UA-cam comments
@@internetdumbass It's a shame UA-cam makes it very hard to search the comments.
In theory it is possible but only by loading all comments (which can take quite a while). I'm affraid this won't change anytime soon because YT has no real concurrents; as long as their website generates revenue, why would they improve it?
I sent feedback to YT asking to make the comments searchable and I wonder if it was ever read by a human.
On the subject of driverless cars: There is a good Kyle Hill video that talks about how an ai beat the world champion at the game Go in 2016. But in 2022, amateur go players were able to beat that ai consistently. He brought up a great point that computers are incredible at recognizing patterns, but ultimately don’t understand what a game is, what a board is, what a piece is.
This is the thing that scares me the most about automation in anything involving humans: we assume that computers can make leaps in logic like humans. It comes as naturally to us as breathing. Ai looks like it does on the surface, but it is relying entirely on a predetermined information set. Taking that for granted is extremely dangerous when it comes to the unpredictability of humans.
The fact about Go is nonsense. AlphaGo is not available for ANYBODY to use! Unfortunately, Google has kept it under wraps. AlphaGo is miles above most of the existing AI programs. The same program also claimed to beat Stockfish handily. And Stockfish is already the overlord of Chess. You may not like it, but AI is superior to humans in more ways than you can ever imagine. Its still not perfect in being a general intelligence but that day is not too far off either. AlphaGo is actually meant to be a prototype GAI and its coming.
Personally, I am not too happy about what AI would do to our society. But, I am not going to be ignorant about what is coming my way either.
Replacing human work with machines is great, if the work and situation fits it. Repetitive tasks, common safety check etc. all those thins are fields where machines excel because (assuming correct maintenance) they don't miss anything that is presents to them.
Can driverless cars make roads safer? Yes. Under one condition: There are only self driving cars and you don't allow humans to ever steer again. Because if you do, all assumptions are out of the window.
for now, that is
Really, if AIs do become self-aware one day, they will be so completely alien to us we might not even recognise it. The way they work is just so insanely different from our minds.
Yeah AI might suck as a driver, but AI cars will allow us to watch titkok for the entire ride, instead of just at red lights and on straighaways and when there isn't too much traffic - ALL the time - and if anything goes wrong, the robotcar is at fault. It's such a perfect promise, that people will overlook any problems that don't interfere with their personal comfort.
I was really hoping someone would make a video like this. As a scientist with some knowledge of the state at which research stands in this field i found Veritasiums video to be nothing but corporate BS.
but i am sure his video about hair and baldness, sponsored by head&shoulders, is totally trustworthy content
XD
He's just uncritical period if you look at his defense comment in the comments section.
Had an off feeling about him a while ago, glad I stopped watching.
🌫️ SERCH ADITYA RATHORE-HE ALSO MAKES INFORMATIVE CONTENT LIKE TOM NICHOLAS
@@LeavingGoose046 I was skeptical of him the since I saw him summon Bill Nye and Neil D.Tyson to act as overseers on an argument about mechanical engineering and fluid dynamics, none of them actually has any real experience in these fields.
@@spinyslasher6586 Wait really? He had access to those two and *that's* what he did?
This is why I have such respect for Tom Scott - who, when a VPN company tried to change the sponsored video more than he was comfortable with, dropped the sponsorship and released the video he wanted and obscured the name of the VPN company. He chose to release an educational video he believed in instead of releasing an ad that'd have made him a decent amount of money.
This is why I only get my information from Toms
@@abderrahimaourir ua-cam.com/video/WVDQEoe6ZWY/v-deo.html
I was just going to comment about the irony of this video about distortion by financial interests containing a VPN sponsorship spot without any critique... Tom Scott's video really turned me off VPN sponsorships and I cringe and skip through every time I see them now xD
+
Fuck yes! That doesn't mean he isn't swiping cash behind the scenes though and feeding you information he gets paid for royally. Distraction by whataboutism is the perfect way to hide things in plain sight. It's used constantly by companies, politicians, etc., even with scandals they create themselves to draw away attention.
Not so much by mr. Scott though. You can actually check his integrity by checking the facts. I'm just saying that, while his actions and words are evidence that support his integrity, if it was the only evidence, I would be very cautious and even more cautious after that video. The reason you notice it as evidence and it _is_ evidence is because it fits the pattern of Tom's integrity. There's a long unbroken pattern. Tom's good people.
looking forward to watching this one! i've generally loved most of veritasium's work, but the episode in question (and a few others) really rubbed me the wrong way
what Veritasium video is this about?
@@benjene according to the video description, this uses that godawful puff piece he did for waymo as a jumping off point, although i hope it gets into the bill gates interview he did which might have actually been worse
I came to making this video from a similar place.
@@freekboy658 I noped out when they did that video which was basically one long ad for shampoo.
Tough to believe the bar goes lower but here we are.
the DNA ancestory video he made is even worse imo.
The ad read promising that a VPN with single-handedly guarantee online security is a nice touch
Hardcore cringe moment that wasn't it. Lost a lot of respect for him right there
the ad didn't bother me that much but i found the costa (coca-cola) product placement interesting. @@Kfdhjgethfdtgh774rvbjs
@@Kfdhjgethfdtgh774rvbjs I don't think it's a reason to lose respect - almost all VPN sponsor spots read the exact same. But I completely agree that they need to add the caveat that "Your traffic is hidden from your ISP, however, the VPN operator can see everything you are doing if they want to".
Wasn't too long ago we found out a major "no logs" VPN provider was actually keeping logs and happily handed them to law enforcement. I wanna say it was Nord, but I can't remember for sure.
Oh and the idea that a VPN alone makes you completely anonymous is laughable - anyone with a shred of knowledge about how web browsing works knows that's total BS.
The irony is honestly kinda sad, imagine making a video criticising another UA-camr of corporate propaganda and doing said propaganda in the same fucking video, worst part is I kinda agree with most of his points.
@@daxie__3210 yup 100%
94% of car accidents are caused by humans is one of the most useless stats ever when you think about it. There are very few other factors that can cause accidents at the moment. It's like saying 99% of boating accidents involve water.
Sober drivers on their way to commit 72% of all lethal car crashes
I know right? People say that only a few lethal accidents have been caused by Autopilot and self-driving cars and I'm like no shit sherlock, self-driving cars have been deployed in the past 5 years and are driven by a super tiny minority of the population. How on earth can you compare those two.
To be fair, the quote is "94% of car accidents are caused by human ERROR", which is to say, a person fucked up, rather than a mechanical failure, a meteor strike, an IED, etc.
This is one of the reasons why Statistics is widely looked down upon in the collective field of Mathematics, given how little grasp of it one requires in order for one to use it to misrepresent, misinterpret, or outright lie about a given data set.
You wouldn't be able to pull off the same nonsense with Topology, Number Theory, or Geometry.
@@BrentODell yeah that's my point, caused and error are pretty much equivalent and that's not many factors which are very unlikely like IED's
Love Veritasium, so this will be a great blow to my current perception of not only that channel, but many "scientifically informative YT channels" (looking at you kurzgesagt, SciShow, and Vsauce). I had forgotten to be critical of the channels I just put in the (I trust you box).
That's a wake up call alright. Often, these science education channels veer too close to a technocratic worldview. Veritassium and Kurzgesagt especially give me that vibe.
I'm glad at least Tom Scott isn't like those other people
@@aprofondir no one is immune to propaganda
I remember kind of dropping Kurzgesagt after a number of their videos on meat/dairy, environmentalism, and health having some dubious rhetoric
I've been subscribed for years, and I just unsubbed from all of them. What a gaggle of fucking shills.
We don't need self driving cars. We need massive, free, public transport.
I mean, massive, free, AI driven public transport doesn't seem like a terrible idea once it's viable. The problem is the capital to build it is being funded by investors who expect to then capture all the value.
Better public transit is only a good thing, however there are a couple of issues with a public transit only system:
1. Great solution for urban areas but less so for towns and rural regions.
2. If anything causes a "hiccup" an entire city can be shut down.
3. Giving the government complete control of anything can be dangerous, obviously
4. Costs money the government probably doesn't have to build and maintain.
5. Nobody wants to walk to the bus station in cold rainy weather
And we need to price in the externalities, that way public transportation becomes feasible everywhere and not just places with high density. And even if it doesn't, it does once you factor in the greater productivity of people who can move around freely and therefore get to meet more people who will become useful contacts in the future.
We probably want both, but separated from each other
@@happycamperds9917 I've been to plenty of islands with a bus you can call. It works great
I love how they think that adding more cars "that are more aware" would stop rush hour from being a thing. You know how you get less cars on the road? You build a subway system, a bus system, a train system. Any kind of collective traffic where one driver can ferry 30-700 people in one go. I take the car, the train, and the subway to get to work every day. The only reason I no longer take the bus instead of the car is because the bus times did not fit me on my way home. A good collective traffic system will make people go "oh I no longer need the car" and they will stop using it. This is the case for most of my coworkers, who live a lot closer to work.
It... Would, actually. CPG Grey did a great video on how synchronized autonomous vehicles would completely eliminate traffic jams, as the majority of slowdown comes from a compounding delay of cars moving after being stopped.
This video is kinda trash for the sheer fact that it ignores legitimate arguments for autonomous cars simply because 'Grrrr.. car bad... Train good... Grrr...'
Don't get me wrong, I want robust public transit systems. I would love to be able to take a high speed train from any major city to any other major city. I would love well maintained subways and bus systems. But to immediately discount the fact that, factually, truthfully, synchronized autonomous vehicles would eliminate the majority of rush hour waits is intellectually dishonest.
@@greed0599Adam something did a video on CGP Grey's video, on why it doesn't. Roads are for people, not for cars.
@@staff4226 Again, that simply does not track with reality and the physics of it. Anti-car advocates jump through hoops to try and disprove the idea of synchronized self-driving cars removing traffic jams solely because, in their heads, there can never be a world in which cars can be improved to solve any of the problems they have.
@@greed0599 I don't disagree that cars can be improved. It's just that if you tape a bunch of cars together with only one big engine pulling them, and place them at low friction rails, you suddenly get much better improvements.
@@staff4226 But that's not the argument for or against automated cars. I am, in fact, in favor of trains and public transit. I don't drive, I take the bus and tram everywhere. But you're never going to 100% eliminate automobiles, so why not improve them all and acknowledge that improvement can be made. Nothing is ever going to be 100% bad or good.
The legendary Pulitzer prize-winning journalist, Chris Hedges, recently made a video on this topic. The corporatization of Science in the US. You will find things like Nutella-funded "scientific" studies that claim that sweets greatly improve your health.🤔
Hah imagine that being a recent thing. Remember that controversial study funded by sugar related companies to make fat look worse than sugar? It’s been happening for a while. Always look at authorship for journals before you make conclusions and especially look into a journal if it’s being mentioned a lot.
Philosophy Tube also talks about this in the video on "Ignorance". Doubt is our product.
This is hardly new. But yes it´s getting worse. We need more public research.
Jesus, what a dreadful thought. Science classes brought to you by Nestle. Even worse is that corporations funding faulty research is something that is decades, maybe even a century old if we talk about pharma or tobacco industry.
@@XMysticHerox The problem isn't a lack of Publicly funded research, the problem is that biased corporate-funded research is used to muddy the waters and is then placed on the same pedestal as impartial publically funded research in the media and public discourse.
The irony is that taking the time to explain in detail the additional measures of 3D mapping of the road, real-time monitoring by support staff and other human involvement, Waymo actually WOULD make themselves look good; maybe slightly less cool and futuristic, but way more safe.
'Not quite as cool and futuristic, but definitely Waymo safe'
I'd love that slogan, though I know it would never happen and would almost certainly be counterproductive for the company because safety, as foundational is it is in reality, doesn't appeal to human emotions like "cool" and "futuristic" do, and emotions (unfortunately) motivate people to buy things far more than careful conscious consideration does. Most people really do think that the majority of their past and present purchases are made due primarily to logical decisions, but psychological and sociological data prove that that is very, very often not actually the case as much as people want to believe.
@@erichanson3369 Maybe, maybe not. I think the product has lost a lot of credibility due to this and so has Derek. Just one of those things...
Showing that aspect makes people start to think about safety and potentially questioning if it's actually tight and not with flaws. Like if they say "it works by X Y" the viewer might come to ask "but what if Z?"
Another great video, Tom.
When watching Veritasium, Vsauce, Kurzgesagt and all the such, I usually turn off my critical eye. It's like watching a 3blue1brown or numberphile video. I just lay back and absorb whatever scientific information is given to me, trusting that it was well-researched and unbiased.
This makes them the amazing vehicles for corporations to advertise. When you're sold something without knowing you're sold something.
It's a little sad that I now have to be sceptical of the UA-camrs I've learned to trust for the past few years.
Good work as always. Keep it up!
When you hear about the social behaviour of ant colonies you can have that sort of trust. But if you look, you will always see people calling them out for mistakes even in that sort of content. But those are harmless, to an extent. Minutiae about hard science facts do not have a solid impact on everyday life as, say, discussions about GMOs or organic foods or information security which these channels are equally likely to approach and which may provide them with even more engagement that "drier" pure science topics
Look up Folding Ideas video about VSauce.
The idea that the channel about pictures that look like dicks and boobs reinvented its image enough that people took it for reputable is hilarious
even sabine hossenfelder was just recently sent onto the slippery slope of dis-infotainment. getting her video on quackery sponsored by brian keating, then promoting his channel. he is a grifter who invites far right propagandists and participates in the pseudoscientific pragerU (fake-) university scam, funded by rightwing fracking billionair cash with the goal to further the republican agenda of privatizing everything, cut social spending, increase giant corporations federal subsedies
Be careful. 3Blue1Brown is a shill for the big math corperations!
@@Camerooony11 He can shill all he wants with those cute dreamy face. Same with Steve Mould
This reminds me a lot of the early days of TV when you could see your favorite cartoon characters whip out their favorite brand of cigarette and advertise to you directly in the content. Here it is even more insidious, in my opinion, because it is content that purports to be informational/educational. There were regulations that drew bounds between ad and content then, why not now?
I grew up in an era of paid late-night programming. Veritasium's Waymo video reminded me of how I used to fall asleep at 2:00 am.
I want to jump on your question on why there 'aren't regulations separating advertising from other content by stating that while cable television, it's infrastructure are a lot more centralized and 'public' in nature, therefore easier to regulate, the internet is by contrast much more amorphous, and UA-cam is a private company with a scary monopoly on video sharing. Therefore, 1. UA-cam, being a monopolistic social media of sorts, actually benefits from both ubiquitous advertising and not having the same inherent accountability as governments and centralized infrastructure, and 2. Consequentially, internet activity of this nature is (in my view, for better and worse) barely regulatable, at least not without employing drastic measures that may do more harm than good.
@@willthealmightygamer6368 The government doesn't enforce existing regulations. People need to stop blindly trusting producers. We are WAY better off than we were 30 years ago. You have any idea how much work it used to take to find out the government had DEFINITELY liked to the public in 1990? It took years, it takes days, sometimes only hours now.
We've always had garbage information, it's far easier to identify it now.
On this platform - "This is a sponsored video" is spoken at some point, and the words "paid promotion" briefly will appear onscreen. That's it. It's self-regulatory, but individuals can report violators. In the USA, the regulators are in a revolving door relationship with industry, so the distinction between educational material and a sales pitch just isn't in their world view.
Because, to put it a little too simply, corporations can lobby politicians.
The point about "educational" and "news" youtube channels focusing on cars as the default mode of fixing issue is something that has bothered me for some time. "Hmm, cars are producing too many emissions, making them electric will fix our issues." and "People die because humans are bad drivers, lets make computers do it." have been repeated throughout major online media channels, when busses and trains get rid of both of those issues, as well as have other benefits such as encouraging more sustainable and equitable urban design.
Adam something whenever he talks about some stupid techbro vehicle-project
Because Americans have been brought up on the idea that owning a car is the embodiment of American individualism, which is a load of sh!te ofc, greater investment into public transport instead of reforming cars is the way to go, but that can't really happen when the automotive industry lobbies the government to stamp out any sort of transport reform that would actually benefit people, it's dystopian, it's gone so far that cities are designed in such a way that owning a car is a necessity, it's a vicious cycle and many people have been brainwashed to such a degree that they're even proud of the fact that they live in a country with barely functioning public transportation,
Without a core understanding of civic design, it's really easy to fall into the trap of "not thinking about it", or thinking that the american way of life is default, that there's nothing wrong with it, which just leads to car-as-default. To be frank, it's very easy to miss civic design as even being a factor here, because there's really very little published on the subject in youtube or in terms of academia relative to other shit. I've seen like 4 civic design channels above 100k, that's about all I've seen to be honest. The only one I've really seen tackle the problem of car-dependent infrastructure in a way that was coherent was "not just bikes".
@@emilycampbell6375 Absolutely. To go a step further, Americans have been taught to think about things only in terms of individualism and their inability to imagine any non-car solutions to public transportation is a symptom of that.
Like, for Americans, the solution to global warming is to buy eco-friendly products. The solution to racism is to be nicer. The solution to poverty is to work harder. etc. etc.
@@guy-sl3kr so severe that any form of collectivism are often labeled as "socialism", as if they don't know about Japan
Viewers have parasocial relationships with their favorite youtubers. Sponsors know that and want to exploit it. Sponsored content is deeply exploitive. It's greed over integrity when creators cash those cheques.
This must be an Adam Something video in disguise because the answer was "Just build a fuxking train" all along.
Thank you for the well thought out, fantastically structured video essay. Very big shout out to the subtitling. It's a small thing for most people, but for me it makes or breaks a good video essay. Thank you so much for including them, and doing them well, too!
That is an answer, but it isn't a solution. I personally have never heard of trains randomly popping out of the ground, ready to be used. If it is more efficient to first have self driving cars to suit areas only covered by train in distant future, so be it.
The difference between answer and solution, is that a solution is actually specific, and not just a general statement. In general Adam Something's videos remind me a lot of propaganda, just that I think that some of the content is actually true.
@@jan-lukas I mean, in London, the stations are so close together and the trains come so regularly that they might as well pop out of the ground ready for you. If you wait for more than 6 minutes for a train in London, it's either a bank holiday or something has gone horribly wrong.
I was thinking the same way. Or an RMTransit video too.
@@jan-lukas there's always a smaller train.
Metros, trams.
And u can walk a last kilometre.
I have been living without a vehical for 5-6 years now
I do have 2 bikes but they are rarely needed. Bcz I live in India and there's train for everything here where I live there are metros trains suburb trains. Etc. And metro at my place also provide rent bikes. For going to college etc.
@@jan-lukas there's something called "bus", which doesn't require rails immediately.
It rings a bit hollow when you push your own sponsor too
What's really irking me here is the lack of data, you brought up a smattering of cherrypicked anecdotes and sensationalised news headlines, as your counter evidence against driverless cars. Sure if you look hard enough, any technology is going to have failures somewhere. But the absolutely critical thing to get across to the public is that they are still safER. Because human drivers really aren't safe. A human is a squishy, error prone piece of hardware that suffers all kind of health issues, judgement problems, opinions, tiredness and intoxication, that have turned our roads into a constant churn of death for the past 70 years.
Car accidents are by far the most common non-disease cause of death, they ydeserve far more intense calls for actions than they currently get, and certainly more than gun crime gets. The fact that the tech isn't perfect, so we should wait until it is, is an endlessly moving goalpost that ignores the realities of thousands of people dying worldwide because of cars every day
And your claim that since derek's video is propaganda, that you don't have to "engage with the minutiae" of his arguments, is an intellectually dishonest handwave.
So far i'm 8 minutes in and you have repeatedly used opinions and anecdotes in an attempt to fight data. You've not demonstrated that you have any more integrity than those you disparage, and i'm struggling to see a reason to continue
To be 100% fair, VPN sponsors oftentimes also make you say talking points that are either just barely true or just straight up a lie in those sponsored segments. But those are never a part of fully integrated videos so it's different.
It is different, but still not right. Because the main content of the video was not influenced, it is different. But that does not make the misinformation justified. (This should be applied to all UA-cam sponsorships. I don’t know if Tom’s sponsored segment had misinformation.)
The VPN ad just hammers home the point that if the information is sponsored in any capacity, even if it's in the same video that's debunking another video for completely misrepresenting reality, it will be complete and utter bullshit. Be it Gandhi, god himself, an actual saint, they will lie and/or knowingly misrepresent facts through their teeth.
@@TheLolilol321 I sincerely doubt that all sponsorships corrupt content completely. What if the sponsorship was just about getting the recipient/UA-cam to engage with a certain topic on their own terms (and such sponsorship is disclosed.) The main issue stems from the direct control that a UA-camr can give to the sponsor, best exemplified by SnazzyLabs’ comment on this video. That being said, a bias can still form when a UA-camr is given money to simply research a certain product, but if the the tuber provided all the sources to the audience and encourage them to engage in due diligence themselves before making a final decision on the product, then most bias should be eradicated.
Honestly, things would be different if Veritassium provided the sources of the information directly in the video/description. And although it is a different scenario, things would also be different if Tom said directly in the ad that he got his information about the vpn , but and vpns in general, directly from the vpn company.
@@TheLolilol321 And what if the sponsorship is just “Hey, this thing exists. Look into it, maybe.” Like the many icons on nascar cars and such. At the very least, there’s no misinformation in those instances. Other issues, certainly, but usually far less impact on the audience’s decision.
@@xzylo451 An exaggeration to be sure, yeah, citing sources in the video right before the claims would indeed be a nice solid step to take.
Corporations get their mouthpiece and the UA-camr can indicate that they do not necessarily personally vouch for the information.
I did say sponsored "information" though, which a logo with text is not.
"You convinced me, drivers are unreliable and traffic is a dangerous problem beacause of the chaos it produces"
"So you gonna support our self driving car start up?"
"I'm gonna fund public railways"
"wait no"
Public railways, bicycle paths, and more walkable cities. Tada, far less car traffic and related injuries.
Edit: Less air and noise pollution as well.
@@Cybermaul Also superimportant: saving a LOOOOT of energy. People thinks of Hydrogen/Electric cars as "green" but even if we can solve all the problems related to the production/efficiency, they still consume lot of energy (generating and encasing hydrogen is a complete mess and it will never EVER be for free). Cars are just that inefficient to transport people compared to train, buses, bycicles and..feet, and there is nothing we can do about it unless we move to a universe where thermodynamics do not apply, or discover magic.
This comment needs more attention 😂
@@blackest3314 Nobody forces you to convert your electricity to hydrogen and back. The most popular non-ICE cars on the road don't use hydrogen.
No thanks I don't like walking miles to wait for a half hour for a dirty bus I have to wear a mask on, which then moves slow as fuck and drops me off a 30 minute walk away from the grocery store, then have to repeat this process to go home when I could have driven there and back by the time i even got there with public transit
I completely forgot how angry Bill Gate's book makes me. It's disgusting that he tried to sell a book of extremely basic climate change information, preaching "personal responsibility" while he has the money, power and influence to directly make the massive industry changes necessary to actually solve the climate crisis but never will.
In his book he does actually say that we do need governmental solutions to the problems, but doesn't really talk about all the subsidies fossil fuels already get,
He just talks about subsidies for consumers to close the gap between green energy and fossil fuels and carbon taxes and stuff like that.
He's a billionaire so of course he's not really advocating for massive change to a system that worked out for him.
@@arsenalfanatic09 and sidesteps the latter part of OP’s comment by saying “well, hey, capitalism makes self sustaining enterprises, so it’s way better for the environment if I invest in companies to do the changes for me (while I make dividends from them all, whether they actually help the climate or not)”.
🟪 SERCH ADITYA RATHORE-HE ALSO MAKES INFORMATIVE CONTENT LIKE TOM NICHOLAS
Bill Gates does not have even 0.0001% of the money, power, and influence needed to solve climate change.
Preach brother" 🙌
this video is a wet dream. I already disliked veritasium so this is just a cherry on top
Veritasium's content is great in general. but this is a friendly reminder that we should be always thinking for ourselves and be critical of the things we hear from anyone, no matter their qualifications.
I was shocked recently by one of their oh-so-sensational physics videos about speed of light. It cherry-picked outdated methods dating back to Galilei whilst it argued with 1900s papers by Einstein - at the same time, left out methods, empirical evidence, and entire areas of physics that not only came decades after, but also do provide evidence on numerous "ooooooh soooo mysterious" claims in the video.
Just one example: speed of light and its (potential) directional dependency... it refers to antiquated methodology, a nowadays unnecessary and frankly ludicrously misleading thought experiment flawed on numerous levels, AND leaves out the Michelson-Morley thought experiment.
Not to mention how it leaves out the arrival of Maxwell, atomic clocks, and shockingly precise interferometry. And all the evidence that demolishes the numerous "ooooohhhhh mystery...look at that" nonsense.
Seeing something like this I admit: I at least have to treat the entire channel as highly suspicious in its purpose & agenda, because 1. if presenter has erudite background, then cannot make such videos, it would be impossible without a strange agenda, 2. if presenter really understands only what is in the video then must never have made a video about this topic.
Not sure which is worse, but to me, Veritasium seems to be an at-best sensationalist "education" channel, at worst tries to raise doubts about known areas of science with some non-scientific (take your pick) agenda.
@david Sh. AGREED!!
his content is great? it´s largely a clickbait channel full of half truths and inaccuracies for scientifically illitarate people that want to feel like they learned something while beeing misinformed. it´s just sensationalism, "education" is just its marketing theme while at its core it´s mostly just an advertising vehicle for his many, many, many sponsors.
i mean the guy is a multimillionaire a hundred times over, you don´t get there by plaing fair. and it´s not like he´s using his ill gotten gains to give out grants, like real educators like sean carol for example with less than 1% of what he has do.
in his pinned reply to tom, he´s seriously citing an "academic study" that´s been bought and paid for by waymo. he knows damn well, that this piece of blatant propaganda isn´t worth the paper it´s printed on. yet he still has the gall to present that as an argument. i´m looking forward to his video advertising the health benefits of smoking, since there were plenty of "acadamic studies" bought and paid for by phillip morris and co. as well that showed just that.
i´m stressing this because this isn´t an oversight, that´s not a tiny oversimplification or a little clickbait sensationalism, this is straight up lying to your face. and not only that he´s misusing "science" as a tool to fluff up his lies when calling it an academic study.
for me, someone capable of something like that has lost every shred of credibility.
i mean i only personally object to your first sentence. i hardly couldn´t agree more with the second.
but if you apply that critical eye to veratsium, theres hardly anything left that´s of any substance at all.
to the extent that it's true, it´s trivial and to the extent that it´s not trivial, it isn´t true.
so in short, you´re wasting your time watching veratasium, when there are so much better creators out there
@@LeventeZone thank you 👍
@@symmetrie_bruch Exactly. I can only use "great" as sarcasm. I was frankly astonished of the content on this channel. Either someone deliberately ignores decades of physics (and/or cherry picks things), or truly doesn't know. In both cases, should not establish a channel with "veritas" in its name. Rule of thumb, if such amazing names are given to something, usually it is overcompensating for other things... and best to avoid.
Hearing that clip at 44:45 where Muller talks about how people can be happier because commuting and being stuck in traffic sucks had me nearly shouting "TRAINS!" at my screen.
You can't just put a train in the middle of a city. That's not a feasible solution for the hundreds of cities in the United States of America alone.
@@electrochipvoidsoul1219 Lol, have you seen american roads? If you have more than 3 lanes in a street, you can just replace the two central ones with trains. Well, they would be trams, but the point still stands.
You could put a train on every single street that has more than 2 lanes.
And some cities are just doing exactly that.
And let's say some streets can't do that for whatever reason, putting a bus on that road is just as easy.
@@electrochipvoidsoul1219 they did that with roads 70 years ago
@@electrochipvoidsoul1219 You can. You can also put a bus, metro and tram in the middle of a city. In fact, the only places you can't put them is outside of cities in small villages in the middle of nowhere. And that is mostly because in the middle of nowhere in the US often means in the middle of nowhere instead of a 10 minute drive from the next village over.
Ah, a fellow AdamSomething viewer?
I love both channels but I love this youtube "peer review" even more. Initially it was a hard pill to swallow - especially the Johnny Harris video, but I'm glad to have come across this because it makes me a more responsible and informed consumer of youtube content.
and then youtube removes dislike counts ...
@@BlueEdgeTechno "Return UA-cam Dislikes" extension while it still works.
More ‘education’ channels should do stuff like this. Every other form of education, science and engineering for example face scrutiny and review so why not UA-cam content as well.
If you are more responsible please don't listen to the video on this channel. The bold claim that self driving tecnhology is just "a pipe dream" shows Tom nows nothing about the topic and his just trying really hard to find the clickbait views. Which is ironically the exact same thing he is criticising veratassium of doing. He is spreading missinformation for the money. In this case the youtube views of a clickbait title.
@@paulogaspar8295 Nobody is arguing that self driving tech isn't something we should be pursuing. The main issue is that self driving tech doesn't work well with human drivers involved, or at least not well enough to squash the perception it's unsafe.
We need to massively redesign transport for self driving cars to work. Unfortunately that won't be happening anytime soon so self driving tech will most likely be limited to interstate transport where conditions are much more predictable.
And lets be honest here, the main issue with the US transport system is that it's designed with one goal in mind: cars. That needs to change before all these technologies can be fully utilized. Until the US begins to change how cities are designed, zoned, etc, none of these problems are going away. Everything is patchwork until the US drops the mindset of cars first.
Clearly, the "94% human error" doesn't say anything about the safety of AI drivers since they haven't even been pooled into the sample. To prove anything about safety either way, there needs to be a direct comparison between humans and AI, and it needs to be expressed in some sort of a ratioed measurement, such as "number of accidents per 100000 miles" or the like.
I agree that a proper comparison needs to be made. AI shouldn't be compared to purely hypothetical 'perfect' driving. Also, when I look at some car crash compilations here on yt, it seems like 99.9999% of accidents are due to human error. Even for example in heavy fog or on icy roads, it is still human error to drive too fast in such conditions. (imo)
Is this sarcasm? NHTSA has been putting out these statistic for well over 25 years, and AI driver reports from various companies (and the government agencies keeping tabs on them) have also been publicly reported with "number of accidents per x miles" since before the pandemic. The standard number of miles is 100 million vehicle miles traveled, so 100 thousand is almost too on the nose to be a random guess on your part. They even have break downs to causes of accidents as reported/discovered by the police reports to help narrow down what or who is at fault. These are USA specific, unfortunately, but that was the target of the comparison to begin with, and will hopefully be updated with other countries as autonomy is rolled out, if possible. That being said, not all governments track these sorts of things.
@@PoikSpirit Ah, but the AI cars are generally only being driven on good roads in good weather.
@@donaldhobson8873 Yep, fair point. Granted, that's just for now. As the technology matures, they will have their use cases expanded too. I believe those details are also in crash logs (reported in public records). I'll have to double check. Still most human accidents happen in easily avoided, but overwhelming situations. Like crowded parking lots. Here at least, the usual bot shines (at avoiding crashes). Keep in mind, I am not claiming they are ready for the prime time. Anyone who says that is trying to sell you something (and in the case of Tesla, they may be lying by omission quite heavily).
@@donaldhobson8873 SF driving is atrocious! I used to live there and it was always stressful in the city. NY and LA were much worse though.
What a masterpiece. Honestly in my opinion this is one of the most valuable videos you've made. The criticism of the specific Veritasium video is great but especially pointing out the implications of sponsored educational content in general is something that deserved way more attention than it got so far, and you've made an incredible video to give it the attention it needs.
That's very kind. It's always difficult to avoid these feeling a bit like "hit pieces" in some regards (and, yes, I add a bit of theatrics to make it enjoyable), however my focus is always on the broader trends that these examples throw up and I hope that comes across in the video.
i agree
🟢 SERCH ADITYA RATHORE-HE ALSO MAKES INFORMATIVE CONTENT LIKE TOM NICHOLAS
@@Tom_Nicholas To be quite honest, though I haven't finished watching your video, I always end up thinking "if there's one type of content creator people should be especially demanding and harsh with, it's educators". How could someone have the pretense of educating people while also doing sponsored videos and generally responding poorly to good faith criticism? Like, if nazis like PewDiePie did that (and he might have done that, idk) I wouldn't bat an eyelid because I know better than to ascribe educational value (or indeed, any value) to his videos, but the Veritasiums and Kurzgesagts of the world should know better and be more humble when caught with their pants down, especially if they're already doing very well for themselves financially. This is a long-winded way of saying that I support what you do. Keep up the great work!
@The Program if he'd made those two separate videos, both videos would be weaker because the two points reinforce each other. Veritasium is wrong about self-driving cars in a way that specifically benefits Waygo, and it's disingenuous to try and divorce the nature of that wrongness from the corporate relationship between his channel and Waygo
I've seen a fuller version of the "space-saving" argument, it isn't a complete non-starter, it just assumes that with the added safety of driverless cars, we would be able to have cars travelling much closer together, at much higher speeds, thereby having less road space taken up by the same "human transport bandwidth" of cars. This is completely correct, and the configuration of many cars tailgating close together at high speeds has been tested, and named: it's called a train.
I was going to add this as well. Congestion is typically not caused by car volume but by car spacing adjustments made (relatively) slowly by human drivers, as well as lane changes propagating slow downs.
I think the best counterargument is the one you and Tom made though: public transport solves this and many other problems, driverless or no.
Even then, Induced demand means that that will simply make more people take that road
Every time some tech bro comes up with a "solution" to traffic congestion it turns out to be just public transport but worse.
@@RaunienTheFirst oh god I remember when I was an automated car Stan and my great idea was to have the cars link up together as they go down the road to take advantage of reduced drag and increased travel density and now I realize that that is called a train
@@land_and_air1250 Yeah, it's a funny thing. I used to be a stan for it, too, and I even went into computer engineering and AI to study things like this, but imo the tech developed for self-driving cars is best used for anything but cars; just make trains and bikelanes, for goodness sake, and put the AI to good use on autonomous Mars rovers and stuff.
This is gonna be a spicy one
Very spicy title. Grab the popcorn. You know there's going to be a back and forth.
Yes
Gonna say this video has aged like fine wine. The @veritasium response to it has aged like milk. Especially in light of the massive scaling back of autonomous driving programs across the industry, the revelation of cooked numbers by Tesla and Waymo, and the drastic reduction of investment in these projects. Thanks to Tom for the video, and I await Veritasium's meo culpa.
Funnily enough, Veritasium's driverless cars video is what lead me to the Not Just Bikes and Adam Something videos about urbanism and car-centric city planning, which in the end made me realise how bullshit that Veritasium video was
veritasium is a dumbass
I came to them from cgp grays bullshit video about autonomous vehicles
As nice as it would be, you can dream and plan for your future, while still living and acknowledging where you're at now.
AI cars would solve a vast majority of our societies issues in the transportation sector in an affordable manner.
@@RebelTvShka More affordable than a literal tram?
Edit: I mean I suppose if you're going by the least additional infrastructure required, but like. Mass rollout of autonomous cars is certainly not more affordable than basic public transit.
@@RebelTvShka oh heck no dude what
There's a really nice quote from nuclear historian Joseph Morone which I feel is applicable here:
_This is not a story of technology run amok - although that is how many people understand it to be. The history of nuclear power is a history of political and economic and social decisions being made about a technology, and the key decisions weren’t being made by the technologists. They were done in the business room._
_What science and technology gives you is a range of possibilities. And those possibilities can take you in any number of directions. It’s potentially a liberating force. But to get there, society has to stop sleepwalking, and realise it’s not a scientific choice. It’s not an engineering choice. It’s a moral choice._
I think my frustration with the Veritasium video comes from a wider frustration in the belief that *this* form of the technology is the inevitable form it must take. And if that was the form it must be the right form. And to me the blind faith in science is a very wrong headed view of science. The existence of a technology does not automatically guarantee utopia. Without holistic planning and policy, negative unintended consequences are likely, and Science Tube routinely forgetting that will always be a source of frustration to me.
Autonomous vehicles will not automatically drive us to a better transportation future. We have to take the wheel ourselves.
Great quote! Thanks!
You're right. We can dream of a future with all-around self-driving vehicles, but the truth is: We don't know if that's the future for us. Just stating, that it _will_ be the future and now we need to get there asap may leave other options and ideas underdeveloped.
PS: I love your videos, and I've watched all 4 of them. :D
You nailed it with that closing paragraph
I think what often happens is that technology becomes a stand in for a company or institution that is using said technology and that is intentional. People basically always think of technology as an automatically good thing and with good reason seeing how much our lives have improved due to it and that is why companies and the army love to associate themselves with technology and having propaganda to make it seem like they are that technology. It's a great way to pre-emptively derail the conversation, if I criticize Tesla I'll inevitably be hit with the accusation that I hate electric vehicles as if Tesla somehow is the concept of electric vehicles and we never get to discussing the problems with the company Tesla.
yeah mate, absolutely. i think we're coming into an era now where its too easy to see a technology as the future & talk down to people who truely think through if something is right or wrong in nuanced ways. like blockchain.
Having a degree in Artificial Intelligence and working in the AI field for almost a decade now, the Veritasium video really irked me when I watched it for its completely uncritical portrayal of Waymo and driverless cars. Him doubling down makes it even worse.
Content creators should be extremely prudent when doing a sponsor integration like this, because as we say in Dutch: "trust comes on foot, and goes by horse."
Well, now it "goes by driverless car". Which mostly seems to mean "in a circle".
I'm not sure I understand the saying???
@@HxH2011DRA It takes a long time to build trust, and it's lost very quickly.
@@hititwithit Jooost
@@hititwithit Joooost
Driverless cars is the biggest scam since FTX
Seeing all the top UA-camrs commenting on this video makes me feel this is some high level decoding shit where everyone is getting their opinions out.
I'm regularly rechecking the comments to wait for others UA-camrs comments, like Physics Girl
It is
I feel like an outsider wandering into the drama having no clue what's going on 😂
i read your comment as "highly decorated" which also works!
Tom Nicholas is one of the best youtubers when it comes down to tackling big issues like these. He has criticized a few massive ecational youtubers in the past.
“Or whether better options already exist.” … So says Adam Something, “TRAAAAAAAAAAAAAAAAAAAAINS”.
i was looking for this exact comment
Eh trains are not a solution tho. They need a lot of people to use them to be feasible and they aren't exactly flexible
always a pleasure to find another Adam Something spectator in the comments
@@victoriap1561 It can definitely be flexible. Train isn’t just limited to heavy duty subway or passenger train it could be a light rail or even a tram. If there are encompassing system in place, utilizing a network of various public transport can be just as flexible as private vehicle. You just never get a chance to see a world outside of this dystopian reality where automotive industry has total control in shaping our infrastructure
@@victoriap1561 If we match the amount of cars, and length of rail to the demand of the city it's being built in, then there wont be an issue of having "enough people". Trains arent just built 1 size. They're flexible because they can be built to suit the city's needs and population density. Cars can be added or taken off. Multiple engine cars can run at once during peak hours. It employs locals, allows cities to become more walkable, reduces emissions/improves air quality, and... so much more.
You say trains need a lot of people to run efficiently, ok... There are a lot of people in cities. lmao. That's why they work better than sitting in a traffic jam.
I know new futures are hard to imagine, so maybe check out "Not Just Bikes" youtube channel if you're interested in learning more.
I've been feeling iffy about the channel ever since he did an episode on DNA and how it caught the Golden State Killer. The video seemed to gloss over the negatives of mass DNA surveillance and not explore them very fully at all, focusing 95% of the video on a company whose business is the use of said DNA. Not exactly an unbiased or balanced look. It just really rubbed me the wrong way.
Same hat! That video also made me extremely wary. This video basically took that concern and cranked it to 11.
I already had an off feeling about him years ago. I knew something was up with him.
worse yet, in my opinion, he alluded to it many times and liked comments that mentioned it. However he never actually addresses that huge issue and really get into the surveillance problem
Maybe cos the guy's got money and doesn't really give a shit which yeah it's concerning
Absolutely this, that was a disturbingly unbalanced video with very little introspection in the slightest.
While I think doing a whole sponsored video is off-brand for someone claiming to be a source of true information, this expose is actual garbage. Just a few points: As for the flight, I can't speak as to what Veritasium's point was about what it showed, it definitely did not show the danger of automation, because the landing was not fully automated, and only automation-aided, and the crash happened because of the human's wrong choice. As for the 94% study, you claim their methodology includes any cases where any blame whatsoever can be laid on the driver, even in an unreasonable way as it would be in your example. Yet you show zero substatiation for that claim about the methodology of that study. Then, of course, you claim that Veritasium tells us knowing about the lidar and cameras is enough to have a complete understanding of how the system works, and, again, show absolutely no evidence of that. Seems to me like you are trying to put ridiculous words in his mouth.
How long till Tom pulls an Adam Something and just says "build a train."
"But communism" or something - probably
I love that guy
I went down to comments for this one, thanks a lot
BUILD A TRAIN. Many of them.
BUILD A TRAIN. Many of them.
Veritasium/Waymo deliberately leaving out human created maps and other human labor is a great example of Potemkin AI-where a system is presented as "real" AI when in reality regular human workers operate critical parts of the system. When a company claims to have an AI your first question should be "who's behind the curtain pulling the levers?"
i don't think potemkin ai is a real term
@@heartache5742 they probably mean mechanical turk
Like the "self-sustaining" train in one of Chris Evans movies. Train creator/designer/owner claims the train is self-sustaining, but later, it turns out otherwise. Not gonna spoil it for other ppl.
@@heartache5742 The term comes from an 2018 article of the same name by Jathan Sadowski (it's a great read). It's inspired by the Mechanical Turk.
@@heartache5742 Any term is a real term as soon as people start using it. And as described, it's a fine reapplication of the longstanding term Potemkin.
I really like that this video seems to be careful to not just be "automation bad" or something like that, especially given that automated systems *can* improve safety a lot in a lot of contexts (just not every possible one).
Also trains are good, actually. And more easily automated, too, I guess ^^
Trains have always been good, anyone who says otherwise is a shill or an idiot.
@@SECONDQUEST You enjoy your Anti-American trains, I'll be watching you from the highway while enjoying my freedom to be stuck in traffic for 6 hours. 😎
I LIKE TRAINS 🚞 🚆
@@SECONDQUEST
What about their effect on the enivironment?
train good; car bad.
Having read Veritasium's response, I am beyond disappointed in them. Its pretty obvious to me, that no matter how good or honest a video is, if the subject of the video is about a sponsor it is unethical to try and pass it off as anything other than an advertisement. Any attempt to say "well the video was actually unbiased" misses the point entirely. If they truly believe that driverless cars are the future and that we need to support the technology and their video needed to made exactly the way it was, they then should not have undercut their argument by taking the money.
Tom, this is a great video. What irks me though, as a programming student, is the fact that you critisize veritasium (rightfully so), but fail to see that VPN ads are very misleading too. I am not saying that the two are equally bad, I agree that the Wayme placement was worse, but VPN ads are also very disingeniuous. Tom Scott has made a video about it (ua-cam.com/video/WVDQEoe6ZWY/v-deo.html) which explains most problems. Besides those, claiming that a VPN protects agains big companies snooping on you is completely false. Big companies nowadays gather data through fingerprinting, account logins and cookies, all of which work regardless of whether you are using a VPN or not. There are ways to stop these things from happening and they are free (like using privacy plugins and using less of the big companies services). This makes me even more against these ads, since they infir that you need to pay for a VPN to be safe, while this is not true at all (it doesn't make you any safer in most circumstances).
I commented this elsewhere, but I think it's 2 different beasts. Is there some level of hypocrisy there? maybe, but it's not like tom nicholas did an entire 10 minute video praising how much safer VPNs are while saying "btw surfshark good, here's an interview with surfshark head of operations, btw im using surfshark now.". It's a very delimited section of the video that is not integrated with the content in any way, and it is the integrated ad that is truly dangerous. VPN ads irks me too though, and they never focus on the only real reason people would actually use them (region-locked content).
Veritasium is in the science and tech sector of youtube, he undoubtedly *knows* he is misleading people willingly. He is trusted to speak about science and tech specifically, so making a full length advert on science and tech that subtly pretends to be an unbiased video is misleading on another level from Tom Nicholas's sponsorship.
Tom Nicholas's sponsorship is on something he isn't expected to know about, and although there can be a level of exploitation of this fact, if this fact wasn't respected at all, most sponsorships just wouldn't happen on youtube. Most VPN companies lie about the utility of their services and you can recognize this because you are a programming student, but you might not recognize if another commonly sponsored product such as raycons were bad, so if you were a creator, you would need to do fairly detailed research on the sponsor, which isn't something you can expect all creators to be capable of doing correctly. Just now after writing this have I decided to watch the video you linked, at Tom Scott actually respects the point made in this paragraph.
Another point is that when the sponsor and sponsoree are closely linked together, the marketing is much more effective, thus much more valuable. Not only is Nicholas's sponsorship less valuable because it is temporally shorter than veritasiums, but it is also less valuable because it is less effective because nicholas's sponsors aren't chosen to be in line with his content. This raises a further point about future content. If veritasium gets a deal with a company who has the same products/services/interests and vertasiums content, veritasium will have to avoid conflict with the sponsor not only in the video in which the sponsorship is in, but any future videos if he is wishing to get similar sponsorships in the future. He is likely to want to get similar sponsorships in the future for the same reason he got them in the past. On the other hand, Nicholas gets his sponsors from various backgrounds, meaning they are replaceable and that he doesn't need to worry about a conflict of interest in a future video. This means that the non-sponsored parts of Tom Nicholas's videos can always be trusted and aren't at risk of bias because of sponsorships, but veritasium's UNSPONSORED videos can no longer be fully trusted, thus every second of what vertasium is currently outputting in terms of content is somewhat tainted.
@@StupudPerson He isn't 'expected to know' about VPN, the thing he is taking money to push? I mean, if you think that is 'honest' actor who 'couldn't know' he was misleading people or 'VPN is so advanced you can't expect' (any) research to be done. And the point about how VPN is less effective ad, yeah, because they don't care who they pay to put their ad WITHIN a video.. you telling me this guy so dumb he doesn't know how VPN ads work? And can't do research but can make 51 min that I should then believe anything within? ALRIGHT? No.
I really like how Linus Tech Tips does VPN ads on the WAN show. "Want to pirate terabytes of content without getting those annoying cease and desist letters? Try Private Internet Access!" Most of the good uses for a VPN are of somewhat questionable legality and ethicality, so they're kinda hard to advertise honestly, but they are great for gay pirate assassins.
Look, I work in tech and hate VPN ads but these aren't really the same thing. One is a sponsored ad break -- the other is literal editorial control over content being produced.
Honestly tho, I am 99.9% sure you and I both know why people are actually getting VPNs. Very few are getting them for privacy. Most are getting them to get around content restrictions and pirate shit because they don't want a seedbox. But you can't explicitly say "this shit is great for piracy"
I spent about half of this video yelling "public transport" at the Veritasium man, so of course I appreciated the conclusion. Wonderful work as always, Tom.
This is like when Elon Musk proposed this new form of transport that is revolutionary, it will fit in multiple people, travel underground at high speeds.. and everyone was like wow what a visionary.. more enlightened people were just “well that is just a shitty version of the subway”.. people need to get a grip of reality
TRAINS ARE THE BEST
Even long busses are just trains on wheels. Even better, trolleybuses are busses that are entirely electrical, don’t carry their own fuel and are crazy efficient... hmmm, those soviets were onto something.
@@cezarcatalin1406 If drivers now stopped staring at their phones while driving, everything would be perfect. (Sorry, just saw a tram driver do that a couple of days ago.)
Public transport works well in a metropolis. It does not work in most other areas in the same way. Things are too spread out. And in environments where the climate can make walking to distant bus or train stops dangerous, it is even worse. So a combination of public transport and self driving cars works best. For example, if I live in small city and work a late shift, it would be hard to use public transportation to go to the usual places people go in the daytime since public transport shuts down after dark in small cities due to not enough people to justify running it.
@@NandR Public transport can work wonderfully in for example Swiss versions of rural areas that are not as extremely rural as some rural areas in Australia or the US for example.
I almost shouted "YES!" when you mentioned that public transport is a much better alternative than self driving cars.
it is not though? it's terrible in sparsely populated areas, which is like most of the world when you exclude cities which are the only place where public transport is actually good and *should* be the preferred transport method
@@aonodensetsu Why would it be terrible is sparsely-populated areas? Surely, that just means less frequent arrivals?
@@Capybarrrraaaa And exactly that makes it terrible?
Imagine having a bus that arrives once per hour and needs double the time to arrive in the city, further away from your workplace than the parking lot.
Thats public transport in rural areas for you. Unless people are EXTREMELY poor (and not even that really counts cause public transport is expensive as hell), you will never choose to waste up to 2h to/from work.
Whereas self driving cars are much more flexible and can accomodate those situations much better
Hell, i live in the city and need about 50m with public transport (with a tram+bus that both arrive about every 10m), with a bike i need 35m, with a car id need 20m. Public transport is an intermediate solution, but ultimately f sucks for nearly everyone
@@leonhopfl573 I think you're looking at public transport, exclusively, though the lens of a car-heavy society.
Wouldnt having more people in a single car maximize casualties if a car crash happens?
On the topic of driverless cars - it is the last thing we need. What we need is everything else: Trains, Trams, Busses and Bikes.
I was in Freiburg (Breisgau) a short while ago. It is a city with more bikes and public transport than cars in the inner city - it was simply the most beautiful, relaxed city I have ever seen. 🙂
Its already horribly hot in Freiburg without cars, so its good that they only run trams in the most spaces. They need more greenery in the city center for sure
I was so captured by Mullers statements!! I want to be able to quickly step in a cheap, selfdriving vehicle everywhere I go. And I want everyone to be able to do that. To enable this, I propose to put them on rails, to make it safer and easier for the algorithm in all weather, and make them bigger and longer, so they can take many people to their destination at once. Selfdriving vehicles for everyone! That would be an amazing future!
Oh yeah. As the name "self driving vehicle" is quite long, I would propose a new name. Maybe "train" or "tram" or something like that. What do you think?
I think we already have infrastructure in place for cars, and a train won't fit in my garage
@@madthough2298 you won’t need a garage if trams become widespread which in many parts of the world they have
@@dangerous1580 Do Lexus make Trams? I'd want as close an experience to my current form of transportation as possible
@@madthough2298 no, they don’t but there are many other companies who make high quality public transport, and I as someone from Sweden living in the suburbs can vouch and say that public transport is objectively better than cars, you can watch videos from either “not just bikes” or other educational UA-camrs making content about urban and suburban transport and design
You don’t have to walk longer than 1km to the nearest buss stop and almost always less than 500 meters, these busses go to very 5 minutes to every 15 minutes depending on the popularity of the route, and not to mention but this is in the 11th biggest city in Norway with only 61 k people (or less than a tenth the population of San Francisco) the city is named asker
45:22 I detected serious Adam Something public transport advocacy vibes coming on and very happy to see I was right.
Self driving cars is basically the monorail solution in car form
Adam Something, the person who refers to things he disagrees with as "completely retarded" and who just made a video with Vaush, who talks similarly.
@@camelopardalis84 damn, guess I better disregard all of his points because he uses harsh language (btw neither Vaush or Adam use "retarded" to describe the things they don't like/agree with).
@@lolusuck386 Of course they do. That's how I found out they use such language. And it's not "harsh language" - it's a slur based on - at best - disrespect for a marginalised group.
@@camelopardalis84 What video did Adam Something say retarded in? Also Association fallacy is cringe.
On the airplanes, I’m a flight instructor, what you said was pretty correct(aside from the localizer just being one part of the ILS system, in the picture there you can see the second part which is the glide slope) and what veratasium said was technically correct, but both you left out some important details to frame this into a black and white automation good automation bad situation, which it just isn’t. Automation with human watching over the shoulder of the automated system is really the safest way currently. No way would I hop into the back of a driverless vehicle, but that doesn’t mean I wouldn’t want some driverless capability.
For example airplanes nowadays have RNAV and in particular LPV approaches, which basically do the same thing as ILS(though they tend to have higher minimum visibility and ceiling requirements), but are done via GPS and satellites, with no ground based navigation aside from ATC direction(though it’s sometimes not used if you’re doing a practice approach or are in VFR conditions, and while most don’t land the plane for you, it sure does get your close to where the only thing left to do is the flare. Though Garmin recently came out with an emergency autoland system, where the plane will talk with ATC, pick an airport, go to the airport, follow an LPV or ILS approach in, and land it at the push of a button, all be it for smaller airplanes.
Veratasium missed a bit too, for the same reasons you stated about it being a more involved process, but while it may seem impressive that a plane can land in little visibility, that’s basically what the entire instrument rating is for, so you can fly an airplane when you can’t see out a window just by reference to your instruments. Pilots can easily shut off the CAT III autoland and land it themselves no problem, and most do unless it is legally necessary in like zero visibility conditions(usually because they give notoriously hard landings and what’s the point in being a pilot if you can’t fly the thing every once and a while).
Airplanes too are surprisingly much easier to automate than cars. For one, you aren’t dealing with lots of traffic comparatively and where autopilots are mostly going to be used is under ATC guidance(unless you’re landing at an uncontrolled airport). You also have instruments which will pretty much tell you everything you need to know about where the airplane is and what it’s doing, and everything happens in an much more predictable manner. Many airline pilots will tell you that they’re almost like actors following the same script every time when they fly, and while yes you can get things like crazy wind every now and then, you can totally automate corrections in. Even with all this automation, there’s a reason why there are two pilots in the cockpit, and why autonomous planes aren’t a mainstay in aviation. Every pilot whose flown with an autopilot has dealt with an autopilot being stupid, and I’ve had to flip off(both the button and my fingers), the autopilot to actually get the airplane to do what I wanted it to. You need someone to step in because computers get stuff wrong, more often than not. The safety benefit of computers isn’t that they’re smarter than the pilot it’s that they’re consistent. Computers are human inventions after all and if they get bad information, garbage in, garbage out. Even as someone that has used automation to fly, I would never move to a passenger seat if the plane was doing an autoland.
Also your Asiana 214 statement is correct. A lot of the time pilot error related accidents happen these days because of what is called mode confusion, or as the ever present line goes in aviation “what’s it doing now?” Pilots can get confused by automated system just like a driver would get confused by someone else doing something stupid. However that doesn’t mean that “automation caused the accident”. It was human error, and the crew made some really poor decisions, though unlike veratasium is implying here going for a non-autoland was not one of them, like I said before it’s quite common because computers don’t make the smoothest landings and again computers can be stupid. The pilot is at the end of the day responsible for the flight, and just because the computer is acting up doesn’t rid them of that responsibility(either that or the pilots are easy scapegoats for glaring automation issues depending on where you stand on the political spectrum). The 737 Max case being a very rare exception where it was the case because Boeing made the decision to not tell their pilots about the MCAS system onboard their airplane. If they had so much as put in a PowerPoint slide about the system, it would have been called pilot error. It was a pretty bad pick of an accident by Veratasium, the better pick for an accident could have been Air France flight 447 where the autopilot came off because of an instrument error, and the relief pilot incompetently put the airplane into a stall for almost 20,000 feet until the Captain came in and saw he was acting stupid. But by then it was too late.
That all being said I really enjoyed this video. Although I say all these things, really this is like a lawyer hearing two non-lawyers debate law. Like you know they are both wrong on some things, but you weren’t really expecting perfect information. But imperfect information is different from misinformation, and this is definitely growing problem on UA-cam and you did a good job calling it out.
you know? this is the real problem i see with youtube educators: they try to be jacks of all trades when it comes to their content, but they arent. I wish they would, at least, consult neutral experts -if possible- and have a good peer review when writing their scripts and choosing their evidence. It would be a lot better to have channels dedicated to, and only to the discipline of expertise of the one/s who research and write the content.
For this reason I never truly trust youtube content, and less when theres sponsors involved. I would think its quite obvious they wouldnt speak neutrally of the products sponsored, and its just a proposal, but i guess some people take this too seriously.
I'd like to add that they are aircraft out there that can take off and land themselves including managing the guidance, flaps, throttle and landing gears, most notably military unmanned drones like the global hawk and reaper drones. But those also have a human operator present in a control room to take over at anytime.
V interesting read, thanks for sharing such info!
Thank you for sharing what you know! This was valuable context. I knew automation + human teamwork was best but you provided a lot of learning material that helped me understand why.
I've done enough flight school for a PPL (no IFR yet) and read a ton of air accident reports. As a person who's actually a software engineer this issue of aircraft modes horrifies me. Modal systems in software applications are avoided as much as possible in all modern systems because they're simply too hard to use. Users get confused about which mode they're in all the time, and this is even though there are often very clear mode indicators near the user's centre of attention. From what I've seen of the modal systems in aircraft the output of which mode the aircraft is in is not so easy to determine and takes mental effort to happen -- no wonder pilots get confused about which mode they're in (and the fact that the aircraft is free to switch modes without pilot interaction makes this even worse). It would be far more surprising if mode confusion didn't happen, and I can only assume that the consequences for mode confusion are generally benign or air safety would be much worse.
I get your point. I dislike such obviously paid content too. I think, however, you went overboard and now you produced a very one-sided view on the topic as well. Just for the clicks. And therefore ultimately for money too…
"Most of these errors are impossible for a machine to make"
Yes it'll make all new ones.
spaghetti code in real life
Machines don't make mistakes, only people do. Machines simply run; the only mistake is from the person who made it
@@christopherreichle6670 That's true without a doubt for programming in the usual sense, but not so much for the machine learning pipelines that are the core of self-driving systems. You don't really tell the machine to look for a stop sign, for instance, you just show a bunch of pictures and tell it where the stop sign is (along with other signs, traffic lights, cones, pedestrians, other vehicles, and so on). You then set up a machine learning model, typical a neural network, that analyses new images and tries to understand what it's seeing. You, as the designer, can improve its performance by messing with the neural net architecture, or the objective function for the training, or by giving it more training data, or by training it longer, etc. Doing so effectively will result in fewer errors. But errors cannot really be eliminated in this approach, so it is fair to say it's the network that's making the errors (though I suppose morally speaking the responsibility still lies with the designers because it can't lie with anybody else).
I mentioned the image analysis part of it specifically but really there's likely a whole lot of machine learning bits and pieces for which version of the the above apply to a larger or lesser extent.
@@christopherreichle6670 Basic cleaning robots make mistakes. Even if it can be corrected by changing their program.
@@2adamast machines don't make mistakes. They can't think for themselves. Mistakes are accidents made due to an action made under false pretenses. Machines have no false pretenses they just have instruction lists that are chosen from the logic structure that we gave them.
The content of the video is well made and interesting, but funny how he talks about misinformation promoted by sponsorships and then proceeds to tell us how a VPN will make you anonymous.
weird to me how no one just says "use tor"
It would be worrying if Tor Project would ever start buying sponsored ads on YT.
sure. But I think even that shows the point. I believe everyone expects ads to be somewhat manipulative. Tom's talking about a situation where the ad is integrated into the entire video.
Yes, a quick ad break is exactly the same as Veritasium making an entire video for his sponsor wherein he seriously downplays the downsides of the sponsored product.
@@SuperGreentoker literally yes
Thank you. Sorry for the long post but I develop machine learning systems for medical devices. When the the Veritasium video came out I was really irked by it. It was such blatant corporate propaganda and the critical thoughts (if you can really call them that) that were expressed in the video were laughable and clearly designed to be straw mans. I am always annoyed when people overstate the capabilities of technology and even more when they lie with statistics.
Machine learning is just a fancy term for function fitting. Like you did in high-school where you fit a polynom to some data points. That is basically it. There is nothing more to it. Our models may be a lot more complicated and the data sets a lot larger, but in the end it is function fitting. And the fundamental problem is: You cannot make sensible predictions in regions of the input space where there is no data.
And therein lies the problem: It's easy to get data from "normal" situations. Like driving along a highway. And since it is "normal", there is little information in this data according to information theory. But the interesting case is: What happens in difficult situations? You won't get much data there because by their nature they are rare. You can try to simulate it. But the problem is: How do you know whether your simulation is reasonable complete? You don't know what you don't know.
And how do you proportion "normal" situations with critical situations in the training data? Too little and they will get overwhelmed by the "normal" data and too much and your algorithm gets a distorted sample of reality.
And regarding the evaluation: A car driving thousands of miles means nothing on its own. It is the composition of the test dataset that is important. If most of the driving happens on a highway that is not really impressive. We already had a working system in the early 90s (ALVINN) that could do that fairly reliably. Of course we have made some progress since then. But the idea that you could put one of these cars in a highly dynamic environment like New York City or any European city and expect it to work reliably is ridiculous. There is a reason Waymo is located in a car friendly city. Very limited bike traffic and very limited pedestrian traffic makes for a less dynamic environment.
I mostly agree with your points. But in the "not normal" situation, both human and machine will perform poorly. Human just has the advantage of "may be" that machines don't.
i.e. human will be like "I am fucked. But if I do this, then may be it will work or make it a little better". Machines don't have that. So they are like "welp. Gotta sit tight"
@@youtubeusername1489 The problem is significantly more complex than that. humans seem to be better at knowing what they don't know, artificial neural networks clearly do have some similarity to the brain and there have been papers about that, but the amount of similarity is not enough to have all of the benefits yet. humans seem to be much better at understanding a new experience correctly the first time still. artificial neural networks can generalize but they're fragile in ways we don't totally understand yet. ultimately what that means is that small changes that a human wouldn't even pick up on can result in the model not having any idea what's going on. meanwhile it performs at superhuman levels for situations that are imperceptibly different from failure cases. unbeknownst to you, perhaps it's because somewhere in the neural network pink and yellow in particular positions on the image happen to confuse it and you need to slightly move the pink down because no car in the original training data set had pink and yellow in those particular pixels. or something. we can even construct examples that cause a neural network to fail and they are mysteriously too easy to create and don't even seem to change the image. neural networks are amazing, but you're still smarter than them, and artificial ones are not to be trusted yet
@@youtubeusername1489 Humans are vastly better than self-driving software at handling situations they have no training for, because humans have general intelligence. They can use world knowledge and reasoning to navigate situations they've never seen before, and to correctly recognize when there is something unusual going on in the first place.
@@notnullnotvoid sometimes the body language of another driver lets me know when they're about to cut me off or change lanes without signaling. Can a computer do that? Napes.
Fellow engineer working with machine learning here--I think a lot of your specific criticisms could be addressed with GANs. Granted, there's still the question of "how can we know if this is reasonably complete", but I think you'll find that the more important problem with machine learning, especially when it comes to self-driving cars, is an engineering ethics question rather than a technical one.
Eventually, we'll develop self-driving cars that are genuinely better drivers than people. How will we know when we've hit that point? Can we get there without experiments that risk the lives of other people? When someone dies because of the AI, who is responsible? These are really important questions that we don't have answers to yet, and we really should *before* we see *any* of these things on the road. No matter how good your neural net is, it's going to make a wrong choice eventually, and we really need to be ready for that eventuality before it happens.
idk these critics seem very negatively minded in my opinion. You make some good points about sponsorships and probably not enough negativism from Veritasium. But there are no really wrong numbers that are hugely misinterpreted. The point to blame is that most concepts at the moment are more niche uses of the driverless concept than one might expect at first.
Btw did you really blame the car to be too cautious when someone is nearby? As long as the car does not stop forever and for every person nearby i would consider that braking fine
Man it is hard to watch this, with veritasium being one of my favourite channels, but in the spirit of learning, it pays to have a balanced view by hearing both sides. And after watching the video (and being a pilot myself) I have to agree that the Waymo video was a huge hype piece that left out a lot crucial information (like an overreliance on automation being the cause for accidents in the aviation piece for eg.). Another aspect of automation to think about this is "I know that I dont have to worry too much about the condition of my taxi's critical systems (brakes for eg.) because my taxi driver would have checked on them in the interest of self preservation." Its a new issue that would arise out of complete automation and I wouldn't want to be a part of the corporation's "acceptable risk parameters" statistic when it comes to safety and integrity. I'd rather be the cause for my own injury than be part of a cost cutting statistic. TLDR, I'm happy I gave this video a chance and will keep my head about myself when it comes to completely sponsored videos.
ford be like: there is no unacceptable risk parameters
I think the recent Physics Girl's series on hydrogen cars was a pinnacle of quality degradation caused by sponsorship. Instead of making a video about the technical aspect of how hydrogen fuel cells work, she spent about 20 seconds saying that it's "basically magic" and moved on to praising them.
Her videos were the first thing that came to my mind when watching this video.
personally i have never watched an ad on youtube and even remotely thought about buying a product. matter of fact , if i see an ad for something on youtube? i will not buy from that company ever
yeah all the recent videos on hydrogen cars often ignore or gloss over a lot of the massive problems with the technology. They're just like "It'll take some new infrastructure being built but we can get there!" completely ignoring the actual sheer scale of that infrastructure, and the fact that one of the primary advantages of Electric cars is that they don't require anywhere near that level of infrastructure.
@@o0o-jd-o0o95 Well basically every brand ever at this point has advertised on UA-cam
@@BlazeMakesGames Well electric cars require a gigantic amount of electricity infrastructure on the level which requires a ton of investment. Countries like Germany could not hope to supply even a 20% electrified private car environment when there's hogh charging demand. You're right though, hydrogen infrastructure is expensive and fuel cells are less efficient. I do wonder about hydrogen combustion engines though.
"Just build a train, idiot!"
- Adam Something
⬜ SERCH ADITYA RATHORE-HE ALSO MAKES INFORMATIVE CONTENT LIKE TOM NICHOLAS
Haha I get this reference immediately
lmao
"Train Good Car Bad Horse Chaos"
- Guy that sounds like he's been smoking unfiltered cigarettes since the 60's.
Funny enough that a video proporting about corporate propoganda struggles from the same issue. Granted this channel isn't an education channel nor was the entire video based on a sponsored product, but VPN's are almost a worthless product and here is Tom making a biased ad in support of one so he can get paid. Quite commical if you ask me. If you want to be a solution to the issue, maybe finding a useful product to be sponsored by (though they are quite hard to find nowadays), or not be sponsored at all and rely on Patreon.
as an IT engeneer I must ask - since it's not corporate propaganda video, how is VPN making me safer?
I think that just underscores the fact that the real problem isn't that youtubers taking sponsorships, but that companies are willing to spend it and youtubers have to take it.
Yeah, not to disregard the points Tom makes in this and other videos by falling in a tu quoque fallacy, but VPN sponsorship segments are truly questionable in their affirmations. Tom Scot has a good video about that if anyone following this thread is interested.
It's a bit heartbreaking to see that ad in this video, yeah. VPNs are some shady, scammy shit these days and Tom is falling prey to all the things he's accusing Veritasium of doing. Basically, you have to treat any and all advertisements as entirely worthless - not just skippable or dishonest, but actually worthless, empty pieces of noncontent.
@@ssj3gohan456 Spiting facts
@@martinovallejo A great video
As the great Garfield once said
No one is immune to propaganda
But some are immune to propper-uganda.
No, stop spreading misinformation. Garfield said "No one is immune to lasagna"
Veritasium also disappointed me by doing an interview with Bill Gates where he puffs him up as a visionary for predicting that a pandemic would happen and we wouldn't be ready for it, but doesn't ask or mention anything about how Bill Gates himself has contributed to making covid vaccines far harder to access in poor countries
*this
Not that I want to take to for Gates, but wasn't that video filmed before the vaccine even came out?
@@sanders555 not sure, but I went back to the video and it turns out I was wrong, he does address the concerns people have with his interference in preventing the vaccine from being open sourced by completely uncritically allowing Gates to give a very weak defense of his actions. It's actually much worse than I remembered, avoiding the topic would have been preferable, I've gone from feeling disappointed to feeling outright betrayed
@@HamidKarzai It wasn't that long ago that I thought Gates was just another random target of the batshit birther QAnon cult before I realized there was something to it. It's a shame they're so off the rails that they're actually helping to provide cover for actual celebrity villains and their misdeeds by lumping them in with all their fairy tales of Presidential resurrections, pedophilic lizard people, and powerful Satanists with a huge predilection for green energy, universal healthcare, and authoritarian communism.
@@sanders555 if you were a conspiracy, wouldn't you want conspiracy theorists to all be insane? wouldn't you want to do anything you could to cause them to be more likely to go insane, so they can't actually figure out what the conspiracy is because they're too busy assuming everything is a conspiracy :p
conspiracy is a real legal charge for a reason and it does occur, and it typically isn't as all encompassing as conspiracy theorists think. It's typically more mundane, people following incentives locally rather than actually working together in conspiracy, and when they do work together it's in groups that don't break the rules of reality. veritasium himself documented an example of an actual conspiracy, and like most actual conspiracies it was much more finite than a conspiracy theorist would propose. in general it should be expected that if there is a lot of conspiracy in the world it would be coming from many small ones rather than one big one. see also zoe bee's video on how to talk to conspiracy theorists.
Those self-driving cars just recently expanded into Downtown Phoenix. I've encountered them more and more on my drives down there, and they're out in full force during big events. The streets in DT Phx are 2 lanes wide, some areone way, and always have cars parked on the side. This is pretty usual for other major cities but in most of Phx its unusual. These Waymo cars regularly get confused, stopping a lane for hours. If a car is parked just slightly over the line, they stop, thinking the car is pulling out. During First Friday I witnessed a passenger prematurely leave a Waymo, which caused the car to panic and stop dead in its tracks. Both lanes of traffic broke down because people stuck behind the Waymo tried to pass it, but they couldnt get around the Northbound land people in the NB lane couldnt back up to let them pass because of how conjested the street was. I was stuck there for about 30 minutes before the technician was able to walk up to the car and get it moving, letting traffic flow again.
Since I can edit on mobile...
Additionally, the comparison of Self driving vehicle AI systems to Aircraft control system is comparing two different beasts. All commercial flights are heavily monitored and follow generally the same rules of flight (not all countries have the same regulations, but the physics of flight are restrictive enough that the rules are generally universal). If we were to compare SD cars to aircraft in good faith, there would need to either be way less private cars on the road (or dedicated lanes for SD and private cars) or there would need to be way more private, unmonitored aircraft flying at the same elevations as commercial aircraft. Its just two entirely different beasts that you cant compare those systems at all.
The VPN ad spot is a weird choice for this video, especially all the talk of massive companies wanting your data (true) but then saying a VPN can help you rest assured you're not being tracked is complete bullshit, and just technically inaccurate. Most large tech firms (websites in particular) can build fingerprints of specific devices, so changing your IP does piss all for most people who actually want your data, along with this given the number of breaches coming from VPN companies you're actually better off getting sponsored by linode or something with their 1 click private VPN through openvpn, you don't NEED locations anywhere in the world to have a useful vpn, that's used to bypass content filters or restrictions which i don't have a problem with if they would stop acting like no one uses these services to watch netflix shows not available in their area.
It stops websites from being able to track you? this is just completely false in most situations. because a vpn doesn't log you out, nor does it have the ability to do so even if it wanted to, and you're easy to track if you're logged into any website with or without a vpn. marketing vpn's as a catch all solution to internet security in my opinion is on par with the criticisms leveled at derek in this video, that's not even getting into the fact their basically pulling a fallout 76 by saying it's 83% off, and it always is, it's literally a tactic proven to push people into buying a product because "oh, that's a good deal, and it can't possibly be that cheap for long", and the idea that their service is 83% more valuable than they can sell it for making it seem worth more than it is. it would cost me $5/month to host a vpn anywhere a cloud provider has a location which is now most places in the world so they can't possibly expect much more than that for a closed source less transparent version of free software running somewhere in the cloud that you own completely, along with the fact you can pay monthly, often times by the hour for a cloud compute provider which means if you don't need it/want it anymore you can just turn it off, and you only pay for what you used, in contrast to being sold something for two years, which you likely only use for netflix, i doubt anyone actually concerned with internet security would use a public vpn although i could be wrong. also it's a little ironic considering youtube tracks people, they in turn get recommended videos to their taste (possibly this one), at which point they watch the video, some with ads on, you get paid by youtube, then sell them a product with the proposal it will stop them being tracked which if it actually could, would pretty quickly kill any channel recommending it because youtube wouldn't know the first thing about them and we all know the subscribe button means jack shit now anyway, along with the bell for the most part.
In the most basic of examples: Who would believe you're living in Europe when all your amazon packages go to nowheresville, US?
Indeed, as someone who works dealing with information security, uses and hosts his own VPNs all the time, knows what they are and how they work and their limitations. His sponsor spot is just as disingenuous as the waymo video.
Is there a use for VPNs yes, is there uses for waymo self driving cars, yes. Are they what they are being made out to be in these sponsor pieces. No. Lets just say, if I'm using a VPN in a security that I'm not hosting personally, it is by no means the only measure I'm taking, or the most important and is only being used for specific circumstances.
@@EwanMarshall +1
Indeed, there are multiple videos about VPNs that discuss their weaknesses.
Finally see this being called out. I noticed it a lot of times, I think primarily on LinusTechTipps, but never never went to comment on that as it's not a relevant part of the video, par of the course and most people probably don't care and it would be lengthy to explain what it does and what it doesn't (like your comment^^).
Particularly in this video it's somewhat relevant to point it out though, being after the truth and criticising misinformation from other videos, but actually spread misinformation or half-truths in an ad.
Also would be nice to see this comment further up, so more people get a better understanding of the use cases of a VPN.
the surfshark add is very funny in context, considering how vpns dont keep you safe or secure. it just swaps who is snooping on you from your internet service provider to your virtual private network provider. is good for flaunting laws ur government cares about that your vpnp doesnt, but dont expect to never ever get caught
Add break vs corporate propaganda fully integrated into the content. There is a major difference.
@@raulinurminen7299It's not just an ad, it's actively misleading. Of course an honest person wouldn't bullshit you while exposing someone else, right? That's what makes the ad so valuable.
@@raulinurminen7299 What is the difference? It's just a difference in style of presentation. Misinformation is misinformation.
Really appreciated the video until the sponsored part. VPNs can be a part of a security-strategy, but claiming they protect your information in general on the internet is simply false. Most tracking is done via cookies and fingerprinting which a vpn is not protecting against.
I really appreciate this video but tom is in a really easy position to just make explanation video without any feedback video even after the person said give clarification back. Tom will NEVER wrong in a video like this even when CLEARLY he knows vpn is bollocks and still got that sweet sweet SPONSORED money.
He in the position of everything he said is right even after said person clarify, that least it's not cancelation of the person in the video happened.
This is the exact same way Magic of Rahat got accused and lose everything until 2 years after revealed he ALWAYS in the right. I just afraid this will be the precursor of it.
I hate to see those vpn misinformation ads, especially from creators that combat other misinformation. However I would also say that there are by far many worse sponsorships out there.
I think most people use vpns to watch shows in other countries rather than for security reasons.
Yeah I also noticed this. I mostly agree with the overall argument that "veritasium did a bad"... but those claims about VPNs are not entirely true either. Having said that, both videos are entirely different beasts. Tom has a small sponsored section about VPNs and that's it, he doesn't make a whole 10 minute video saying "why surfshark good", which is basically what the waymo video is.
Hate it as well, but do we think he has the luxury of picking sponsors? Also fact fiend has shown how shitty these sponsors can be with the contracts where you have to say what they want otherwise you don't get paid. That's why every audible sponsorship has the person say what they're favourite book is and repeat the whole last line twice, cause they have to to get paid.
I think, public transport companies all over the world should do the same waymo did: sponsor educational yt-channels to make (good and well researched!) videos about the high benefits and all the problems, public transport would solve.
Honestly, this was really great as an introduction to the supposition of neutrality (political and ideological) not only in educational content but in the scientific field as whole, going forward i really hope this topic can be further explored
Sabine Hossenfeld seems to be relatively good at that topic. She most certainly has her own biases, but she's seems so much better than most.
This hits way harder for me than I think it would for a lot of people bc I actually lost my dad in one of those “human error critical moment” accidents and had to face a similar sensationalization of his death and more specifically his blame in it from our local news when they covered it...
my dad died after taking a curve too fast and rolling his car. He was also not wearing a seatbelt. When the news reported on this, both in the local paper and on tv, they put a huge emphasis on my dad /speeding/ around the curve and how dangerous he must’ve been driving to crash like that and how reckless it was bc he had a kid (my brother) in the car with him. They also REALLY emphasized that he wasn’t wearing a seatbelt in a way that kind of implied that he deserved to die for not “taking his safety more seriously”. They even speculated on whether my brother was wearing a seatbelt, basically trying to make my dad out as a bad, dangerous parent based on something they completely made up about the scenario.
Critically, they didn’t include any ACTUAL information about my family or the accident other that the things I listed above and never asked any questions of my brother who survived and witnessed the accident, the state troopers who found their car, or the paramedics who worked on them.
In reality my dad took a curve too fast bc rural back roads where I’m from are notoriously ill maintained, with no guardrails, huge drop offs, loose terrain, and vegetation or other debris blocking the street signs. In this case the curve was supposed to have a speed limit drop with a warning about the curve that had been overgrown by plants, had loose gravel on it that had just been left over from snowfall a few months before and reduced tire control, and overlooked a large embankment with no guardrails or safety measures to stop cars from rolling ALL THE WAY down it. His only failing in the accident was not being able to correct for a bunch of errors on the part of the state in maintaining the road that he didn’t know about until he was actively mid-crash.
The reason he wasn’t wearing a seatbelt is less black and white that the cause of the crash but still very nuanced. My dad had previously survived an accident that killed every single other person in it when he just forgot to buckle up and got thrown out of the car while everyone else was crushed when the cage less convertible rolled. After that he was so scared and so traumatized he was genuinely /scared/ of his seatbelt. He knew it wasn’t logical and we all tried to get him to wear it anyway but trauma isn’t that easy to fix. He still insisted everyone else in his car buckle up.
Which brings me to the last fallacious idea they threw out when reporting; my brother was in the front seat bc he was over the required weight, height, and age for a child to sit in the front, was wearing his seatbelt, and even told me after the accident that my dad had physically held him in the seat, putting as much of his body in the way of the impact and debris as he could to protect my brother.
They basically demonized my dad bc they needed something to blame the accident on and knew they wouldn’t get much from focusing on the bigger issues of the bad roads that politicians in the state didn’t care about fixing and that citizens would support only until they realized that fixing roads and saving lives costs money that’s going to eventually come from them.... needless to say it made it a lot harder for our family to mourn the sudden loss and for my brother to go thru therapy while every news outlet on the tv, radio, internet, and paper in out small town was running lies about my dad to make him at fault.
"the media is the enemy of the people"
Condolences.
My younger brother also died in a head-on crash.
I have often wondered if the oncoming truck had been automated, would it have been “programmed” for proper safe speed. Would the third car near the scene also have avoided the scene by following all “rules of the road?”
We may never avoid all deaths, but seems like we have a safer future ahead of us when “on board” systems double check on the signs to keep us safer.
Either way, we need honest and full disclosure of all options and have the FREEDOM of choice as to whether to protect ourselves. That said, what freedom should we have to endanger others? Perhaps our vehicles should override our choices when we fall asleep, DUI, or excessively speed?
my condolences, this sounds like an awful, heartbreaking experience. I'm very sorry it happened, the way the media jumped to traumatize a grieving family sounds horrid
Thanks for sharing + educating, sorry for your loss.
@@THESLlCK fuck media and its journalist and journalism. These just make things worst
Him using flight 214 as an example in this context and with those implications is OUTRAGEOUS! I don't believe in suing people for everything, but this gets me mad to the point I feel there should be some accountability! And to imply pilots choose not to use auto landing because of hubris or ego just shows how he knows NOTHING about aviation! Guess what, pilots need to train manual landing actually, it's pretty import in the case, you know, in the case the need to do it.
It was absolutely tasteless and Derek is still fully doubling down in his comment of the video, the biggest yikes indeed...
it's funny how many commenters here follow the same channels, and how many commenters are actually owners of those channels. And now I find fellow aviation enthusiasts, too? I guess UA-cam is better at connecting like minds than the FBverse.
I recall from the time of the 214 accident that the biggest contributing factor was that the airport had turned off the ILS system but kept the airport open while they moved the runway itself. This would be like running the Waymo service on live roads during an outage at the central map servers. The automation couldn't save them because the primary automation input was turned off. Also that particular runway isn't easy to aim for without computer assistance.
I agree that a better argument for self-driving vehicles, when it comes to weird trolley-problem-like scenarios, is that there's approximately no human drivers I'd trust to make a split second decision about who to run over either.
But as it is a split second decision it is very defensible. Someone programming who to kill in such a scenario without any pressure is less defensible I think
@@user-hl7ic7wc1r I don't agree. By the time someone has to decide it's too late, but they knew the risks when they drove thousands of pounds of metal into a public place at faster speeds than anyone can run.
The programmer, not being under pressure, can more easily just instruct the AI to hit the fewest people possible, which should always be zero, and when it isn't that's when planning committees need to get together to find out why it happened and prevent it from happening again.
@@amymason156Who is responsible for any accident? Is there a law? How can we discuss about this when Even the Most Basic thing doesnt exist yet?!
@@PM-wt3ye Just like this, that's how. Who's responsible or if there's a law isn't as important as how accidents can be prevented, which we're already aware right now usually result from humans not having the reaction time, situational awareness or ability to communicate they'd need to avoid them. It's easier to get humans out of those situations than to turn everyone into super-cyborgs...
False, it is the MOST important and most basic thingof all. If there is no law in case something terrible happens, theres no need to prevent it from happening, like in any other situation in the world. In case of an accident, there has someone to be hold accountable for. Denying the impotance in this situation is ABSOLUTELY rediculous and shows, that we are FAR away from establishing this correctly...@@amymason156
Man it’s really disappointing to learn that some educational channels you thought were trustworthy do stuff like this. I really appreciate these videos, they’re a stark reminder to always remain critical of the information people give you
I have to steer away from everything vlogbrother-related for such reasons. Hyping Gates, war criminals like Obama, telling their audience Biden is wonderful even after the main election ...
They literally state they are being sponsored... How do you miss that? Even this video is sponsored. That should always factor into your mind when watching a free video. Anytime you see a company giving access to an outside person, you should be critical of what you are watching. It does not take away from the point of Veritasiums video, that self driving cars are and will be safer than humans. And I am a skeptical person about that myself because of all the odd roadways and signs and missing road lines, etc.
@@NandR you expect them to be as neutral as possible even when being sponsored, that's what being an educator should be, I haven't seen the waymo video but I'd expect pros and cons of the technology with the least amount of bias possible, but I guess that's asking too much of today's UA-cam
Any video that is sponsored by the subject of the video is an advertisement. Use your adblock and don't watch it.
What's worse is in his reply to this video he's doubled down and what's worse is he implies he's just as uncritical to any other cause he deems noble, so yeah steer clear of this man.
45:20 I instinctively shouted "Trains! Just build god damn trains, trams and metro!". Thanks Adam Something, I guess...
Conover?
Yeah. After watching Adam's videos for a few weeks, I am honestly baffled how a lot of suburban problems can be effectively tackled by improving public infrastructure. Guy single handedly change my mindset
@@felixstenson1654 No, Adam Something on youtube. Hungarian youtuber who makes great arguments for public transport over even more cars to solve the congestion problem.
@@headerahelix Omg, I fully follow him haha! 😅
I see you're a man of culture as well.
I like you Tom, but was this really the best video to include a paid promotion about a VPN where you add misinformation about what they're able to do for you?
Yes, I first thought he was joking when the promotion came up. The irony is painful.
VPNs do offer some services that people may be honestly interested in. Having a VPN ad break is not the issue the video is talking about in any way
@@kirklandday The issue is that he, like many creators misrepresent the capabilities of such VPN services. In that regard it's pretty damn close to what he critizises. The only difference is that it's an ad break and not the entire video.
@@shiniri2647 That's kind of the point though. This video isn't being made to say "sponsorships are bad", it's just to encourage larger 'educational' content creators to draw a solid line between what the commercial bit and 'educational' bit is, which Tom does very well with his VPN sponsorship.
If he made an entire video going on about how good VPNs are, that would be the problem but he isn't. I do see the slight irony but I don't think it takes anything away from the point of the video
"Human error causes 94% of traffic incidents, therefore AI should drive intead of humans"
There has been not a single traffic accident caused by a body pillow behind the wheel. Therefore, all cars should be driven by body pillows.
my 02 bodypillow is better driver than me
The most notorious "educator" in Italy is probably Piero Angela (now 92 and still very smart and lucid) he always worked for the public television and when a private offered him considerable more money he refused to keep his work as independent as possible and said "I eat two times per day, I don't need more butter or more meat"
notorious has more of a negative message, i think you might be looking for the word "famous"
Putting the word educator in quotations also has somewhat negative connotations.
We, from latin languages (Brasil here), use notorious because has the positive meaning for us, notable exists?
48:00 the worst part is that self driving public transport would be way easier to achieve (and significantly safer) than creating self driving cars
Right?! They follow prescribed paths/routes anyway :D
Yeah, same thing with electric cars. Electrifying many forms of public transport could be hugely beneficial and cost effective (slower speed, central hubs makes charging easier, higher purchase price is amortized over the fleet, etc.) but the focus is almost all on consumer electric cars because that's where the money is and because Americans see it as their flashy new tech.
You Europeans don't understand a thing. A big portion of the world doesn't have the climate for people to walk outside at 9 am and 12 pm for half an hour without showing up to work like an unshowered mess. If y'all don't invent driverless cars to sell to us, the rest of the world will.
@@mmmk6322 shhh let them shot themselves in the foot. Qhen the state mandated public transport gets limited to some arbitrary gvernment mandated standard, they will still support it, but the schadenfreude will be wholesome
That already exists actually. The Copenhagen metro is self driving and most high speed rail systems are basically fully automated and have been since their invention. As it turns out automation is a lot more feasible when you're in a controlled environment like a rail vehicle.