I disagree that AI is hurting "creative jobs" as much as you claim. Not everything that an artist makes is creative or even artistic, if a company could replace you with AI generated content without any repercussion, is because what they were using before (real people's work) was so bland and devoid of creativity that no one can notice it anyway. Jingles were mostly forgettable but once in a while there were good ones, AI is essentially very similar in that regard, sometimes it comes up with good stuff in a sea of mediocre uninspired stuff. In music this isn't AI's fault either, this is been going on since the 90's when they found out that autotuning a pop singer was much less hassle and more lucrative than signing a rock band that could create the next stairway to heaven or bohemian rhapsody. Good music is being produced a lot, but you dont hear it because mainstream is flooded with low effort urban music, so only niche listeners get to enjoy the newer bohemian rhapsodies that come from artists that could be making money elswhere but sacrifice in popularity to create what their masterpieces. My point being: If at some point our society needed a lot of "creative" professionals it was not because they wanted to create art, it's because they had skills to embelish and make stuff more appealing. That is not creativity, that's just valoring technical prowess, paying someone because they're more skilled at drawing, painting anf that can execute your vision better than a regular person that isn't trained in that. If an AI can do your taxes better than an accountant, would you claim AI is killing creative accountability ?
this is your first video I have seen (maybe recommended via Drew Gooden?) but excellent, amazing, I cannot compliment you enough. Very impressed and delighted on every front. You have a new sub! Looking forward to future (and past) videos! This is what YT is supposed to be! Keep up the phenomenal work!
You know the "AI now" with Willsmith is just Will Smith joking on the subject, it's actually him on video, not AI production. There are many points you push that are deeply ill informed. While I share a lot of your anxiety relative to the impact of AI it's important to be accurate and to portray the real issues without going into full panic mode. You are not informing people better than the ultra positive tech bros, you are just sharing as much disinformation as they do. What you are doing is called in psychology catastrophization, it's the unconscious act of taking the worse interpretations of events (or comments) and assuming that they are the reality. Reality tend to be way more nuanced. You are not serving your own objectives by doing this because anyone who knows anything of the subject will dismiss you out of hand (rightfully) as someone who don't know the subject. Creating a disproportionate panic only muddies the water and pushes the discourse to extreme irrational reactions not based on facts but feelings and end up having unforseen and unwanted consequences.
I saw someone say that they want AI to do basic mindless chores or work so they have time to do art, and not AI doing art so that they have time to do chores or work. I truly agree with this.
if they're training AI to do basic mindless chores, that will also have negative consequences. If it gets better, people will lose jobs. They already have been. Trust me, cleaning is not an understaffed job. And if they're good enough to clean or work, they can literally make art as well??? Much like now.
No one cared about computers taking the jobs of good singers and instrumentalists. Spent years learning how to sing or play an instrument? Sorry, someone that spent no time learning that used autotune or pasted instruments in a program has your job instead. And no one cared. Or AI potentially taking away truckers jobs with self driving trucks, and no one cares. People should just admit they only care about art being in danger because it's what they've attempted a career in and don't care about others careers.
Protip: When searching for generic images most search engines allow you to put "before:(insert a year here)" if you want only images before a year so you can put "before:2022" if you want to filter out most ai images.
Imagine, in 2050 we're still searching for *before: 2022.* That's such a sad thought. "Yeah all the people in the pics are wearing super outdated outfits. But they all have 5 fingers, so it's just more efficient."
So UA-cam lets open ai use hundreds of videos to produce generated schlock without the creators consent and calls it "fair use," but when a small content creator uses a few clips of a Disney film in their analysis/video essay, its a copyright violation. What a world we live in.
When it comes to business, particularly businesses like UA-cam, if you are able to do something for free that means you are the product. Including uploading. I remember the early days of Facebook, one of the terms of the user agreement was (and still is I think) that anything you post on there is actually their property and they can do whatever they want with it, including using it in ads. Places like UA-cam are no different. They will protect the Disney's of the world because Disney owns their IP and they are brutal about protecting IP, whereas the average UA-cam user is on their own because UA-cam actually considers themselves the owners of everything on here.
Keep in mind why would Google call out all these other companies for their data mining when they are quietly doing the same exact thing for their own AI products
Am I the only one who really really misses the old internet, like I already was missing the old internet where your real name wasn't attached to everything, and they were like specialty forums of people who shared your interests.. But now we can't even Google stuff properly
I don't miss it because it still exists ! Check out the whole "small web" or "indie web" scene or even neocities. Making a personal website is super fun a useful and it's making a real comeback!
I probably am not part of that old internet you're talking about but I still remember when you didn't need to create an account to get anything from the internet. If I wanna scroll in a photo gallery I need to make an account, if I wanna make a trashy beat I need an account, if I want a free cache cleaner app I need an account. Why does an online game need to know my government name...
I'm using the internet since about 2004, everything was so interesting back then. Now every online device is just a terminal to sell you things, including your watch and car.
My mom used to transcribe calls for companies as work, but then her work implemented AI to transcribe the calls, leading to work calls having sentences in them such as: “Thanks for watching!” “This venue has the best beat” and putting in random websites that DONT EXIST into the calls, when the caller said nothing like that.
My mom edits translated books and 90% of them are translated by AI and make no sense. They'll change the name of a character halfway through the story, mix up character's backstories, etc so she now basically has to not only check for spelling errors but also continuity errors, which isn't in the job description since it's assumed that by the time they're translating an already published book, someone else caught those errors. They really need to improve AI comprehension 💀💀
AI companies: "everything should be replaced by auto-generated slop" also AI companies: "we're having trouble finding good training data, it seems like everything has been replaced by auto-generated slop"
People say AI is only getting better in the future. Funnily enough it will get worse because its being trained on its own data now. AI peaked and openAI and other companies are scrambling to find more untapped data that they can squeeze out of the internet for "better" models. Truth is, they need to find a better way of training the AI. This method they are using of scrubbing the web is unsustainable.
@@fumaholics well not really exactly anymore now AI is finally becoming physical for like robots like optimus gen 2 from tesla and the openai fobot they will be able to go outside and see the world touch and hear just like humans do and just like a baby they will grow and it will take time but eventually they will reach a level of philosophy unmatched in any realm of human endeavour
@@gamers-xh3uc Yeah, but ignorance is bliss, it's easier to think that because AI is currently just regurgitating that means it's "peaked". I don't think people quite realize that humans evolved the exact same way. We imitate. AI just has a lot more to start off with than prehistoric man so of course it's overwhelmed and doesn't know how to use what's already been done to create new, better things but it's not as far off as we want to think. No art is truly original, nothing we do or think is. Humans overrate themselves, our greatest achievement is creating AI that will exceed and replace us. But looking at what we have done to the planet, it's a shame we didn't get replaced by AI sooner ...
have you caught that commercial from the Olympics? where a dad asks the AI to "help" his young daughter write a fan letter? What's the fun in fanmail if you do not put it in your own words? Writing to your hero should be something you want to do, not something you want to pass along like a business email to your boss
exactly!!! all the AI commercials make me feel so weird. there was one from Hundayi with a clearly ai-generated background (which i think was like, the stylistic choice?) and it just makes me think, "you can't bother to pay people to... drive around in the desert or whatever?" it makes no sense to me
Using AI to write fanmail sounds like creating spam. I mean it's different if it's just for polishing up your letter but to create some generic ai letter using a prompt... I don't get it
Maybe they were disadvantaged and therefore weren't given enough tax dollars to go to a good school that didn't have systemic programs in place like metal detectors. Maybe you're just prejudice? ?
I’ve been seeing ads on porn sites for AI face swap services lately. They even straight up say it’s “good to use for ex girlfriends, celebrities, or people you know in the world.” It’s sickening
my brother is a PHD doing medical data analysis for a research department at an in state college and he uses AI regularly to organize and detect patterns in the data. The AI doesn't make up the data, the AI doesn't do the research, the AI doesn't even get to know what the data even means. It's so incredibly useful and time saving and accelerates the process by days worth of time. That is the kind of thing I think AI is meant to be used for, helping you speed up a task not dong it all for you
I use AI to summarize results, create templates and translate things, but at the end of the day I still need to check and edit things. I might save a couple of hours of work, but I still need to do a quality check on everything to make sure AI's not altering the information. This kind of use isn't as glamorous, but leads to more accurate results.
And even for that it s still dubious. Is it worth it to put out of work research assistants to gain a few hours on the processing of these datas ? Is it really a gain for society as a whole ?
That's how all AI works and is supposed to work. It's just normies have been spooked by propaganda about making pretty pictures with just text, without constraints or consistency or anything actually useful. Because someone somewhere was selling this generic, non-specific kind of asset and they liked money. If you need to make some assets of a pink elephant monster for a game or a movie or something you can't just ask for it and get something you can actually use. You need to do a lot of the work and keep the AI use within rigid workflows where it's narrowed down to what you actually envision and working on mostly what you yourself made. It's just supposed to make you faster, which means you make more assets faster.
@@etienne8110 Yes it's worth making things faster and better. You're advocating for stalling things just to pad hours and wages. Late stage capitalism thinking.
CTO of ChatGPT when AI answers questions in a harmful way: "It's very human to make up an answer, humans rarely say 'I don't know'." CTO of ChatGPT answering a question where the truth would be harmful to her: "I don't know."
What's so annoying to me about that statement too, is so many layers of things. One-- no, I don't agree with this weird generalization about human nature. There are plenty of instances where people say "i don't know" there's a whole commonly used "idk" acronym for it, the only kind of person who would never say "i don't know" would be someone who has an incentive not to, like a politician or a salesman. The thing people may be misinterpreting as "not being willing to say idk" is when people instinctively try to offer some kind of guess to convey they are engaging with what was asked of them, but that is not the same as outright lying and claiming authority on a subject. Two-- WHY would a purely algorithm based "intelligence" behave in a way that implies inefficient amounts of human-like pride or social shame or some kind of sociopathic pathological lying habit and Three-- if such a human trait believably existed in your programming-- why would you sell an AI that is prone to that as if it is a useful engine for information searching and not simply make a program that is unemotional??? lmao its just so bad
I know this was a throwaway bit in the video but a lot of VA's have taken extremely hard stances about their voice being used even for joke things like silly character covers because it's genuinely terrifying to think that someone can make your voice say any number of absolutely horrific things but also because voice work is as much of an artistic skill as physical arts. I've seen as recently as this week someone who did VA work for a company have to start up a legal dispute with them because they used AI to supplement their work without permission or compensation
If someone used my voice through AI without permission, those would be fighting words for me. Fists are flying. Genuinely offensive, disgusting and an insult to my humanity.
I have always thought it was a bad sign how many people thought the AI generated Spongebob song covers were just sooo funny and sooo good, meanwhile I'm sitting here like "Okay it's a cute idea to have Patrick Star sing a Ringo Starr song, but that doesn't mean it's okay to just... steal the guy's voice?"
"Why would I learn to draw if the AI can do it better than me" is something I've been seeing a lot of in the drawing community. Imposter Syndrome has always been something artists have to work through but it's so much harder to feel like "a real artist" when someone can click a button a produce something more visually (or audibly) pleasing than something you spent weeks on, if not longer. It's heartbreaking watching young artists give up before they ever really get started
@@mrsticky005 A lot of people love drawing but are surrounded by negativity (from themselves and/or others like school art teachers or parents) and they convince themselves that even though they want it doesn't mean it's for them. Just because someone has a passion doesn't mean they feel they are allowed to feel that passion. That's why it's a Syndrome and they aren't actually imposters. I personally stopped drawing because of it until I was like 29 and had been through a bunch of therapy
the best thing we can do to combat ai generated image crap everywhere is to do our best to encourage those young artists. it's hell out here for them and we just have to do what we can to support them and encourage them onwards. Oh and ofc let them know that people who use ai image generation are pathetic hacks who are too lazy to put in the effort learning real art, and they're stronger than those people could EVER be by just TRYING to learn. every young artist who picks up the pen on their own instead of giving in and being a worthless ai image user is so strong and admirable
It's far more involved than just clicking a button - have you tried to use AI image generation? They want you to jump through a bunch of hoops just to use it and then you need to feed it good data and prompts to get anything even remotely approaching useful heh
@@Beckelby Welcome to capitalism where we can't have time to just enjoy life, we have to be "productive" and find a way to monetize our hobbies if we even want to afford to do them in the first place heh
As an artist, I really appreciate you grabbing MSPaint, drawing and image by yourself (those mountains genuinely look good, unironically and unbiased, you are cooking!) and using it in your video! It was genuinely a nice touch that really shows your humor and sensibilities and I will appreciate this a lot more than somebody going for that unrealistic polish AI is trying to replicate so badly! Jank is life, people! Embrace it! 🧡
I love how she said that "humans don't say 'I don't know'," but when she was asked if the AI was trained on videos from youtube, instagram, and facebook, she literally says "I don't know/I'm not sure."
She’s also wrong. I’ve seen several different AI programs say things like “I don’t know” or similar. For a professional on AI, she sure doesn’t know how it works at all.
I will say I don't know all the time! I'll often make a guess about something at the same time, but I'll make it clear that it's a guess. I hate misinformation. It's so, so dangerous
And it's not even true! Humans lie and make stuff up in certain situations yes, but doesn't apply to all the topics and situations, and not to the level of the hallucinations produced by AI, where things that don't make sense in the context are present. For example, making stuff up on the spot about a certain recipe we have no idea about would be unthinkable for a normal human, why even do that? Specially suggesting adding glue to pizza, outside of a joke context...
Personally what I am excited for is every single legal case defense being "Um your honor, thats AI" and needing to bog down an already bogged down system waiting for an "AI expert" (or an AI, AI detector) to validate it.
That SEEMS practical until you consider that most security footage at places like stores or banks has all the resolution of a malfunctioning vintage TV set. Then the judge would say, "This video is not high enough res to be AI."
“Co pilot for every profession” + “we will just make up an answer” makes me TERRIFIED for the future of healthcare if nurses & doctors rely ok copilot 😬
absolutely it’s scary how far it’s already getting.. just got back from EAA Airventure where they had a seminar for a digital copilot advertised to long haul SOLO pilots. They want fedex pilots to have AI helping them fly (reading out generated procedures, telling them where to go, which runway to use etc.) why not use an algorithm!!! Algorithms are more reliable and accurate, and can actually retain knowledge about your checklists and stuff. AI just makes shit up!!! planes already have such smart/high tech instruments in them you just have to learn how to use them. I just hope the people in these professions are smart enough to not use them. although a lot of times it’s up to higher-ups and it seems like they tend to like AI
Tbh I do agree that AI is being used for useless stuff, like arts and music. There is a bit of reason for video generation which is having a 'world model' for the world but I don't think it is best used in that area. There is some merit to what you guys are saying. The reason there is a lot of "AI copilot" is because humans aren't perfect just as AI isn't. But humans + AI would be better than just only AI or only human. You are overestimating algorithms, humans design those algorithms and sometimes humans forget things (miss out edge cases). Also, humans get tired of doing the same thing over and over again. I cannot overstate the humans + AI thought. Ngl though doctors are not perfect too. You're forgetting the other positive effect, places without doctors, would get free healthcare even though it wouldn't be perfect. It is better than nothing. If implemented correctly, AI can stop making shit up. Claude AI is already building towards that. Point here is, AI is new technology just like computers and algorithms before. It just started a few years ago.
But what if AI produces results which are better then doctors and nurses... it is literally worse to not use them, you know humans aren't infallible creatures..????? There have already been research where these new technologies ARE better than doctors/nurses in specific domains, and this is only going to increase as it gets better.
I had a survey from my university about how staff feel about AI implementation for our university. One question that concerned me was "how do you feel about AI being used for grading?" The issue with AI grading assignments is that students will most likely get the mindset that if the professor or TA doesn't care enough to put in the effort to look at my work to give me feedback to do better, why the hell should I put in the effort and spend thousands to do this?
@@greeny.official Exactly. I could also see this encouraging cheating as well. They really can't hold it against students that use AI to do their assignments for them if they use AI to grade the papers!
I'm uni rn close to graduating from a CS course. We have students using AI to do assessments or cheat on tests (even though they were warned they'd get a fat 0 if they did that) and professors using it to write lecture material. Like what is even the point of being there if you're not learning or teaching?
Which leads to the obvious extension of "teaching to the test" : "Writing for the algorithm". Which must be incredible writing, I bet [He said sarcastically]
@@omnipenne9101how can you be close to graduating and not already know that universities are primary centres for research and academia? Teaching is basically a necessary distraction to your lecturers... Most of them really don't want to be lecturing and tutoring you - but they are forced to by the University as part of the agreement to let them pursue their academic interest. And the learning part is 100% on the student.
This was such a good break down of everything wrong with AI. I'm so happy that you hit on the idea of training redundancy. Ever since tech bros started bragging about how AI will only get better, I've been thinking that itll probably only get worse because itll be forced to train on itself once it steals everyone elses work. Like a xerox of a xerox, itll slowly decay into nothing. An ouroboros of false information and too many fingers per hand
30:00 So fun fact, this is actually a huge problem in fandoms. Like, full on poisoning entire fandoms. I've mentioned this on Tumblr and I'll do it again here. The way fandoms work, is a bunch of people get together over something they like and start creating stuff for each other, they bond over shared interpretations of the media, sometimes they bond over different interpretations of the media, and the whole thing is one happy family (that argues over their favorite ship). When a fandom gets older, the people in it get older too, eventually you have some main people in the fandom. There are a few older people at the top that the other people look up to. Kind of how you've got your big boy UA-camrs amongst everyone else. These bigger people in the fandoms pave the way for the newer (and often younger) people to grow better at art/writing/ whatever. It's a healthy system. Like how the older generation raises the younger ones. But boy when you throw ai into the system does everything go horribly wrong. Pre-ai, if you wanted to make a piece of art, there was no one who could do it the way you wanted it, except for you. So even when you're a nervous, self-conscious kid in a fandom, you eventually tested the waters and got in. You learned new skills and made friends. If you wanted to make something, but didn't trust yourself, the only thing you could do is learn to trust yourself. Let yourself suck so that you can be good at it eventually. But when ai is there suddenly these kids have an out. You want to create something, but you don't trust yourself, you don't believe in yourself, so you have the ai do it for you. And who would blame you? There's a button that will do it for you, and you're scared of doing it yourself! So you push that bottom. You let ai do it for you, and boom! You've got a thingy! You've effectively skipped the tutorial level that teaches you to believe in yourself, and to do your work. Now you're in the big kid pond and can't do anything. If AI fails you, if someone insults you, you don't know how to bounce back and you just crumble. And on the outside, to all the big kids, you're some talentless weirdo that thinks ai is the answer. On the outside, you look like the problem. Now people are pointing and laughing at the victims of ai, who didn't know the dangers beforehand. Ai is robbing kids of skills they _need_ to learn for fandoms, and for real life. It's absolutely horrible. And it's so dangerous for these kids. So whenever I see the summary of a fanfic say something like "yeah, I used a little bit of ai, sorry! I just wanted it to be good, I did _some_ of it! So enjoy!" I die inside and think "who told you that a crappy robot was better than you? Who convinced you that the robot's horrible job was better than you could achieve? Cause they lied. With a bit of practice, you could blow this out of the water." Ai is the death of creativity, and self-esteem, we're killing Gen Alpha and the younger half of Gen Z. In more ways than just UA-cam brainrot.
We've somehow as a civilization decided that potentially all future generations of artists, composers, writers etc. need to be culled in the name of efficiency and "democratizing art". It's hard to believe that this is where we are as a species.
@@hillehai it's all getting so industrial. I can't believe I live in a time period where companies figured out how to mass produce freaking ART. CREATIVITY. Things like ai that roleplays a character are a problem too. People used to get together and each choose a character to play for the rp. But now that chat room ai exists people aren't going to each other for the fear that they'll be judged. And instead the robots. Which not only lowers the quality of the plotline, but ai feels far too fake to satisfy a social quota that keeps us mentally healthy. It's exacerbating the crippling loneliness that's breaking younger generations (well, everyone really, but kids are hit the hardest). This whole thing is creating enormous groups of people that don't know the first thing about talking to each other. (I'm talking way too much sorry)
This AI thing dominating the creative scene is why I am in a fallout with the current art community and had lost my job as a designer. Now I am considering either working in a book retail like Barnes and Noble or enter the military. Sad to see all my former classmates at art school now work in menial jobs, with one of them now working as a fry cook in a local McDonalds.
wow this is a really terrible attitude to have towards people just expressing themselves with a creative tool that isn't hurting anyone. Real "old man yells at cloud" moment here.
I have 6 years of experience in customer service and fraud prevention, but I cannot find a job because companies would rather save money with infuriatingly useless AI chatbots that end up requiring a cx agent anyways. Edit: somehow I am being called a liar and that I proved myself wrong because peopled actually don't get it that companies are DOWNSIZING THEIR WORKFORCE IN FAVOUR OF AI I genuinely didn't think I needed to explain that lmao, one company I worked for went from 100+ agents to 30 within 5 months.
@@alext3811 how will they made profit if no one uses their products? those days i was helping my mom with a problem with her bank, their customer support was all AI-based and it required a lot of effort to get to talk to a human being, but their site wasn't even available and the phone lines were shut down since december last year. she ditched them 🤷♀
@@kazu9445the theory and idea that most of the Internet is just bots making comments for other bots to interact with and that there are entire circles of the Internet comprised of non human automated engagement farming
What I don't understand is why are we as a society giving robots better treatment and possibilities than actual humans. AI has take n over art, music, video content, etc etc. And yet we are still working all day in minimum wage jobs. I don't want to work for 8+ hours every day for AI to make me preety pictures for me to stare at. I want AI to work for us so that we can enjoy the good things in life. I just don't understand.
I'll definitely quote you in the future, you perfectly nailed why Gen AI in creatives is so fucking unhelpful and harmful, it's all corporate and soulless
This is the crux of my issues with AI. Creating art is one of the biggest and most intrinsically human ways to fulfill our short time on this planet, and making a living out of it was already a path filled with hardship. Then these companies show up and just airdrop steamrollers all over this scenario without a care. I'll never understand the pro-generative AI crowd, to me it's almost as if they hate and resent artists...
Because humans are cheaper for many things. And because we don't have any economic model to support people not working to earn money. There's no actual right to food, water or shelter - only the right to buy it, if you can afford to. So if there isn't enough work available to keep everyone employed, everything goes to shit really fast. It's a trap that we are all stuck in, collectively. Governments must constantly strive to grow the economy, to create more jobs, to keep generating enough work to keep everyone employed. Doesn't matter if the job actually matters at all, that's another person saved from poverty by letting them work eight hours a day writing leaflets for a company that sells Disney-branded umbrellas.
big labels suing ai music companies bc their revenue is getting siphoned away is fucking peak irony. literally "stealing a meme vs having your meme stolen" shit
Maybe it's easy for AI to copy their music because they keep making the same thing over and over? Oh, look, it's another 4-4 pop song singing a love tribute to a nameless partner.
@@vylbird8014 I know this is sarcastic but you are unironically right. As long as most ppl produce the same things in the same way, AI will be just as good. If you get original, that's when no AI will be able to emulate you.
What really gets me is how useful I know AI can be in places such as the medical field where it can do stuff like detect cancers long before a human could, and yet all the biggest AI companies are working on the most useless and immoral areas.
@@439801RS (preface I hope this sarcasm is obvious haha) People don't have to waste time on the arts anymore thanks to our AI overlords! No more expressing your humanity ever! Now stop yapping and get back to producing surplus labor value you lazy sheep- I mean my wonderful employees!
as an artist the number of people supporting generative AI is disheartening and demotivating. i just had an argument with a guy asking me whats /really/ wrong with genAI (even tho i was like, doesnt really matter, its stealing) and i kept trying to tell him it doesnt have capacity for creativity like we humans do. he kept going to "how do you know that/cant you prove that" and "what if it could someday." i gave up. it just feels so useless and idk what we should do now
you cant argue with NPCs.. they feel proud as "producers" if some stealing machine transform for them plain text into audio file.. Its thinking pattern of such people, they are basicaly thiefs too... Maybe all of this is just test of civilisation if we can sustain spirutualy or shift to braindeads..
I refer to it as AI imagery. No most people who don't know the answer to something say "I don't know." Imagine asking a stranger on the street how do you cook a pizza, and they just make some BS thing up instead of saying "Who are you, weirdo?" Nothing AI does is "human like". Honestly, that's a real good point. If google wants to claim ownership to the videos on UA-cam, then they're liable for defamation when one of those creators goes off the rails. Its contradictory to what they've been arguing in front of Congress that they shouldn't be held accountable for the creations of those on the platform.
Actually this is not quite right. There are plenty of cases of humans with brain issues that will retroactively hallacinate and provide incorrect explanations bc they don't know and are completely unaware themselves that is what they are doing. Confabulation is the term used to refer to this phenomena of the human psyche related commonly to amnesia or dementia. That is essentially how an AI with a brain fart responds as well. Also this gal does not understand what an AI is doing. It is not just copy and pasting what it has been trained on, it is distilling what it has been trained on down to more fundamental concepts that it can then itterate upon and that is why it can accurately and consistently reproduce pluasible output. It is arguble that humans do the samething when we learn how to produce writing, images, music, etc. I honestly don't understand why so many people so desperately want to deny that ML can learn real knowledge and then apply that knowledge in new ways...but that is exactly what makes AI useful in the first place and exactly why it is being applied in the first place. To find patterns and concepts in the data that humans cannot find without the AI, and if humans knew exactly what those patterns and concepts were we would not have issues with hallicination or aligment problems in the first place. Also google does not claim ownership of the vids, but you do sign a licensing agreement with them when you agree to terms and conditions that gives them certian rights to use your content. It is legal bc you agree to it, it is also a standard agreement with all social media platforms. It does not make them liable for what you post, it gives them certain permissions to use what you post as a condition of your ability to use their platform to host your content.
@@memegazer do you not see the issue of LLMs and genAI being used in schools and shoved in the results of google searches? you wouldn't want a judge hallucinating when doing their job, but you had a judge use chatgpt, which cited cases that never happened. you wouldn't want a customer support agent to tell customers about a policy that doesn't exist, but a airlines chatbot did just that. that's the issue with ai hallucinating, it giving false information in a PROFESSIONAL setting, and in the case of the airlines chatbot it led to the airlines getting sued. it wouldnt be bad if employees and entire jobs were being replaced by said hallucinating LLMs under the guise of productivity.
Sorry to hijack this but... saying that there's NOT ENOUGHT DATA means that they will steal EVEN MORE of our data, even if they don't say it in their privacy policies. DUMBPHONES PEOPLE! RETURN TO DUMBPHONES! SEND A MESSAGE!
@@memegazer professional Alphabet glazer. Someone likes giving up their privacy for an orwellian society. Maybe if you worship harder, the AI algorithm will increase your digital social credit score.
i had an old friend who shared that there was an ai generated nsfw picture of him floating around and that it was not him and to please delete it. i felt so bad for him because thats just so scary! another thing is remember when scammers would say "i have a picture of u doing stuff and if u dont give me $1 billion i will send it to all ur contacts"? like back then, they didnt actually have the pictures.. but now think. with this ai stuff? anyone can have anything on u its so scary
Actually, I think this can actually normalise having nsfw images floating around. Like imagine if your real NSFW images got leaked. Due to AI, it's easy for you to now deny that it's a real picture. You can claim it's AI. When everyone has NSFW images of themselves floating around on the internet, nobody will care that there are NSFW images floating around on the internet
As someone who lives in Toronto, I remember the coverage of the Jennifer Pan case when it happened. Not once were there images of her smiling like a party girl. Seeing those ai-generated images for the documentary immediately made me say "that's not the same person."
In the 1950s people imagined future where robots to all the boring tasks and humans enjoy life, create art and music. Nowdays robots create art and music, while humans spend their lives doing boring tasks.
As a cybersecurity and privacy professional: the amount of people who want to focus so much on A I....but do not have the infrastructure to support a healthy or mature IT environment is huge. Hugeee.
@JimAllen-Persona Those these folks have great counsel, some of them don't understand the technology. Out of team of 10, you may have one person that understands A.I. deeper than a UA-cam video. I've seen this problem with Data Subject Access requests, for example. Someone wants their data deleted (counsel can understand that) but how do they defend not being able to pull it from an data set used by A.I? They don't know. 😂
yeah - I was actually looking to invest in cloud computing and quantum computing. I'm far, far from an expert but can tell hardly anyone is currently poised to support any of this.
Scarlett Johnson voice in OpenAI is literally "We invented the Torment Nexus from the Sci Fi novel "please don't invent the torment nexus" it surprises me how much tec CEOs all have no ability in media interpretation and see any tec stuff in movie and go "uh shiny"
@@rrbella He did say Her was his favorite movie and its funnier to imagine he was being honest but is so bad interpreting media he only likes the movie cuz "cool IA thingie" (tho he was probably just lying just because)
@@every_username_is_taken I meant the irony of Her being a movie about Joaquin Phoenix falling in love with AI Scarlett Johasson and kinda of a cautionary tale of how tecnology pulls us apart while pretending to be humane. Then the mf actually tries to get Scarlett Johanson to voice hi irl IA without noticing the immense irony
Companies branding ai working as it normally does as a "hallucination" is so funny, it would be like rolling random dice to try to solve a math problem and saying the dice are hallucinating when they dont give you the answer. These machines were built to string words together, and have no way of actually cooresponding them to reality or even understanding the strings of words theyve created in the first place
It's so obvious that these corporations are lying when they talk about "hallucinations" as if they're a bug that can be solved and not an uncontrollable and fundamental aspect of the technology.
All this talk about AI taking over the world is just a humble brag saying "oh no we were too good when we made this thing and now it's going to replace us"... sure buddy.
Deskmats like that are such a cool merch option, holy shit. I love deskmats and they're not quite my style but such a unique merch option, love the creativity
I read a comment on a different video about this topic that summarized my thoughts on "AI" very well: "Why would I want to read a book that nobody bothered to write?"
Why would you enjoy a picture no one created (a photo). And yet they're typically is far more human interactions and creativity still involved in good A.i. work. While I totally understand a lot of criticism around A.i. (particular types, not all are equal), but many appear to be riding a bandwagon of hate.
@@Fastlan3photography is created by human with tool , but its is still human that that compose everything,edit, and do many more things. They don't tell their camera to take nice pictrue.
Honestly, it's kinda cool to see how the rise of AI accidentally gave birth to another wave of "internet ugly", meaning people presenting themselves as more human again, instead of trying to archieve the corporate perfection look in all of their work. I'd love to see this being adapted by the traditional media aswell. Before any journalist or editorial person even thinks of using AI imagery for their articles, they should just open MS paint and draw something themselves to the best of their abilities, even if they never drew anything in their life. There is just so much more there. An actual human vision. An expression of someone's creativity AI can't replicate.
I’m a journalist, and whenever we get a spelling correction (which people often lump in with insults) I just take it as - well at least you know it wasn’t written by AI!
Hmm… More messy you say? That sounds like an aesthetic we can generate superficially just like everything else. I may eat my words but I do think „AI“ is the perfect tool to imitate anything put on the internet in a way that’s good enough for corporate hacks as long as these idiots are allowed access to everything that’s on the internet.
@yummyliciosable Concept art exists as a form of communication between the people with the ideas and the artists that will make the idea work. It's not just there to look cool. It's a blueprint. Using AI for inspiration is not the same as using AI for official concept art. Concept art has a specific purpose. It needs to be readable. AI, with its weird blurry details, is not ideal for a project where a bunch of different artists working on a deadline have to figure out what the weird blurry details are supposed to be.
I like the type of AI that can help doctors recognize tumours and help people in that sort of way. The AI that just does art isn't really that useful to us as a species. All the AI people call it a tool, but it's more akin to straight up outsourcing, like you said.
I’ve noticed lately a rise in AI generated crochet and knit patterns, both for the image of the product and for the pattern itself, and it’s so insidious because unless you have been crocheting or knitting for several years, you will not be able to tell the difference between real images and patterns and AI generated ones. People are getting scammed out of their money to be given a pattern that makes NO sense and cannot physically be done, and it could keep people from ever picking it up as a new skill, lest they be scammed
I find it wild that you aren't the first comment I've seen mention crochet/knitting. Not even the innocent facebook grandmas are safe from their past-time. I find it wild that the AI content farms have invaded something as niche and innocuous as Knitting/Crochet, literally any avenue they can find to make a profit these days.
@@Lemonpeppergoblin I’m in a bunch of Facebook groups for knitting and crochet and you have no idea how many times I’ve had to tell someone that the image they’re looking for a pattern for is AI. I post about once a month in those groups with a check mark list about how to tell if what you’re looking at is AI, but the newer ones to the craft still don’t really understand how to tell it apart, and I genuinely hate it for them. I’ve even got random swatches I made up to try to show them what real stitches and AI generated ones look like, but newbies eyes aren’t trained to see the difference yet so it mostly looks the same to them
With every passing year I'm more and more nostalgic for the 2000s to 2014ish internet. The current net blows and it doesn't seem like it'll get better any time soon.
I do like that the current internet is more pro-woman in the surface, but if you scratch it it’s just the same misogynist cess pit it’s always been because it’s just a mirror of society and society hasn’t really made female existence more comfortable or safer yet.
@@xp7575 The confusion probably comes from how absolutely stupid of a response it is if you read it as intended. I was born in the rural South in the 90s. We had internet genius.
@@kylegonewild obviously there were a few people with internet access in rural areas back in the day but it was rare, rednecks getting online en masse this century ruined the internet
The problem isn't rural areas it's phones. Phones should never have access to the internet. Or at least very limited access. Phones are the root cause of a lot of issues with the internet, with gaming, everything.
My friend and I entered an art contest a few months ago and she lost to very obvious ai “art”. She spent months on a self portrait oil painting that even when putting my bias aside was amazingly beautiful and definitely deserved to win. There was ai competing in the charcoal category with me too but luckily someone looked at competing pieces before the judges actually scored anything and bombarded the hosts of the competition to remove the piece and they did eventually. It’s incredibly frustrating especially considering part of the first place prize was a scholarship that my friend definitely deserves and needs, and the contest hosts were very hesitant to remove the ai “art”.
Understandably frustrating but it's also very funny to imagine an AI 'charcoal' work's sender getting a scholarship to an art school and being handed an actual piece of charcoal and suede and get told to get to work.
@@jasminv8653 honestly yeah when you forget about how awful it is the whole thing is really funny 😭😭 like what did the person submitting the “art” expect to get out of an art scholarship???
@@heatherduke7703 normally yeah, but this one took digital scans of the peices for some reason. I agree though, all physical media competitions should require submission of the physical piece
@@vonn_kate there's machines that do all of that too, but they're expensive and take up a lot of space. The engineering problem is pretty big and costly to make a machine that just "does your laundry" where as programming an AI replicate human digital output is literally a processing power problem.
Why are you still using FB? Genuine question. I quit over 10 years ago when the warning signs were already there (albeit different warning signs). It just seems like a wasteland of arguing and misinformation farms.
I recently had an actual human try and steal my art to pass off as her own. She's a scammer and the people that know about her and are warning the ones falling for it alerted me to it. I had no idea it was happening and I never thought in my wildest dreams that would happen. Now I have to worry about a no name, no face, no body having computer stealing my work just for funsies and money that I desperately need. I hate this timeline.
@novasfashionfunchannel8287 So she's this woman who had gastric bypass but lied about it and just tells people she's chronically sick for unknown reasons and is constantly getting "donations" for her "hospital bills" while also constantly buying luxury items. So she's an "artist" or at least claims to be but anytime people ask to see her work she says "just Google my name and you'll find my fb page." We have the same name which is the name of my fb art page except I have a different middle initial. She even brought it up on a live on night to show people and everyone was like, "great work but what's the R stand for?" That's how they started figuring out that it wasn't her work and found me. She's not actually an artist at all. She worked at an art bar, pretended to own it but never did, and still will just say to Google her name, knowing it brings up my art page, to her followers. I had to make a post on my art page with a bunch of pics of myself and letting everyone know that she's trying to steal credit for my work. But the people of the sub have been lovely and I appreciate them letting me know about it. It's been wild.
My final ad break for your video was legitimately for the exact Adobe AI tool you spent minutes of the video railing against, and it killed me. My husband happened to overhear just enough of the video prior, that we spent the whole ad gasping for air together. Then I gave him the context he missed, and we spent another few minutes nearly crying. Truly special.
I legit used Notion for the entire duration of my degree for every lecture's notes, every assignment, group work... until they added AI. specifically bringing up a whole dropdown window every time you press the spacebar at the start of a new paragraph, which i did constantly. literally the fastest i've ever stopped using a piece of software since there was NO WAY to disable it.
18:23 - they love this term. "publicly available" Just because something is publicly available that doesn't mean it's free to use. An artist still owns the rights to their artwork after posting it on instagram. Just like how if you park your car on the street, it's publicly available, but it's still theft if someone takes it. What that term actually means is "we could access it", not that they had the rights to use it. Don't let them bamboozle you with wooly language.
I remember the term "publicly available" being used by some pro-AI guy screeching with that term and how artist were being "entitled" just because an artist didn't want their art being used in AI image training and then complaining on how they were against "progress" in doing copyright claim or something. I don't even care how much someone would say how there's just as much bad human made art and the like, I can't really see AI art being nothing more than a fancy toy that'll be used by a business suit as a "cost saving measure" and AI generated art, no matter how good it looks to be, to be meh compared to a human made drawing.
computers are better at manipulating and processing data than restricting its use. people love to think like they're in control of everything the put on the internet, but really, people barely have any control the moment something is out there on the web. that's just fundamentally how things work, it's kinda hard to restrict computers from computing in specific ways
18:51 "publicly available" DOES NOT MEAN "publicly available to use" that's not how freaking copyright works. just because a song is on youtube doesn't mean i can put it in a commercial.
if an AI is trained on a song and creates a new song based on parts of the original that’s baaaaad, but if a person listens to the same song and is inspired by it to create a new song reminiscent of the original that’s gooooooood. make it make sense lmao
Trouble for the "AI" companies is that they'd never be able to afford the training data (particularly high-quality training data) if they actually had to pay for it. Then there's the whole issue of curating the data.
It makes me think of how I can go to a publicly available store and walk inside and see all the stuff just on display, but that doesn't mean it's free to stuff my cart and walk out without any consequence. There's so much messed up about this
One of the best takes on AI I’ve seen. I work in the political sect of the world and it’s been a huge issue with interns using AI to write memos so false information gets passed up the chain of command and many times if it’s not “consequential” it could easily be overlooked then given to the public.
17:36 UA-cam inserted an ad break right between the previous clip and your reaction, so this face was the first thing I saw after coming back from a weird candy bar ad. 💀🤣
doodling something quick like that has personality. even so, it seems a bit silly to make such a big deal over a background image. like, nobody's going to get hurt over a background image being AI
@@noxxyy It hurts your soul when you have to look at. Especially when it's everywhere. It takes a psychic toll on you when you're surrounded by garbage all the time. It's the same kind of attack on your soul as watching the same ad over and over every time you click a video.
@@noxxyy You'd be surprised. I follow a smaller content creator (about 10% the subscribers of Gabi) and they used some AI generated stuff ironically and had people saying they didn't like it.
@@whitemakesright2177 my soul seems rather fine, and i don't see ads to begin with, so... yeah, i think people are just being a bit over-the-top with all that, people should care when someone's genuinely using AI to trick people, not when they're just using it for quick and easy filler for backgrounds and placeholders and memery
Imagine a computer that just flat out lies to you when it doesn't know the answer, or even just because it feels like it at that particular moment. That'd go down great in any big company I'm sure.
Its just a matter of time when a product sold to a corporation like "improvement of the workflow" makes a huge mistake that leaves an entire corporation at risk. Just then they will be questioning the AI tools. Meanwhile tech developers will keep on selling bullsh*t stamped with AI
@@happytofu5 Not quite. Auto-complete would recommend the same word no matter how many time you erased and retyped, because as you said it's using a set of instructions and usage data to come to a conclusion. If AI worked like that, every text prompt for an image would come back with the same result. It's to simulate the "human" element, something you absolutely do not want a computer to be mimicking. Giving a moustache you never asked for to an AI-generated crab is like adding an extra zero to a quarterly earnings report because a human "might" make such a mistake, even if it wasn't asked for. :p
@@Cantimule Quite indeed, actually. "glorified auto completion" does not mean "literal auto-complete your Android phone uses." The difference is auto-complete maintains a history so prior inputs impact future outputs. The generative algorithms are black boxes where the output does not depend upon prior inputs and indeed does not consider them. The function of bots like ChatGPT is still ultimately just trying to guess what the next correct word is and output that. All that other shit you mentioned is irrelevant to the analogy they drew.
That lady said that AI is really human and that when we don't know something we make it up. So why would I ask AI something instead of a human seeing as it would be the same? Or better to ask a human. At least you can tell your friend is lying 😅
Funniest thing about this all is it enables the worst behavior from the worst people, my own experience of generative model misuse was asking dad to apologies for being a bad person (not going to provide context), he offloaded responsibility of texting me to chatgpt over the course of like 2-3 months of discussion, i thought he finally did some thinking and gave an ok apology, and I was ready to let him back into my life after 5 years, only for me to realize each message was written by chatgpt after 5 minuets of thought and inputting some word salad (which he proudly provided after I called him out on it).
Posted this on Drew's video, too, but I really appreciate creators with platforms such as yours speaking out on this. A lot of us in the art community have been really deflated by this whole thing, feeling like no one cares the threat this poses to us--and the world at large, really--so it's really meaningful to see folks like you guys getting our backs on this.
I’m a tutor at my uni’s writing center, and I had a student say, to my face, that he wishes there was an “AI room” in the library where students could go and get their essays written and/or edited and/or “peer reviewed” by AI. Guess I’m chopped liver, then. Thanks for wishing that AI steals my job.
I wanted parts of this long time ago. We could have had all the lectures on video for the last few decades, on disks. Students watching the videos, and maybe having some software teaching and testing them. Official tests would be under human supervision, of course.
The powered weaving loom took the job of thousands of people weaving by hand, and that's a good thing. Now that weavers aren't needed, we can all get jobs making Playstations and stuff. We wouldn't have had the labour spare to do that, if we still needed to make everything by hand. Physical jobs are automated out of existence, and that's good cos they're usually crap and unrewarding jobs, and society becomes richer because things get cheaper. If an AI can do 90% of the job of a manuscript editor, why not? I don't imagine it's fun work. You can do something else instead. Sure it sucks for you in the short term, but on an historical scale, it's infinitely better for humanity to have progress. It'll likely make it cheaper to be a professional author, so perhaps more people will be able to do that, using their language skills creatively instead of de-mangling bad grammar all day. It's not jobs that people need, it's money. An ideal world would have machines do all the work, and money shared out fairly, so we can all chill out and do whatever we like.
I'm an esl "tutor" rn (which on the site I use, 90% of the time its just talking to people), but it's been slowly forcing ai on us more and more. they started generating ai summaries of each class and they now have a weird ai chatbox for students to use, but it's insanely scary because you KNOW they ARE secretly using lessons to train ai-- either selling to third parties, OR using it to replace actual human tutors in the future. it's definitely not legal either way and honestly morally fucked up, it's just sad because we're about to see a LOT of professions die. WHICH I know is a thing that just happens when there's progress. difference is, I don't like that it's replacing things tutors or storyboard artists 😭
@@greenaum how is manuscript editing a physical job? Also, if you’ve ever been irritated by a word processor or your phone autocorrecting something that was already correct, you’ve experienced why this is not a good application for AI. Language is deeply complex, grammar rules can and should be broken at times for artistic effect, and humans are vastly better at explaining to other humans what could make their writing better.
@@greenaum Problem is, this is a really idealized future that i frankly, can't see happening even generations forward. Yeah I'm all for my menial work all be taken care of ai but as of now, ai is not used like that. Ai is being used as this piece of tech utilized by big out of touch companies to replace the work of humans, even though it'll completely compromise their end result, and making everything worse in the process, all for just saving a few bucks (late stage capitalism amirite). Hell, ai of today is not really "inteligent", it doesn't have a mind of its own, ai is just buzzword for marketing, this version of ai is a glorified algorithm. Even then the ideal world with ai we all dreamed of might not even use this technology/algorithm, but now we must deal with the sheer shittyness of ai that's fueled with corporate greed. and if I'm being honest the trajectory of our future is more distopian than utopian, it's not really enstilling confidence of the future when the work that's ai is currently replacing is actually the creative ones and not the boring menial work that's it's supposed to.
Several years ago, I naively assumed that large language models would be trained from sources like academic journals and text books and other things that are generally understood to be truthful. But that would just be very illegal! It wouldn't be any more illegal than scraping countless youtube videos and reddit posts for the data, but academic journals and publishing houses actually have the legal representation to defend the work they publish from being blatantly stolen!
Fun Fact! They actually *did* train the original generations of machine learning systems on open source literary and academic sources! And then they ran into the same problem A.I. Bros whine about now - it "wasn't enough data". So companies started buying data from companies like facebook, and stealing art by scraping the Internet and pretending the aggregate mix had no ethical concerns!
And even if it was trained solely on academic resources, it would still churn out false information because it doesn’t have an understanding of truth or being factual. It’s only trying to put words together in a pattern that is similar enough to what it’s been trained on to pass as human writing.
I still can't believe that Will Smith eating spaghetti is a benchmark for AI video generation. The Turing Test of the future is going to involve a virtual dinner with Will Smith, and spaghetti is on the menu.
It drives me nuts that Google used to give me the exact answer I was looking for at the top - and now it gives me an ai response that answers my question in the absolute wrong way. Congrats on your release - they look beautiful next year when I'm a nurse I will be able to have one myself :)
For some really weird reason I don't have that feature on all the devices that I use Google on. Maybe it's because I haven't updated it yet but now I'm certainly not going to update it now
@eclipsemoon01 I looked up how to get rid of it - and you can't. So don't update anything !!! Its so bizarre that Google AI is worse than Google is.... I guess it's just too much info idk
"maybe the problem isn't the lack of data" damn right there's literally research PROVING that, even, higher quality data (instead of "literally everything") results in WAY higher efficiency, you waste WAY less resources on training, and get very good results at a FRACTION of the data... the problem is: EFFORT, lol, the companies do not want to put in the EFFORT to make actual high-quality data, they want FREE data without bothering to pay people there are actual ethical datasets, though, where the data is explicitly agreed on, but y'know...why waste money and time, when you can just steal data?
It's okay, we can just have AI decide what's good data and lousy data, before we feed it to the training AI. (In case it's not obvious - that's not a serious suggestion)
So this is a symptom of the multi-billion dollar gigacorporations skimping on the singular thing that makes or breaks this tech? Thus leading their output to get consistently shittier as it keeps getting inbred generation after generation? Amazing. A story as old as time is the richest bastard on earth saving money on one thing he should really spend as much money on as possible, and self-destructing as a result. Can't wait for this bubble to pop and every rich imbecile involved to NOT learn their lesson because once again, no, the most intelligent people on earth do not get the most money. Please, we've learned this lesson 1288567 times, it's time to actually learn from it
All of the points you make in this video are exactly why I’m going to be majoring in Computer Science and AI. I want to be able to learn and create a better use for mainstream AI that isn’t generative. It’s heartbreaking to see how something I’ve been wanting to do for a large part of my life change into a product that (almost) everyone hates. So far, I’m not quite sure how to change the industry, but I have years to plan and figure it out. So please do not lose hope for the future of ai. There are so many students (me!) and people that are working to improve the situation
I spent 15 years working my way up from homless to director at an FI adjacent company. Then they got a new ceo who decided the company would no longer sell products and instead use their industry connections to become a data broker and sell decades worth of data they'd collected on customers. Then they laid off most of us and now I'm unemployed and close to becoming homeless again. Scraping and selling data on people is pure evil in so many ways.
It's an unfortunate natural conclusion to advertisement, marketing and capitalism in general. People are a data point to be manipulated to keep the money flowing.
AI has made most search features completely useless. When they started pushing SEO content I thought it was bad, now with AI it’s a hellscape. It’s so so frustrating for trying to find cooking recipes, and even worse for trying to find academic research articles. I feel like I’m constantly having to fight the algorithms to get to what I actually want, the thing that I *specificaly* asked for. 🥲😒
What's funny is to most people, when you watched terminator or jurassic park, you'd walk away thinking "man, that would suck" but it seems like all the people pushing ai and genetic cloning watched those movies and thought "that's literally the world i want to live in"
Yeah, the Torment Nexus meme strikes again... Malcolm's quote about "scientist preoccupied with whether or not they could without thinking if they should" feels very relevant nowadays.
@@Adagamante i swear i saw a video years back about a company that makes clone dogs with glow in the dark toenails, and their end goal was to eventually revive the mammoth species and reintroduce it to the wild lol. As if that wouldn't be the most catastrophic thing to introduce to an ecosystem
@@Countdownsmiles what a reasonable goal, right? As if the current day megafauna isn't endangered across a myriad of species... Focus on cloning those first, you buffoons!
It's almost as if, we learned from the past when technophobes warned us against "playing God", and we went ahead anyway, and developed life-saving medical technology, and world communications networks, and international transportation.
@@AlexReynard Nobody said anything about medicine bro. We're talking about bringing animals from 200 million years ago back to life lol. Learning from the past is not something i would credit our world leaders for lol
I'd say the thing that sets photography and blender type things is that, you take the photo, unlike ai. And the built in models for blender is made for you to use, not stolen from others. It can be changed by you. It can be created by you
Hey gabi! About Spotify's curated "mood music" playlists: The songs there are put together by just a handful of ghost producers under a special contract which pretty much guarantees them a spot on the playlist in exchange for a much lower (about half) royalty payment per stream. Spotify heavily promotes these playlists because if people listen to this music instead of literally anything else, they don't have to pay out as much in royalties. And I guess they prefer it to look like there are diverse artists from all over the world contributing to these playlists rather than just a few dudes in Sweden, so each one of them are pretending to be hundreds of artists. The reason it all sounds so similar is 1. readily available sample loop packs and 2. because they want it to be like that I guess??? (This info is a couple of months old, so maybe some of these ghosts producers have started using AI generated music as a base by now)
I love that at the beginning you tried to use the AI "agent" like a search engine by entering search terms, the exact thing people have stopped doing in search engines, entering questions instead. And now the AI actually REQUIRES questions to be able to work at all.
not really, the AI chatbots will still lock on keywords. You don't need to create a sentence for them to work. See for example the prompts of text or image generators, it's just a long string of words separated by commas
@@Call-me-Al from my personal experience, yeah! I've been finding more and more luck finding results when I ask... which is something that I was actually told NOT to do when I was young, and rather use keywords. Silly.
@@1BergerVongSchlauigkeitHer I blame voice assistants on phones for it. As soon as people could start searching for things by talking to their phone, they seemed to treat it like a person and ask questions instead of just treating it like a search engine.
Most people won't listen to me because they are so ubiquitous today (especially in America), but the same thing happened with cars. Look up the "Motor vehicle fatality rate in U.S. by year" on Wikipedia, and see the solid red graph. Now, imagine a similar "technological advancement" being rammed through by people that stand to make life changing amounts of money from it, faster than the public is able to really consider the repercussions and come up with ways to make the tech more safe (but unfortunately slightly less convenient/fun, like speed governors for the example of cars), that are then able to sweep a curve like that under the rug of operator error and after long enough, "oh but can you imagine life without it?!" as they actively worked behind the scenes to dismantle all the ways we previously lived without it. Now, death is an extreme example, but it's also the most likely to get something done about it and *that* didn't even work in that example. In fact, most recently, the numbers are steadily ticking back up after a long period of decline... but I'll save that discussion for another time and get off my soap box for now.
I’m desperately hoping it can either be heavily regulated with laws that protect stolen copyright, environmental impact, or transparency. AI users should have to be required transparently label things as AI the same way UA-camrs have to clearly state that they’re being sponsored or how A&W got in trouble for putting “aged vanilla” on their cans when they don’t use anything of the sort in their drinks. That, or people find better ways to outsmart AI like that program that makes your art impossible for AI to decipher.
@@Shoulderpads-mcgeeI think AI is very quickly going to run into the problem that it’s burning a ton of money for something that’s just not profitable so they’re going to basically speedrun the enshittification that’s plagued the other major tech platforms. Like, ramping prices way up, degrading models to be cheaper, firing off tons of staff and demanding the remainder replace their workload. While watching the AI hype men squirm will probably be funny, it’s going to be rough for the average employee of these companies.
@@bree77749 Cars weren't a technology "rammed through." A working EV had already been built before the 20th century. Model T (the first real mass market car) didn't enter production until 1908. It took decades of lobbying and shady business practices to make cars a necessity for so much of American life and it's not like there was no public pushback lol. You can still legally put your horse and buggy on the road right now. Nothing stopping you. Why would these car companies not have lobbied against this? Forcing the Amish to integrate or walk everywhere is a small but untapped market for revenue. I lived down the road from an Amish family for 20 years. Just gotta have safety components like reflectors, something you'd legally be required to have on your bicycle if you went on the highway. It's no secret older cars were more dangerous. And the unfortunate silver bullet to all the arguments against such a car centric society is that enough people just...like their car or rely on it too much to survive now. Most of our issues with cars are not cars themselves but the infrastructure we built for them. The likelihood of dying in any random low-mid speed car accident in a modern vehicle is actually quite low. There's just so many cars on the roads every day that it's a numbers game.
I think the standard is to say I don't know when you don't know something. If you are making up an answer rather than saying I don't know then that is called a lie not an hallucination. I wonder if the CTO of Open AI is projecting?
22:10 the funniest part of companies pretending that they own whatever is posted in their platforms is that this implies they're the proud owners of a whole lot of highly illegal material like you know what involving children that gets posted on their platform
As an artist and privacy advocate i stopped fighting for ai laws when i brought this issue up like 2 years ago. When creeps were using it for this exact reason , and people were still defending it . like it was no big deal or a joke. Online users will still take it as a joke or meme until it happen to their sister or daughter. 10:04
@@seanbell69 from what ive experienced, people usually wake up once the issue reaches their front door, when they're no longer a spectator but another person having to deal with it. for those who still don't care about the repercussions , what can i do? I cannot change a persons deep rooted morals, it is up to them to develop some form of empathy at that point. what i had done and will keep doing is reporting instances of people using it for csam and targeted deep faked revenge etc, but when it comes to speaking to people about the Encompassing side effects, i lost the energy
I think the issue is that people don't understand there are existing defamation laws if something fake is presented as real for one, so it is not like there is no legal recourse that exists at all, bc there is. The issue however that many don't like is you really don't have any right to have your likeness not ever being used by somebody else without your consent or approval. Consent was never the legal issue in such cases. And a lot of freeom of speech exists expressly to give people a right to say or depict things regardless if others approve or consent. Take a case like Flynt v Farwell where courts decided that you have certain rights to use a person's likeness without their consent provided it cannot be reasonably argued that it is defamation or libel bc it is being presented as factual or real. I think the difficult thing would be writing laws that do not overly infringe on freedom of speech, even if there are valid cases to be made about non-consensual use of likeness. It is not that it is a joke, it is that regulations with good intent to target valid objectional things, could wind up being vague and wind up shutting down legitmate forms of freedom of speech. So it is kinda like how you can't really outlaw photoshop, even though you can use that tool to remove one persons head and paste it onto another persons body. The ability to do that might give creeps an oprotunity to make images others do not consent to, but the consent is not issue that one should be concerned about in terms of legality imo.
@@memegazer You do have a right to your likeness. It's commonly referred to as a "right of portrait". In the US this is limited to your likeness being used to advertise a product, but in many countries a broader version of this right exists. Here in the Netherlands the right to portrait applies by default, with exceptions being made for example for the purposes of journalism. That is pretty much the same in all of Europe. Americans are often unaware of this, but in much of the world you cannot just film someone without consent, even in public. Also, regarding this specific issue: there are many other laws where a distinction is made when something is "for purposes of sexual gratification". That could easily be used for this issue as well.
Not to mention the fact that ANYTHING you look up online has ten thousand AI responses up front. Including reference or scientific images that people use for IDing things like animals and plants. Not only is it irritating, but it can be downright DANGEROUS for those that trust those resources for accurate information.
100% and as an artist, makes my job so harder, and makes me waste a shit load of time that would be spent Learning styles, and how, say a a person might look under water. Or just anything to do with the ocean and like 90% WERE FUCKING AI UGh
Yeah, trying to find legitimate historical art or pictures to figure out costume or architectural or prop details is starting to become a complete crapshoot.
@@animatrix1490 i can only imagine! I'm an artist myself, so I tend to Google a bunch of random things for reference material. I can't even begin to think about the kind of horrors you'll find when trying to do homework or do replica pieces. Sounds downright AWFUL
@@EranRaye I was working on a fantasy comic from late 2020 to late 2022 and I wanted to base the visuals on the Republic of Georgia's historical dress, architecture, and culture (ended up with some Turkey, Azerbaijan, etc. in there). During that time, I was able to find quite a bit of material through Google images--some of it also led me to some very cool articles and Reddit threads. This year in like May and June I wanted to research some historical dress, hairstyles, country manors, etc., from late 1700s or early 1800s France and it was almost impossible to find stuff. I had to narrow and narrow my search terms and ended up finding out it was way easier to just not even bother with the Images tab. Finding articles from actual academic websites and archives (which would often include images with some actual citations or sources) was so much more effective than trying to dig through and differentiate real stuff from all the slop and garbage in the Google image results. *Wikipedia* gave me better material than Google. It really didn't take very long for it to get VERY bad.
good ai: they accidentally made an ai that could recognize cancer cells (was supposed to distinguish different kinds of donuts from each other or something) bad ai: type words and out pops an image (is trying to replace creatives)
In the 1800s, painters made arguments against cameras, saying that they were trying to replace creatives. 'Photography is only for people too lazy to learn how to paint'. Did creativity go extinct with cameras, now that people had increased access to reference images for their art?
@@AlexReynardexcept with cameras you still have to put in effort to get a good angle, lighting, and subject to photograph. you don’t have to do anything with ai except write a sentence and wait for it to spit a picture back at you.
@@seanbell69 What do you mean?, The big people behind AI is also Rich people making more money than us, they're not trying to better the world, they're just trying to replace as many people they can with AI, EXCEPT the ones at the very top, just so they have more money to line in their pockets.
@@AlexReynardbut no one was entering photographs into painting competitions and tricking people into thinking the photo was anything besides a photo. It also wasn’t trying to replace artists. Maybe we didn’t need painters for portraits or for illustrating war any more because photography is a better historical record but it wasn’t being used to replace painting or drawing as an artistic expression. Meanwhile ai seeks to take jobs away from artists like the animated show on Netflix that used ai backgrounds rather than paying an artist to do their job. The camera also wasn’t built by stealing paintings the way ai is built on stealing people’s art.
i love how they created one big plagiarism machine.. and then instantly were shocked that all it can do is just recreate all the stuff it stole just... worse
There was definitely the idea in mind that humans create art based on things we've seen, so why can't an AI? But human creativity is taking the things we've seen and making them different somehow. AI can't do that, it only knows how to mimick. You can't code creativity.
@@fresch4395 Yeah, a prompt invented by a human. Tell an AI to draw something that doesn't exist. It'll just give you an amalgamation of other human-made artwork depicting things which don't exist. AI doesn't have the capability to imagine.
@@fresch4395 it doesn't though, if you think that you have been fooled when it comes to LLM's its just really good predictive text its the same with music and images i have been using it every day for the last month in my software engineering job and man does it tell people very confidently to do very dangerous, insecure things and its because its been taught off a bunch of terrible answers from stackover flow... absolutely no checks on whether they are good answers or not
@@fresch4395if the AI didnt have 1000’s of _human made_ data points to draw from it would not be able to create anything. Unlike humans, it doesnt take the time to look at various artworks and perceive them through its own life experiences, preferences and abilities. It has none of those. It can create, but it is by no means creative.
As an artist in feature animation, I'm more worried that AI art will be "enough" for most people because they don't know it could be better than if an artist worked on it. If you don't know what is possible, you can't compare and you will be impressed by what AI gives you even though it is, by definition, just a 3-dimensional database giving you results (or pixel colour) by sheer statistics. Even if you have the best works in the dataset, it will just return the top of the bell curve, which, although it can be shifted for slightly better results in the algorithm, is still worse than the best work in the dataset and always, by definition, mediocre. Even an all-knowing AI will never have lived one perspective or one life, so it will never be able to tell stories or tales to humans. If you have "lived" multiple paths or multiple lives with all perspectives of the internet, books, etc., do you have a perspective at all? So, if most people are ok with that, I'm more sad that I will not be able to consume good art than I fear of losing my job.
Which is why I'm glad I've started commissioning artists to do background and character art for my stories. It certainly made me appreciate the effort and skill going into each picture.
I've been hiring artists for a game recently and at least from my perspective AI is just not a viable replacement. Specifically if we talk animation my bet is that after hype is done it will do half the work in realistic 3d animation (and maybe 20% stylized or 2d), but it will do the boring stuff. Retargering clips, cleaning mocap, rigging, decent but absolutely not shippable first pass on text-to-face, etc. It will be one more tool that sometimes can really deliver, sometimes not, but one of the 100 tools needed to do sth right. In my case it provides better translations than we got from humans (yes, humans could do it even better, but our budget was enough for very cheap rushed translation and AI does mediocre for nearly free and if you get native speaker testers to go over it, result is passable for our quality target). As for mediocre slop, the blackpill is - it's not the fault of AI, or even market economy, but its been unleashed by the customers themselves (or maybe, blind "consumers"). If ppl stop playing or watching slop spigot will be closed down and more budget goes to quality productions. There was not 1 year in this century without quality games/videos, we can collectively keep buying that and paying the ppl making it. Unfortunately as a % the "consumer" spending on slop is increasing, tolerance of "consumers" for shi%ty practices is increasing, this is why it goes the wrong way.
@@Peteruspl Yes, I agree completely. That's exactly my fear that consumers will be okay with mediocre quality. The highest-grossing animated film is still the live-action Lion King, followed by the Inside Out sequel and Frozen 2, so the financial incentive will be pushing sequels and live-action remakes. I see many people unhappy about sequels and live-action remakes (at least in my bubble), but the box office doesn't reflect that, so I'd imagine it will continue... I fear the same could happen with AI art, that it will just be enough for most people. I have been writing AI tools for years for my own workflow but I find generative AI generally useless for my workflow (except for looking for inspiration). The AI rigging tools also seem to have an actual benefit to animators in terms of performance and quality of deformations but that's also not considered generative AI. Translation/Transcription seems to be a great benefit to generative AI, though. I have also seen demos where lip-syncs are adjusted by the language which could be quite cool (especially for deaf people - if a movie is well animated deaf people can actually lip-read - though it depends on the stylisation) I only fear of a future where generative AI will be used as final or just quickly comped or adjusted to make it slightly better for a final product.
This reminds me of a video about it that discussed 1984 for like the quadrillionth time. This time about "newspeak" and how it can be used to limit worldview. Quite frankly. The idea that people in the future won't be able to conceive anything better than AI is a terrifying thought.
Anyone else feel like search engines used to be able to give way more accurate results with way more vague information compared to today? I feel like i could search a picture or meme with 3 to 4 words and nowadays i have to write an entire sentence and as it gets more descriptive the results just vanish completely.
Yep. And that's obvious: A lot of the same information is thrown on the pile in the course of 4 years, so more shit to sift through. And then. Companies just take bribes (SEO fees/advertisement dollars) to prioritize that info source, without validation.
I used to use Google image search in my job as a language teacher. But when I try and search something nowadays, the results are pretty random at times and might include raunchy images for stuff that is not raunchy by default.
Google is useless now. It ignores your keywords (or sometimes even surreptitiously replaces them with more lucrative ones in order to direct you to shopping sites), and the web is so thoroughly polluted with human- and AI-generated slop that exists for no other reason than to game the AdSense program.
I absolutely love Yuumei's art, she keep getting harassed about her art being ai despite her art blogs existing for forever AND despite how she uploads a video of her making just about every piece
Yes her art isn’t AI generated, but she has proven to be using AI art generators in her free time. People aren’t mad at her because her art is AI generated, which it isn’t, they are mad at her for blatantly supporting the use of AI art despite advocating for environmental protection. She equates the energy used for generating art as the same as buying supplies and drawing on a computer. As an artist, I find her takes to be incredibly hypocritical.
Not even showing speedpaints is enough to prove your art isn't ai anymore sadly - I've seen stuff showing off a new ai tool that can create a fake speedpaint of a piece of art you feed it
For one of my previous full time jobs as a graphic designer, I made logos for their multiple websites and advertisements, stuff like that. I also took product photography. Then little by little my boss was like “this isn’t going fast enough, let’s use this ai program to filter out all of the backgrounds of these photos” which ended up having pieces of the product cut off, random sections having blurred edges-it looked awful. And then they were like “let’s use this ai to make the logos instead” and “let’s use this handy new ai program to write our content pages and product instructions.” It was disgusting to be a part of
One of my favorite recent instances of AI messing up was on a podcast where they asked ChatGPT "what time is it?", and instead of the AI explaining that it doesn't have the resources to give a time accurate to their location, the AI just said a random time. And the time it gave didn't even have the correct minute.
Happened in Brazil: Volkswagen used AI to make a commercial using the image of a dead singer. Even though the daughter gave the "ok" for the image usage, the fans argue that she would NEVER do a commercial for a company she does not share her values and vision with. Imagine having your image used after your death for something you disagree in life?
It's you repeatedly walking in on screen like a glitch but continuing from where you left off verbally so smoothly lmfao was messing with my brain for a second. Honestly this AI shit is just so ridiculous, and sad for legitimate artist and striving artist. Thank you for sharing your pov ❤
One other argument I've seen brought up is that all of these companies are also using data from dead people, who by definition could not agree to these updated terms of service and legally cannot have their content being used in training models (and I imagine this also works for anyone who's abandoned a platform w/o deleting all their stuff, ie most deviantArt accounts)
Tbf (as an illustrator myself), humans have been trying to dramatically underpay and undervalue (or not pay) artists for decades. This unfortunately just feels like the end point for humans wanting all the things for as cheap as possible. Not exclusive to, but definitely including art and music. It sucks.
You sound like conservatives yelling about how unskilled migrants are taking your jobs. If a machine can make art better than you, maybe the problem isn't the machine.
My graduation thesis was about Gerative/Generative A.I in mid 2023, we can't forget that some of the initial model training for a certain company was made using very very cheap labor from other countries and was prejudicial to the worker's mental health since they were exposed to very disturbing prompts during model training, because humans had to flag and filter prompts/images/results as "ok" or as not allowed/nsfw/breaking TOU...
@@Unrayorojo Times magazine's arcticles " 150 African Workers for ChatGPT, TikTok and Facebook Vote to Unionize at Landmark Nairobi Meeting" Perrigo, 2023 and "OpenAI Used Kenyan Workers on Less Than $2 Per Hour to Make ChatGPT Less Toxic" Perrigo, 2023. These may give you a hint XD (can't post links since comments usually dont allow it!)
@@moongirl8807 At the time of writing my thesis we didn't see that point in specific, but i wouldn't doubt. One topic my thesis partner and I came across was the sexualization of people using those "make A.I portraits with your pictures" which i'm sure most people used at some point in 2022-2023, some which had a lot of reports of doing that to women, and most of the arcticles i've found were about asian people, either the models made sexualized/fetishized images or got their nationality wrong, like accentuating east asian features on southeast asian people, and so on...
@@moongirl8807 About what i've just posted: technologyreview's arcticle "Lensa Ai undressed me without consent" and also analyticsinsight arcticle "Lensas fetish for asian women shows sexist standards", and the last arcticle is catntran at medium, titled "Does lensa ai thinks all asian people look the same"
I wholeheartedly agree, as someone in cs I often see people humanizing AI or getting really wrong infos, but this video is just right. Corporate greed is always bad, there's no denying that, but I also think that AI is just the "new big thing" that will fizzle out of the spotlight soon enough, see cryptos, nfts, and so on, because like you said, AI has been around for a while now, it's nothing new
thanks for going over all the grey areas involved in distinguishing "generative" AI "art" from using AI as a tool for the artistic process and other tedious processes.
It really is depressing that whenever I want to search an image it has come to this, I tried to find some references for practicing on drawing anime and like it's one thing if it was even decent, but I saw like loads of just bad AI anime, there was one I was search "anime with braids" and it was so revolting to see the girl's hair had seeped into her body creating hair for hands, it was super scary
generally, for drawing references for anime, you want to look at image repositories that let you add your own tags for filtering (pixiv, safebooru etc) - most of them are ahead of the game and let you blacklist the "ai-generated" tag, which can help a lot in finding actually useful stuff.
@@minhuang8848 okay but let’s just take your point about color palettes and fashion what happens when a company like H&M or American Eagle decides that using AI is easier and cheaper to make their new lines with AI rather than paying fashion designers to do so it’s not really a problem on a small scale but the problem lies with companies trying to save a buck off the backs of people who have already done the work rather than crediting and paying those people
@minhuang8848 Booo, 'let's have machines do all the artwork and games' Real AI actually can help, gen AI only steals and is done for a way to cut jobs and costs, as well as use too much energy to damage the environment. Watch the video.
The worst part of Windows Recall is that they are basically forcing everyone to upgrade to it. They plan to stop all support services for windows 10 in october of next year, which makes using it dangerous and makes you vulnerable to online threats.
Right! I own a windows 10 pc that my mom had gotten from her job for us to use at home. She doesn’t use the pc at all so I have it. Mind you we can’t afford to buy a whole new one, like it took ages js to get this one, and now I keep getting notifications like “oh upgrade ur pc cause it’ll stop working soon” like… oh you wanna make MY pc stop working even tho it’s already ass? Okay, then YOU buy me a new one cause wtf 💀
@@albert2006xpIt is for now, but ppl have been able to get it to work on virtual machines without the processor, so that seemed to be more of a marketing restriction than a hardware limitation. Knowing microsoft, theyll just push it to all possible computers eventually with little to no notice.
As a future game dev if I have to see one more AI person talking about how they are going to make art or a whole video game with AI I'm going to jump off a cliff. Its supposed to be about the journey not the end result. I've been learning art with blender and coding for unity from scratch to build quality games it's insane
I guess you can delete your whole project and start a new one without any remorse and still ask people to pay for the fact that you WHERE delevoping a game, yes? After all, who cares about the end results, its only about developer's journey, not the end product for people to enjoy...
Congrats on learning art! You could probably also reach out to newer artists who are looking to practice new skills to collaborate or if you have any budget commission
my desk mats are out for a limited time until september 4th, 2024! gabibelle.shop/
I got one! Super cute! I drink my coffee standing 2 ft away because I’m so scared I’ll spill on it.
I disagree that AI is hurting "creative jobs" as much as you claim. Not everything that an artist makes is creative or even artistic, if a company could replace you with AI generated content without any repercussion, is because what they were using before (real people's work) was so bland and devoid of creativity that no one can notice it anyway. Jingles were mostly forgettable but once in a while there were good ones, AI is essentially very similar in that regard, sometimes it comes up with good stuff in a sea of mediocre uninspired stuff.
In music this isn't AI's fault either, this is been going on since the 90's when they found out that autotuning a pop singer was much less hassle and more lucrative than signing a rock band that could create the next stairway to heaven or bohemian rhapsody. Good music is being produced a lot, but you dont hear it because mainstream is flooded with low effort urban music, so only niche listeners get to enjoy the newer bohemian rhapsodies that come from artists that could be making money elswhere but sacrifice in popularity to create what their masterpieces.
My point being: If at some point our society needed a lot of "creative" professionals it was not because they wanted to create art, it's because they had skills to embelish and make stuff more appealing. That is not creativity, that's just valoring technical prowess, paying someone because they're more skilled at drawing, painting anf that can execute your vision better than a regular person that isn't trained in that. If an AI can do your taxes better than an accountant, would you claim AI is killing creative accountability ?
this is your first video I have seen (maybe recommended via Drew Gooden?) but excellent, amazing, I cannot compliment you enough. Very impressed and delighted on every front. You have a new sub! Looking forward to future (and past) videos!
This is what YT is supposed to be! Keep up the phenomenal work!
Everything wrong with youtube: people buy a microphone and suddenly think they're a quirky online personality and use all lower case letters
You know the "AI now" with Willsmith is just Will Smith joking on the subject, it's actually him on video, not AI production. There are many points you push that are deeply ill informed. While I share a lot of your anxiety relative to the impact of AI it's important to be accurate and to portray the real issues without going into full panic mode. You are not informing people better than the ultra positive tech bros, you are just sharing as much disinformation as they do. What you are doing is called in psychology catastrophization, it's the unconscious act of taking the worse interpretations of events (or comments) and assuming that they are the reality. Reality tend to be way more nuanced. You are not serving your own objectives by doing this because anyone who knows anything of the subject will dismiss you out of hand (rightfully) as someone who don't know the subject. Creating a disproportionate panic only muddies the water and pushes the discourse to extreme irrational reactions not based on facts but feelings and end up having unforseen and unwanted consequences.
I saw someone say that they want AI to do basic mindless chores or work so they have time to do art, and not AI doing art so that they have time to do chores or work. I truly agree with this.
And doing art poorly since an AI will eventually be better.
if they're training AI to do basic mindless chores, that will also have negative consequences.
If it gets better, people will lose jobs. They already have been. Trust me, cleaning is not an understaffed job. And if they're good enough to clean or work, they can literally make art as well??? Much like now.
No one cared about computers taking the jobs of good singers and instrumentalists. Spent years learning how to sing or play an instrument? Sorry, someone that spent no time learning that used autotune or pasted instruments in a program has your job instead. And no one cared. Or AI potentially taking away truckers jobs with self driving trucks, and no one cares. People should just admit they only care about art being in danger because it's what they've attempted a career in and don't care about others careers.
People keep repeating that same thing over and over again.
That "It's not like the world will end" quote 😂 😭 💀💀💀
Protip: When searching for generic images most search engines allow you to put "before:(insert a year here)" if you want only images before a year so you can put "before:2022" if you want to filter out most ai images.
Yus
Thank you!
Imagine, in 2050 we're still searching for *before: 2022.* That's such a sad thought.
"Yeah all the people in the pics are wearing super outdated outfits. But they all have 5 fingers, so it's just more efficient."
Quite literally, The Before Times
@@Strogman25 more like “Yeah, their outfits are outdated… but hey, at least they’re real!” 😂😅
So UA-cam lets open ai use hundreds of videos to produce generated schlock without the creators consent and calls it "fair use," but when a small content creator uses a few clips of a Disney film in their analysis/video essay, its a copyright violation. What a world we live in.
Campanies have more rights than people obviously
When it comes to business, particularly businesses like UA-cam, if you are able to do something for free that means you are the product. Including uploading. I remember the early days of Facebook, one of the terms of the user agreement was (and still is I think) that anything you post on there is actually their property and they can do whatever they want with it, including using it in ads. Places like UA-cam are no different. They will protect the Disney's of the world because Disney owns their IP and they are brutal about protecting IP, whereas the average UA-cam user is on their own because UA-cam actually considers themselves the owners of everything on here.
I am pretty sure the youtube terms of service say they can do something like that
@@Bandikit campanies have more rights than people after all
Keep in mind why would Google call out all these other companies for their data mining when they are quietly doing the same exact thing for their own AI products
Am I the only one who really really misses the old internet, like I already was missing the old internet where your real name wasn't attached to everything, and they were like specialty forums of people who shared your interests.. But now we can't even Google stuff properly
You're not the only one. I hate the enshittification of everything.
I don't miss it because it still exists ! Check out the whole "small web" or "indie web" scene or even neocities. Making a personal website is super fun a useful and it's making a real comeback!
@@jessicamoore8903 best comment ive read tdy ngl
I probably am not part of that old internet you're talking about but I still remember when you didn't need to create an account to get anything from the internet. If I wanna scroll in a photo gallery I need to make an account, if I wanna make a trashy beat I need an account, if I want a free cache cleaner app I need an account. Why does an online game need to know my government name...
I'm using the internet since about 2004, everything was so interesting back then. Now every online device is just a terminal to sell you things, including your watch and car.
My mom used to transcribe calls for companies as work, but then her work implemented AI to transcribe the calls, leading to work calls having sentences in them such as: “Thanks for watching!” “This venue has the best beat” and putting in random websites that DONT EXIST into the calls, when the caller said nothing like that.
iconic, ai undermining intrusive corporate recordkeeping by making shit up. even the computet doesnt want The Man listening to your calls
My mom edits translated books and 90% of them are translated by AI and make no sense. They'll change the name of a character halfway through the story, mix up character's backstories, etc so she now basically has to not only check for spelling errors but also continuity errors, which isn't in the job description since it's assumed that by the time they're translating an already published book, someone else caught those errors.
They really need to improve AI comprehension 💀💀
@@asmodeus8053I’d love to have a job like that! Having to keep track of the continuity is too much, though.
@@asmodeus8053that's wild it really sounds like ai made your mom's job harder and probably more time consuming.
@@asmodeus8053 or just stop using it
AI companies: "everything should be replaced by auto-generated slop"
also AI companies: "we're having trouble finding good training data, it seems like everything has been replaced by auto-generated slop"
bingo
People say AI is only getting better in the future. Funnily enough it will get worse because its being trained on its own data now. AI peaked and openAI and other companies are scrambling to find more untapped data that they can squeeze out of the internet for "better" models. Truth is, they need to find a better way of training the AI. This method they are using of scrubbing the web is unsustainable.
It's like back in the day when someone would xerox copy a copy of a copy of a copy... Or like a game of "telephone".
@@fumaholics well not really exactly anymore now AI is finally becoming physical for like robots like optimus gen 2 from tesla and the openai fobot they will be able to go outside and see the world touch and hear just like humans do and just like a baby they will grow and it will take time but eventually they will reach a level of philosophy unmatched in any realm of human endeavour
@@gamers-xh3uc Yeah, but ignorance is bliss, it's easier to think that because AI is currently just regurgitating that means it's "peaked". I don't think people quite realize that humans evolved the exact same way. We imitate. AI just has a lot more to start off with than prehistoric man so of course it's overwhelmed and doesn't know how to use what's already been done to create new, better things but it's not as far off as we want to think.
No art is truly original, nothing we do or think is. Humans overrate themselves, our greatest achievement is creating AI that will exceed and replace us. But looking at what we have done to the planet, it's a shame we didn't get replaced by AI sooner ...
have you caught that commercial from the Olympics? where a dad asks the AI to "help" his young daughter write a fan letter? What's the fun in fanmail if you do not put it in your own words? Writing to your hero should be something you want to do, not something you want to pass along like a business email to your boss
Lol I have used ai to write emails to my boss!
exactly!!! all the AI commercials make me feel so weird. there was one from Hundayi with a clearly ai-generated background (which i think was like, the stylistic choice?) and it just makes me think, "you can't bother to pay people to... drive around in the desert or whatever?" it makes no sense to me
Using AI to write fanmail sounds like creating spam. I mean it's different if it's just for polishing up your letter but to create some generic ai letter using a prompt... I don't get it
I saw that. It was just sad. Corporations really don't know the meanings of the words "heart" or "soul," do they?
Maybe they were disadvantaged and therefore weren't given enough tax dollars to go to a good school that didn't have systemic programs in place like metal detectors. Maybe you're just prejudice? ?
I’ve been seeing ads on porn sites for AI face swap services lately. They even straight up say it’s “good to use for ex girlfriends, celebrities, or people you know in the world.” It’s sickening
Quite.
That is gross wtffff
i have seen it with celebrities' faces when they were clearly children
my brother is a PHD doing medical data analysis for a research department at an in state college and he uses AI regularly to organize and detect patterns in the data. The AI doesn't make up the data, the AI doesn't do the research, the AI doesn't even get to know what the data even means. It's so incredibly useful and time saving and accelerates the process by days worth of time. That is the kind of thing I think AI is meant to be used for, helping you speed up a task not dong it all for you
Your comment is good...
But your typo at the end is even better 😂
I use AI to summarize results, create templates and translate things, but at the end of the day I still need to check and edit things. I might save a couple of hours of work, but I still need to do a quality check on everything to make sure AI's not altering the information. This kind of use isn't as glamorous, but leads to more accurate results.
And even for that it s still dubious.
Is it worth it to put out of work research assistants to gain a few hours on the processing of these datas ? Is it really a gain for society as a whole ?
That's how all AI works and is supposed to work. It's just normies have been spooked by propaganda about making pretty pictures with just text, without constraints or consistency or anything actually useful. Because someone somewhere was selling this generic, non-specific kind of asset and they liked money.
If you need to make some assets of a pink elephant monster for a game or a movie or something you can't just ask for it and get something you can actually use. You need to do a lot of the work and keep the AI use within rigid workflows where it's narrowed down to what you actually envision and working on mostly what you yourself made. It's just supposed to make you faster, which means you make more assets faster.
@@etienne8110 Yes it's worth making things faster and better. You're advocating for stalling things just to pad hours and wages. Late stage capitalism thinking.
CTO of ChatGPT when AI answers questions in a harmful way: "It's very human to make up an answer, humans rarely say 'I don't know'."
CTO of ChatGPT answering a question where the truth would be harmful to her: "I don't know."
at least the ai only lies bc it's stupid i guess 💀
But srsly tho, most allistic people do have a huuuuuuge problem with admitting “fault” (literally just not being perfect)
What's so annoying to me about that statement too, is so many layers of things.
One-- no, I don't agree with this weird generalization about human nature. There are plenty of instances where people say "i don't know" there's a whole commonly used "idk" acronym for it, the only kind of person who would never say "i don't know" would be someone who has an incentive not to, like a politician or a salesman. The thing people may be misinterpreting as "not being willing to say idk" is when people instinctively try to offer some kind of guess to convey they are engaging with what was asked of them, but that is not the same as outright lying and claiming authority on a subject.
Two-- WHY would a purely algorithm based "intelligence" behave in a way that implies inefficient amounts of human-like pride or social shame or some kind of sociopathic pathological lying habit
and Three-- if such a human trait believably existed in your programming-- why would you sell an AI that is prone to that as if it is a useful engine for information searching and not simply make a program that is unemotional??? lmao its just so bad
I know this was a throwaway bit in the video but a lot of VA's have taken extremely hard stances about their voice being used even for joke things like silly character covers because it's genuinely terrifying to think that someone can make your voice say any number of absolutely horrific things but also because voice work is as much of an artistic skill as physical arts. I've seen as recently as this week someone who did VA work for a company have to start up a legal dispute with them because they used AI to supplement their work without permission or compensation
If someone used my voice through AI without permission, those would be fighting words for me. Fists are flying. Genuinely offensive, disgusting and an insult to my humanity.
A lot of Veterans Administrations?
@@anthonymeade7345 VA = voice actors
I have always thought it was a bad sign how many people thought the AI generated Spongebob song covers were just sooo funny and sooo good, meanwhile I'm sitting here like "Okay it's a cute idea to have Patrick Star sing a Ringo Starr song, but that doesn't mean it's okay to just... steal the guy's voice?"
@@anthonymeade7345 VA means "voice actors" in this context.
"Why would I learn to draw if the AI can do it better than me" is something I've been seeing a lot of in the drawing community. Imposter Syndrome has always been something artists have to work through but it's so much harder to feel like "a real artist" when someone can click a button a produce something more visually (or audibly) pleasing than something you spent weeks on, if not longer. It's heartbreaking watching young artists give up before they ever really get started
Because unlike the imposters
I actually love drawing itself.
Do you?
@@mrsticky005 A lot of people love drawing but are surrounded by negativity (from themselves and/or others like school art teachers or parents) and they convince themselves that even though they want it doesn't mean it's for them. Just because someone has a passion doesn't mean they feel they are allowed to feel that passion. That's why it's a Syndrome and they aren't actually imposters. I personally stopped drawing because of it until I was like 29 and had been through a bunch of therapy
the best thing we can do to combat ai generated image crap everywhere is to do our best to encourage those young artists. it's hell out here for them and we just have to do what we can to support them and encourage them onwards. Oh and ofc let them know that people who use ai image generation are pathetic hacks who are too lazy to put in the effort learning real art, and they're stronger than those people could EVER be by just TRYING to learn. every young artist who picks up the pen on their own instead of giving in and being a worthless ai image user is so strong and admirable
It's far more involved than just clicking a button - have you tried to use AI image generation? They want you to jump through a bunch of hoops just to use it and then you need to feed it good data and prompts to get anything even remotely approaching useful heh
@@Beckelby Welcome to capitalism where we can't have time to just enjoy life, we have to be "productive" and find a way to monetize our hobbies if we even want to afford to do them in the first place heh
As an artist, I really appreciate you grabbing MSPaint, drawing and image by yourself (those mountains genuinely look good, unironically and unbiased, you are cooking!) and using it in your video! It was genuinely a nice touch that really shows your humor and sensibilities and I will appreciate this a lot more than somebody going for that unrealistic polish AI is trying to replicate so badly!
Jank is life, people! Embrace it! 🧡
Totally agree
It's iconic
Hey does anyone know where I can find the ai generated Gabi birthday suit pics? Asking for a friend
@@brian8507🤮🤮
@@brian8507up your own ass! hope this helps
I love how she said that "humans don't say 'I don't know'," but when she was asked if the AI was trained on videos from youtube, instagram, and facebook, she literally says "I don't know/I'm not sure."
She’s also wrong. I’ve seen several different AI programs say things like “I don’t know” or similar. For a professional on AI, she sure doesn’t know how it works at all.
I believe she knows how it works. She is seemingly intentionally NOT admitting to stealing public content to avoid legal action by creators@@枒
I will say I don't know all the time! I'll often make a guess about something at the same time, but I'll make it clear that it's a guess. I hate misinformation. It's so, so dangerous
@@msjkrameybeing able to say "I don't know" totally shows the level of intelligence someone has.
And it's not even true!
Humans lie and make stuff up in certain situations yes, but doesn't apply to all the topics and situations, and not to the level of the hallucinations produced by AI, where things that don't make sense in the context are present.
For example, making stuff up on the spot about a certain recipe we have no idea about would be unthinkable for a normal human, why even do that? Specially suggesting adding glue to pizza, outside of a joke context...
Personally what I am excited for is every single legal case defense being "Um your honor, thats AI" and needing to bog down an already bogged down system waiting for an "AI expert" (or an AI, AI detector) to validate it.
I need a new Ace Attorney game just so they can do this LMAO
I'm waiting for the AI wars!
I'm scared that crime evidence can be just made up so easily, that can go wrong so fast
That SEEMS practical until you consider that most security footage at places like stores or banks has all the resolution of a malfunctioning vintage TV set.
Then the judge would say, "This video is not high enough res to be AI."
@@Lesrevesdhiver nah, AI can generate low res images that looks like Security cam footage
"and these are the ones of me clothed" really hit hard
My favourite google ai fail is when it was asked for fruits that end in um, it responded with things like raspberum and tomatum and coconut 💀
Applum my favorite frum
Well, coconut is correct, a coconut is considered a fruit. So, 33%! ☠️
add 5 to your age,.,,,..
To be fair that is how I'd respond to that question.
@@renharris6504 coconut doesn't end with um
“Co pilot for every profession” + “we will just make up an answer” makes me TERRIFIED for the future of healthcare if nurses & doctors rely ok copilot 😬
absolutely it’s scary how far it’s already getting.. just got back from EAA Airventure where they had a seminar for a digital copilot advertised to long haul SOLO pilots. They want fedex pilots to have AI helping them fly (reading out generated procedures, telling them where to go, which runway to use etc.) why not use an algorithm!!! Algorithms are more reliable and accurate, and can actually retain knowledge about your checklists and stuff. AI just makes shit up!!! planes already have such smart/high tech instruments in them you just have to learn how to use them.
I just hope the people in these professions are smart enough to not use them. although a lot of times it’s up to higher-ups and it seems like they tend to like AI
@@sami-tw7qk I’m already TERRIFIED of flying and that’s with human pilots 😬
Tbh I do agree that AI is being used for useless stuff, like arts and music. There is a bit of reason for video generation which is having a 'world model' for the world but I don't think it is best used in that area. There is some merit to what you guys are saying. The reason there is a lot of "AI copilot" is because humans aren't perfect just as AI isn't. But humans + AI would be better than just only AI or only human.
You are overestimating algorithms, humans design those algorithms and sometimes humans forget things (miss out edge cases). Also, humans get tired of doing the same thing over and over again. I cannot overstate the humans + AI thought.
Ngl though doctors are not perfect too. You're forgetting the other positive effect, places without doctors, would get free healthcare even though it wouldn't be perfect. It is better than nothing.
If implemented correctly, AI can stop making shit up. Claude AI is already building towards that.
Point here is, AI is new technology just like computers and algorithms before. It just started a few years ago.
AI for the poor, Humans for the rich
But what if AI produces results which are better then doctors and nurses... it is literally worse to not use them, you know humans aren't infallible creatures..?????
There have already been research where these new technologies ARE better than doctors/nurses in specific domains, and this is only going to increase as it gets better.
I had a survey from my university about how staff feel about AI implementation for our university. One question that concerned me was "how do you feel about AI being used for grading?"
The issue with AI grading assignments is that students will most likely get the mindset that if the professor or TA doesn't care enough to put in the effort to look at my work to give me feedback to do better, why the hell should I put in the effort and spend thousands to do this?
no literally like what is the point of going to uni if you're not getting feedback from your professors or ta? like what are THEY getting paid for too
@@greeny.official Exactly. I could also see this encouraging cheating as well. They really can't hold it against students that use AI to do their assignments for them if they use AI to grade the papers!
I'm uni rn close to graduating from a CS course. We have students using AI to do assessments or cheat on tests (even though they were warned they'd get a fat 0 if they did that) and professors using it to write lecture material. Like what is even the point of being there if you're not learning or teaching?
Which leads to the obvious extension of "teaching to the test" : "Writing for the algorithm". Which must be incredible writing, I bet [He said sarcastically]
@@omnipenne9101how can you be close to graduating and not already know that universities are primary centres for research and academia? Teaching is basically a necessary distraction to your lecturers... Most of them really don't want to be lecturing and tutoring you - but they are forced to by the University as part of the agreement to let them pursue their academic interest. And the learning part is 100% on the student.
This was such a good break down of everything wrong with AI. I'm so happy that you hit on the idea of training redundancy. Ever since tech bros started bragging about how AI will only get better, I've been thinking that itll probably only get worse because itll be forced to train on itself once it steals everyone elses work. Like a xerox of a xerox, itll slowly decay into nothing. An ouroboros of false information and too many fingers per hand
30:00
So fun fact, this is actually a huge problem in fandoms. Like, full on poisoning entire fandoms. I've mentioned this on Tumblr and I'll do it again here.
The way fandoms work, is a bunch of people get together over something they like and start creating stuff for each other, they bond over shared interpretations of the media, sometimes they bond over different interpretations of the media, and the whole thing is one happy family (that argues over their favorite ship).
When a fandom gets older, the people in it get older too, eventually you have some main people in the fandom. There are a few older people at the top that the other people look up to. Kind of how you've got your big boy UA-camrs amongst everyone else. These bigger people in the fandoms pave the way for the newer (and often younger) people to grow better at art/writing/ whatever. It's a healthy system. Like how the older generation raises the younger ones.
But boy when you throw ai into the system does everything go horribly wrong.
Pre-ai, if you wanted to make a piece of art, there was no one who could do it the way you wanted it, except for you. So even when you're a nervous, self-conscious kid in a fandom, you eventually tested the waters and got in. You learned new skills and made friends. If you wanted to make something, but didn't trust yourself, the only thing you could do is learn to trust yourself. Let yourself suck so that you can be good at it eventually.
But when ai is there suddenly these kids have an out. You want to create something, but you don't trust yourself, you don't believe in yourself, so you have the ai do it for you. And who would blame you? There's a button that will do it for you, and you're scared of doing it yourself! So you push that bottom. You let ai do it for you, and boom! You've got a thingy! You've effectively skipped the tutorial level that teaches you to believe in yourself, and to do your work.
Now you're in the big kid pond and can't do anything. If AI fails you, if someone insults you, you don't know how to bounce back and you just crumble.
And on the outside, to all the big kids, you're some talentless weirdo that thinks ai is the answer. On the outside, you look like the problem. Now people are pointing and laughing at the victims of ai, who didn't know the dangers beforehand.
Ai is robbing kids of skills they _need_ to learn for fandoms, and for real life. It's absolutely horrible. And it's so dangerous for these kids.
So whenever I see the summary of a fanfic say something like "yeah, I used a little bit of ai, sorry! I just wanted it to be good, I did _some_ of it! So enjoy!" I die inside and think "who told you that a crappy robot was better than you? Who convinced you that the robot's horrible job was better than you could achieve? Cause they lied. With a bit of practice, you could blow this out of the water."
Ai is the death of creativity, and self-esteem, we're killing Gen Alpha and the younger half of Gen Z. In more ways than just UA-cam brainrot.
damn that half an essay👍
We've somehow as a civilization decided that potentially all future generations of artists, composers, writers etc. need to be culled in the name of efficiency and "democratizing art". It's hard to believe that this is where we are as a species.
@@hillehai it's all getting so industrial. I can't believe I live in a time period where companies figured out how to mass produce freaking ART. CREATIVITY.
Things like ai that roleplays a character are a problem too. People used to get together and each choose a character to play for the rp. But now that chat room ai exists people aren't going to each other for the fear that they'll be judged. And instead the robots. Which not only lowers the quality of the plotline, but ai feels far too fake to satisfy a social quota that keeps us mentally healthy. It's exacerbating the crippling loneliness that's breaking younger generations (well, everyone really, but kids are hit the hardest). This whole thing is creating enormous groups of people that don't know the first thing about talking to each other.
(I'm talking way too much sorry)
This AI thing dominating the creative scene is why I am in a fallout with the current art community and had lost my job as a designer. Now I am considering either working in a book retail like Barnes and Noble or enter the military. Sad to see all my former classmates at art school now work in menial jobs, with one of them now working as a fry cook in a local McDonalds.
wow this is a really terrible attitude to have towards people just expressing themselves with a creative tool that isn't hurting anyone. Real "old man yells at cloud" moment here.
I have 6 years of experience in customer service and fraud prevention, but I cannot find a job because companies would rather save money with infuriatingly useless AI chatbots that end up requiring a cx agent anyways.
Edit: somehow I am being called a liar and that I proved myself wrong because peopled actually don't get it that companies are DOWNSIZING THEIR WORKFORCE IN FAVOUR OF AI I genuinely didn't think I needed to explain that lmao, one company I worked for went from 100+ agents to 30 within 5 months.
If I can't talk to a human for technical support the next time my internet goes out I'm abandoning everything and living in the woods
Yep and if they don't use AI they just have their customer support be based out of a far lower cost country than the US
Because it's cheaper, and all they care about is profit.
@@alext3811 how will they made profit if no one uses their products? those days i was helping my mom with a problem with her bank, their customer support was all AI-based and it required a lot of effort to get to talk to a human being, but their site wasn't even available and the phone lines were shut down since december last year. she ditched them 🤷♀
Innumerable jobs are made obsolete by technology and industry. Learn to code
Day by day, the dead internet theory is becoming a reality.
Deffo
dead internet? what's that?
@@kazu9445the theory and idea that most of the Internet is just bots making comments for other bots to interact with and that there are entire circles of the Internet comprised of non human automated engagement farming
@@kazu9445 the idea that the majority of interaction on the internet is not human, but automation and bots.
the amount of ai generated pictures and bot comments on face book is getting crazy
my absolute mantra as someone who works in Marketing and design right now is "If you can't be bothered to write it, I won't be bothered to read it
What I don't understand is why are we as a society giving robots better treatment and possibilities than actual humans. AI has take n over art, music, video content, etc etc. And yet we are still working all day in minimum wage jobs. I don't want to work for 8+ hours every day for AI to make me preety pictures for me to stare at. I want AI to work for us so that we can enjoy the good things in life. I just don't understand.
Yeah ... no. - The rich in unison
I think we should treat robots like how America used to treat minorities.
Because robots don’t deserve rights.
I'll definitely quote you in the future, you perfectly nailed why Gen AI in creatives is so fucking unhelpful and harmful, it's all corporate and soulless
This is the crux of my issues with AI. Creating art is one of the biggest and most intrinsically human ways to fulfill our short time on this planet, and making a living out of it was already a path filled with hardship. Then these companies show up and just airdrop steamrollers all over this scenario without a care. I'll never understand the pro-generative AI crowd, to me it's almost as if they hate and resent artists...
Because humans are cheaper for many things. And because we don't have any economic model to support people not working to earn money. There's no actual right to food, water or shelter - only the right to buy it, if you can afford to. So if there isn't enough work available to keep everyone employed, everything goes to shit really fast. It's a trap that we are all stuck in, collectively. Governments must constantly strive to grow the economy, to create more jobs, to keep generating enough work to keep everyone employed. Doesn't matter if the job actually matters at all, that's another person saved from poverty by letting them work eight hours a day writing leaflets for a company that sells Disney-branded umbrellas.
big labels suing ai music companies bc their revenue is getting siphoned away is fucking peak irony. literally "stealing a meme vs having your meme stolen" shit
Maybe it's easy for AI to copy their music because they keep making the same thing over and over? Oh, look, it's another 4-4 pop song singing a love tribute to a nameless partner.
@@vylbird8014 Haha yes that is for sure part of it
Makes sense Drake the industry rapper used AI for his disses
@@vylbird8014 I know this is sarcastic but you are unironically right. As long as most ppl produce the same things in the same way, AI will be just as good. If you get original, that's when no AI will be able to emulate you.
@@vylbird8014i hope you're not the kind of person who thinks ai generated music will be more inventive and soulful than current pop music.
What really gets me is how useful I know AI can be in places such as the medical field where it can do stuff like detect cancers long before a human could, and yet all the biggest AI companies are working on the most useless and immoral areas.
Are areas that humans have historically enjoyed doing
The rich wanna get richer without needing pesky morality in their way
Thats crony capitalism.
@@fabiankehrer3645 that's just capitalism.
@@439801RS (preface I hope this sarcasm is obvious haha) People don't have to waste time on the arts anymore thanks to our AI overlords! No more expressing your humanity ever! Now stop yapping and get back to producing surplus labor value you lazy sheep- I mean my wonderful employees!
as an artist the number of people supporting generative AI is disheartening and demotivating. i just had an argument with a guy asking me whats /really/ wrong with genAI (even tho i was like, doesnt really matter, its stealing) and i kept trying to tell him it doesnt have capacity for creativity like we humans do. he kept going to "how do you know that/cant you prove that" and "what if it could someday." i gave up.
it just feels so useless and idk what we should do now
Get out, and learn how to grow your own potatoes
you cant argue with NPCs.. they feel proud as "producers" if some stealing machine transform for them plain text into audio file.. Its thinking pattern of such people, they are basicaly thiefs too... Maybe all of this is just test of civilisation if we can sustain spirutualy or shift to braindeads..
I refer to it as AI imagery.
No most people who don't know the answer to something say "I don't know." Imagine asking a stranger on the street how do you cook a pizza, and they just make some BS thing up instead of saying "Who are you, weirdo?" Nothing AI does is "human like".
Honestly, that's a real good point. If google wants to claim ownership to the videos on UA-cam, then they're liable for defamation when one of those creators goes off the rails. Its contradictory to what they've been arguing in front of Congress that they shouldn't be held accountable for the creations of those on the platform.
who are you weirdo 😭😭💀
Actually this is not quite right.
There are plenty of cases of humans with brain issues that will retroactively hallacinate and provide incorrect explanations bc they don't know and are completely unaware themselves that is what they are doing.
Confabulation is the term used to refer to this phenomena of the human psyche related commonly to amnesia or dementia.
That is essentially how an AI with a brain fart responds as well.
Also this gal does not understand what an AI is doing.
It is not just copy and pasting what it has been trained on, it is distilling what it has been trained on down to more fundamental concepts that it can then itterate upon and that is why it can accurately and consistently reproduce pluasible output.
It is arguble that humans do the samething when we learn how to produce writing, images, music, etc.
I honestly don't understand why so many people so desperately want to deny that ML can learn real knowledge and then apply that knowledge in new ways...but that is exactly what makes AI useful in the first place and exactly why it is being applied in the first place.
To find patterns and concepts in the data that humans cannot find without the AI, and if humans knew exactly what those patterns and concepts were we would not have issues with hallicination or aligment problems in the first place.
Also google does not claim ownership of the vids, but you do sign a licensing agreement with them when you agree to terms and conditions that gives them certian rights to use your content.
It is legal bc you agree to it, it is also a standard agreement with all social media platforms.
It does not make them liable for what you post, it gives them certain permissions to use what you post as a condition of your ability to use their platform to host your content.
@@memegazer do you not see the issue of LLMs and genAI being used in schools and shoved in the results of google searches? you wouldn't want a judge hallucinating when doing their job, but you had a judge use chatgpt, which cited cases that never happened. you wouldn't want a customer support agent to tell customers about a policy that doesn't exist, but a airlines chatbot did just that. that's the issue with ai hallucinating, it giving false information in a PROFESSIONAL setting, and in the case of the airlines chatbot it led to the airlines getting sued. it wouldnt be bad if employees and entire jobs were being replaced by said hallucinating LLMs under the guise of productivity.
Sorry to hijack this but... saying that there's NOT ENOUGHT DATA means that they will steal EVEN MORE of our data, even if they don't say it in their privacy policies. DUMBPHONES PEOPLE! RETURN TO DUMBPHONES! SEND A MESSAGE!
@@memegazer professional Alphabet glazer. Someone likes giving up their privacy for an orwellian society. Maybe if you worship harder, the AI algorithm will increase your digital social credit score.
i had an old friend who shared that there was an ai generated nsfw picture of him floating around and that it was not him and to please delete it. i felt so bad for him because thats just so scary! another thing is remember when scammers would say "i have a picture of u doing stuff and if u dont give me $1 billion i will send it to all ur contacts"? like back then, they didnt actually have the pictures.. but now think. with this ai stuff? anyone can have anything on u its so scary
Black Mirror!
Actually, I think this can actually normalise having nsfw images floating around.
Like imagine if your real NSFW images got leaked. Due to AI, it's easy for you to now deny that it's a real picture. You can claim it's AI.
When everyone has NSFW images of themselves floating around on the internet, nobody will care that there are NSFW images floating around on the internet
As someone who lives in Toronto, I remember the coverage of the Jennifer Pan case when it happened. Not once were there images of her smiling like a party girl. Seeing those ai-generated images for the documentary immediately made me say "that's not the same person."
That’s honestly horrifying
What's even the point of doing that? Honestly disturbing
In the 1950s people imagined future where robots to all the boring tasks and humans enjoy life, create art and music. Nowdays robots create art and music, while humans spend their lives doing boring tasks.
How ironic
Robot 1 = human 0
Mole pfp!! lets gooooo!!!
As a cybersecurity and privacy professional: the amount of people who want to focus so much on A I....but do not have the infrastructure to support a healthy or mature IT environment is huge. Hugeee.
nobody cares about the infrastructure until shit stops working
@JimAllen-Persona Those these folks have great counsel, some of them don't understand the technology. Out of team of 10, you may have one person that understands A.I. deeper than a UA-cam video. I've seen this problem with Data Subject Access requests, for example. Someone wants their data deleted (counsel can understand that) but how do they defend not being able to pull it from an data set used by A.I? They don't know. 😂
yeah - I was actually looking to invest in cloud computing and quantum computing. I'm far, far from an expert but can tell hardly anyone is currently poised to support any of this.
@JimAllen-Persona😬
@VilleUTTP what?
Scarlett Johnson voice in OpenAI is literally "We invented the Torment Nexus from the Sci Fi novel "please don't invent the torment nexus" it surprises me how much tec CEOs all have no ability in media interpretation and see any tec stuff in movie and go "uh shiny"
i don’t think the tech CEOs care about anything except profit or glory, to be frank
@@rrbella He did say Her was his favorite movie and its funnier to imagine he was being honest but is so bad interpreting media he only likes the movie cuz "cool IA thingie" (tho he was probably just lying just because)
I don't think it's that deep at all. They're not doing it out of spite, or because they want the world to end, or whatever. They simply do not care.
@@every_username_is_taken I meant the irony of Her being a movie about Joaquin Phoenix falling in love with AI Scarlett Johasson and kinda of a cautionary tale of how tecnology pulls us apart while pretending to be humane. Then the mf actually tries to get Scarlett Johanson to voice hi irl IA without noticing the immense irony
Tech CEOs are the biggest sociopaths. There's a reason such kind of nerds have never been well accepted by the general public.
Companies branding ai working as it normally does as a "hallucination" is so funny, it would be like rolling random dice to try to solve a math problem and saying the dice are hallucinating when they dont give you the answer. These machines were built to string words together, and have no way of actually cooresponding them to reality or even understanding the strings of words theyve created in the first place
How about an image recognition AI that labels a blank wall as a dog, then?
Maybe? We don't even really know what understanding is yet.
It's so obvious that these corporations are lying when they talk about "hallucinations" as if they're a bug that can be solved and not an uncontrollable and fundamental aspect of the technology.
@@niklasneighbor6726 if we can see it in the clouds they can see it in the grains of dots in the static?
All this talk about AI taking over the world is just a humble brag saying "oh no we were too good when we made this thing and now it's going to replace us"... sure buddy.
Deskmats like that are such a cool merch option, holy shit. I love deskmats and they're not quite my style but such a unique merch option, love the creativity
I read a comment on a different video about this topic that summarized my thoughts on "AI" very well: "Why would I want to read a book that nobody bothered to write?"
“What is written without effort is in general read without pleasure.” -Biographiana (about Samuel Johnson) from 1799
Yes, exactly.
Why would you enjoy a picture no one created (a photo). And yet they're typically is far more human interactions and creativity still involved in good A.i. work.
While I totally understand a lot of criticism around A.i. (particular types, not all are equal), but many appear to be riding a bandwagon of hate.
@@Fastlan3 Because they don't improve the service, product, or site. The quality is bad is one such reason.
@@Fastlan3photography is created by human with tool , but its is still human that that compose everything,edit, and do many more things. They don't tell their camera to take nice pictrue.
Honestly, it's kinda cool to see how the rise of AI accidentally gave birth to another wave of "internet ugly", meaning people presenting themselves as more human again, instead of trying to archieve the corporate perfection look in all of their work. I'd love to see this being adapted by the traditional media aswell. Before any journalist or editorial person even thinks of using AI imagery for their articles, they should just open MS paint and draw something themselves to the best of their abilities, even if they never drew anything in their life. There is just so much more there. An actual human vision. An expression of someone's creativity AI can't replicate.
I’m a journalist, and whenever we get a spelling correction (which people often lump in with insults) I just take it as - well at least you know it wasn’t written by AI!
@@regiergal Yeah, it's real hard for someone to put in an error on purpose.
The we demand 5% error in everything you do just to justify humanity, brigade.
…YET! 😭
Hmm… More messy you say? That sounds like an aesthetic we can generate superficially just like everything else.
I may eat my words but I do think „AI“ is the perfect tool to imitate anything put on the internet in a way that’s good enough for corporate hacks as long as these idiots are allowed access to everything that’s on the internet.
I lost my job as a concept artist because my boss just replaced my work with AI concepts. The higher ups didn't know or care, so i was made redundant.
as an artist, i think it's pretty good at concept art
@@yummyliciosablei really don't care how good or bad it is, it's taking jobs away from people.
You are not redundant. Your boss just fell for a marketing scam. Im sure you'll find a place that appreachiates your work.
L boss
@yummyliciosable Concept art exists as a form of communication between the people with the ideas and the artists that will make the idea work. It's not just there to look cool. It's a blueprint. Using AI for inspiration is not the same as using AI for official concept art. Concept art has a specific purpose. It needs to be readable. AI, with its weird blurry details, is not ideal for a project where a bunch of different artists working on a deadline have to figure out what the weird blurry details are supposed to be.
I like the type of AI that can help doctors recognize tumours and help people in that sort of way. The AI that just does art isn't really that useful to us as a species. All the AI people call it a tool, but it's more akin to straight up outsourcing, like you said.
I’ve noticed lately a rise in AI generated crochet and knit patterns, both for the image of the product and for the pattern itself, and it’s so insidious because unless you have been crocheting or knitting for several years, you will not be able to tell the difference between real images and patterns and AI generated ones. People are getting scammed out of their money to be given a pattern that makes NO sense and cannot physically be done, and it could keep people from ever picking it up as a new skill, lest they be scammed
I find it wild that you aren't the first comment I've seen mention crochet/knitting. Not even the innocent facebook grandmas are safe from their past-time. I find it wild that the AI content farms have invaded something as niche and innocuous as Knitting/Crochet, literally any avenue they can find to make a profit these days.
@@Lemonpeppergoblin I’m in a bunch of Facebook groups for knitting and crochet and you have no idea how many times I’ve had to tell someone that the image they’re looking for a pattern for is AI. I post about once a month in those groups with a check mark list about how to tell if what you’re looking at is AI, but the newer ones to the craft still don’t really understand how to tell it apart, and I genuinely hate it for them. I’ve even got random swatches I made up to try to show them what real stitches and AI generated ones look like, but newbies eyes aren’t trained to see the difference yet so it mostly looks the same to them
With every passing year I'm more and more nostalgic for the 2000s to 2014ish internet. The current net blows and it doesn't seem like it'll get better any time soon.
I do like that the current internet is more pro-woman in the surface, but if you scratch it it’s just the same misogynist cess pit it’s always been because it’s just a mirror of society and society hasn’t really made female existence more comfortable or safer yet.
@xp8969 that sounds like you're saying rural communities made the internet worse
@@xp7575 The confusion probably comes from how absolutely stupid of a response it is if you read it as intended. I was born in the rural South in the 90s. We had internet genius.
@@kylegonewild obviously there were a few people with internet access in rural areas back in the day but it was rare, rednecks getting online en masse this century ruined the internet
The problem isn't rural areas it's phones. Phones should never have access to the internet. Or at least very limited access. Phones are the root cause of a lot of issues with the internet, with gaming, everything.
My friend and I entered an art contest a few months ago and she lost to very obvious ai “art”. She spent months on a self portrait oil painting that even when putting my bias aside was amazingly beautiful and definitely deserved to win. There was ai competing in the charcoal category with me too but luckily someone looked at competing pieces before the judges actually scored anything and bombarded the hosts of the competition to remove the piece and they did eventually. It’s incredibly frustrating especially considering part of the first place prize was a scholarship that my friend definitely deserves and needs, and the contest hosts were very hesitant to remove the ai “art”.
Understandably frustrating but it's also very funny to imagine an AI 'charcoal' work's sender getting a scholarship to an art school and being handed an actual piece of charcoal and suede and get told to get to work.
@@jasminv8653 honestly yeah when you forget about how awful it is the whole thing is really funny 😭😭 like what did the person submitting the “art” expect to get out of an art scholarship???
A competition for art made with physical media should require submissions of the originals 🤷♀️
@@heatherduke7703 normally yeah, but this one took digital scans of the peices for some reason. I agree though, all physical media competitions should require submission of the physical piece
The future of art contest, now need to upload all stage of process.
I want robots who can do my laundry so I can paint more, not robots who can paint for me so I can do my laundry
It's called a washing machine.
@@jaybanzia washing machines don't sort, fold, or put away the laundry
@@vonn_kate You want to hire a maid but you can't afford it.
@@vonn_kate you want to hire a maid but you can't afford it.
@@vonn_kate there's machines that do all of that too, but they're expensive and take up a lot of space. The engineering problem is pretty big and costly to make a machine that just "does your laundry" where as programming an AI replicate human digital output is literally a processing power problem.
My Facebook is now just AI generated images of isolated cabins and impossible houses with boomers/idiots commenting on how beautiful those places are.
Tbf more and more of those comments are just bot farms themselves, but yea you're right
I don't facebook, but I've seen my mother's and it's pretty much that but for dogs.
Beautiful cabin crew ❤
Scarlet Johanson 🌹🌹
Why are you still using FB? Genuine question. I quit over 10 years ago when the warning signs were already there (albeit different warning signs). It just seems like a wasteland of arguing and misinformation farms.
@@vylbird8014🐶🐕🦮🐩🐕🦺
I recently had an actual human try and steal my art to pass off as her own. She's a scammer and the people that know about her and are warning the ones falling for it alerted me to it. I had no idea it was happening and I never thought in my wildest dreams that would happen. Now I have to worry about a no name, no face, no body having computer stealing my work just for funsies and money that I desperately need. I hate this timeline.
@VilleUTTP Go away troll
Spill I need to know your timeline story it's so entertaining to me
@novasfashionfunchannel8287 So she's this woman who had gastric bypass but lied about it and just tells people she's chronically sick for unknown reasons and is constantly getting "donations" for her "hospital bills" while also constantly buying luxury items. So she's an "artist" or at least claims to be but anytime people ask to see her work she says "just Google my name and you'll find my fb page." We have the same name which is the name of my fb art page except I have a different middle initial. She even brought it up on a live on night to show people and everyone was like, "great work but what's the R stand for?" That's how they started figuring out that it wasn't her work and found me. She's not actually an artist at all. She worked at an art bar, pretended to own it but never did, and still will just say to Google her name, knowing it brings up my art page, to her followers. I had to make a post on my art page with a bunch of pics of myself and letting everyone know that she's trying to steal credit for my work. But the people of the sub have been lovely and I appreciate them letting me know about it. It's been wild.
Everything is gonna end up like WALL-E soon...
@@kootunesscrewy I don't know what that means
My final ad break for your video was legitimately for the exact Adobe AI tool you spent minutes of the video railing against, and it killed me. My husband happened to overhear just enough of the video prior, that we spent the whole ad gasping for air together. Then I gave him the context he missed, and we spent another few minutes nearly crying. Truly special.
I legit used Notion for the entire duration of my degree for every lecture's notes, every assignment, group work... until they added AI. specifically bringing up a whole dropdown window every time you press the spacebar at the start of a new paragraph, which i did constantly. literally the fastest i've ever stopped using a piece of software since there was NO WAY to disable it.
18:23 - they love this term. "publicly available"
Just because something is publicly available that doesn't mean it's free to use. An artist still owns the rights to their artwork after posting it on instagram. Just like how if you park your car on the street, it's publicly available, but it's still theft if someone takes it. What that term actually means is "we could access it", not that they had the rights to use it.
Don't let them bamboozle you with wooly language.
I remember the term "publicly available" being used by some pro-AI guy screeching with that term and how artist were being "entitled" just because an artist didn't want their art being used in AI image training and then complaining on how they were against "progress" in doing copyright claim or something. I don't even care how much someone would say how there's just as much bad human made art and the like, I can't really see AI art being nothing more than a fancy toy that'll be used by a business suit as a "cost saving measure" and AI generated art, no matter how good it looks to be, to be meh compared to a human made drawing.
The Mona Lisa is "publicly available." So is the Statue of Liberty.
computers are better at manipulating and processing data than restricting its use. people love to think like they're in control of everything the put on the internet, but really, people barely have any control the moment something is out there on the web. that's just fundamentally how things work, it's kinda hard to restrict computers from computing in specific ways
it is free to use yes
@@selectivelysocial7117 its not publicly available, you have to pay to see it
18:51 "publicly available" DOES NOT MEAN "publicly available to use" that's not how freaking copyright works.
just because a song is on youtube doesn't mean i can put it in a commercial.
it's not used directly, for copyright it matters only if the original work is reproduced in significant part.
if an AI is trained on a song and creates a new song based on parts of the original that’s baaaaad, but if a person listens to the same song and is inspired by it to create a new song reminiscent of the original that’s gooooooood. make it make sense lmao
Duck u and duck ur copyright laws
Trouble for the "AI" companies is that they'd never be able to afford the training data (particularly high-quality training data) if they actually had to pay for it. Then there's the whole issue of curating the data.
It makes me think of how I can go to a publicly available store and walk inside and see all the stuff just on display, but that doesn't mean it's free to stuff my cart and walk out without any consequence.
There's so much messed up about this
One of the best takes on AI I’ve seen. I work in the political sect of the world and it’s been a huge issue with interns using AI to write memos so false information gets passed up the chain of command and many times if it’s not “consequential” it could easily be overlooked then given to the public.
Drew goodens vid was pretty good too. Also a video called everything is bland is a good watch
17:36 UA-cam inserted an ad break right between the previous clip and your reaction, so this face was the first thing I saw after coming back from a weird candy bar ad. 💀🤣
7:42 I unironically appreciate the MS paint drawing more. Like, it's cute.
Agreed.
doodling something quick like that has personality. even so, it seems a bit silly to make such a big deal over a background image. like, nobody's going to get hurt over a background image being AI
@@noxxyy It hurts your soul when you have to look at. Especially when it's everywhere. It takes a psychic toll on you when you're surrounded by garbage all the time. It's the same kind of attack on your soul as watching the same ad over and over every time you click a video.
@@noxxyy You'd be surprised. I follow a smaller content creator (about 10% the subscribers of Gabi) and they used some AI generated stuff ironically and had people saying they didn't like it.
@@whitemakesright2177 my soul seems rather fine, and i don't see ads to begin with, so... yeah, i think people are just being a bit over-the-top with all that, people should care when someone's genuinely using AI to trick people, not when they're just using it for quick and easy filler for backgrounds and placeholders and memery
Imagine a computer that just flat out lies to you when it doesn't know the answer, or even just because it feels like it at that particular moment. That'd go down great in any big company I'm sure.
Its just a matter of time when a product sold to a corporation like "improvement of the workflow" makes a huge mistake that leaves an entire corporation at risk. Just then they will be questioning the AI tools. Meanwhile tech developers will keep on selling bullsh*t stamped with AI
Its not feeling like it though. The algorithm is a glorified auto completion that gives you the most likely answer based on its collected data.
@@happytofu5 Not quite. Auto-complete would recommend the same word no matter how many time you erased and retyped, because as you said it's using a set of instructions and usage data to come to a conclusion. If AI worked like that, every text prompt for an image would come back with the same result. It's to simulate the "human" element, something you absolutely do not want a computer to be mimicking.
Giving a moustache you never asked for to an AI-generated crab is like adding an extra zero to a quarterly earnings report because a human "might" make such a mistake, even if it wasn't asked for. :p
@@Cantimule Quite indeed, actually. "glorified auto completion" does not mean "literal auto-complete your Android phone uses." The difference is auto-complete maintains a history so prior inputs impact future outputs. The generative algorithms are black boxes where the output does not depend upon prior inputs and indeed does not consider them. The function of bots like ChatGPT is still ultimately just trying to guess what the next correct word is and output that. All that other shit you mentioned is irrelevant to the analogy they drew.
That lady said that AI is really human and that when we don't know something we make it up. So why would I ask AI something instead of a human seeing as it would be the same? Or better to ask a human. At least you can tell your friend is lying 😅
Funniest thing about this all is it enables the worst behavior from the worst people, my own experience of generative model misuse was asking dad to apologies for being a bad person (not going to provide context), he offloaded responsibility of texting me to chatgpt over the course of like 2-3 months of discussion, i thought he finally did some thinking and gave an ok apology, and I was ready to let him back into my life after 5 years, only for me to realize each message was written by chatgpt after 5 minuets of thought and inputting some word salad (which he proudly provided after I called him out on it).
I love you son. Please let me know if I answered your question okay so I can enhance my models
This is more fked up than I would’ve ever imagined 😭
Unhinged behavior, I'm so sorry.
WHATTT !! IM SO SORRY
Quite possibly the sickest, darkest, most horrifying use of AI I have heard of thus far. My sincere condolences.
chat gpt is like if we let jordan peterson answer every google search
Posted this on Drew's video, too, but I really appreciate creators with platforms such as yours speaking out on this. A lot of us in the art community have been really deflated by this whole thing, feeling like no one cares the threat this poses to us--and the world at large, really--so it's really meaningful to see folks like you guys getting our backs on this.
And I bet it isn’t any better that promising artists are problematic or groom minors.
Kyle Carroza, Animator320, Gery, Jonochrome, the list goes on
I’m a tutor at my uni’s writing center, and I had a student say, to my face, that he wishes there was an “AI room” in the library where students could go and get their essays written and/or edited and/or “peer reviewed” by AI.
Guess I’m chopped liver, then. Thanks for wishing that AI steals my job.
I wanted parts of this long time ago. We could have had all the lectures on video for the last few decades, on disks. Students watching the videos, and maybe having some software teaching and testing them. Official tests would be under human supervision, of course.
The powered weaving loom took the job of thousands of people weaving by hand, and that's a good thing. Now that weavers aren't needed, we can all get jobs making Playstations and stuff. We wouldn't have had the labour spare to do that, if we still needed to make everything by hand. Physical jobs are automated out of existence, and that's good cos they're usually crap and unrewarding jobs, and society becomes richer because things get cheaper.
If an AI can do 90% of the job of a manuscript editor, why not? I don't imagine it's fun work. You can do something else instead. Sure it sucks for you in the short term, but on an historical scale, it's infinitely better for humanity to have progress. It'll likely make it cheaper to be a professional author, so perhaps more people will be able to do that, using their language skills creatively instead of de-mangling bad grammar all day.
It's not jobs that people need, it's money. An ideal world would have machines do all the work, and money shared out fairly, so we can all chill out and do whatever we like.
I'm an esl "tutor" rn (which on the site I use, 90% of the time its just talking to people), but it's been slowly forcing ai on us more and more. they started generating ai summaries of each class and they now have a weird ai chatbox for students to use, but it's insanely scary because you KNOW they ARE secretly using lessons to train ai-- either selling to third parties, OR using it to replace actual human tutors in the future. it's definitely not legal either way and honestly morally fucked up, it's just sad because we're about to see a LOT of professions die. WHICH I know is a thing that just happens when there's progress. difference is, I don't like that it's replacing things tutors or storyboard artists 😭
@@greenaum how is manuscript editing a physical job? Also, if you’ve ever been irritated by a word processor or your phone autocorrecting something that was already correct, you’ve experienced why this is not a good application for AI. Language is deeply complex, grammar rules can and should be broken at times for artistic effect, and humans are vastly better at explaining to other humans what could make their writing better.
@@greenaum
Problem is, this is a really idealized future that i frankly, can't see happening even generations forward.
Yeah I'm all for my menial work all be taken care of ai but as of now, ai is not used like that.
Ai is being used as this piece of tech utilized by big out of touch companies to replace the work of humans, even though it'll completely compromise their end result, and making everything worse in the process, all for just saving a few bucks (late stage capitalism amirite).
Hell, ai of today is not really "inteligent", it doesn't have a mind of its own, ai is just buzzword for marketing, this version of ai is a glorified algorithm.
Even then the ideal world with ai we all dreamed of might not even use this technology/algorithm, but now we must deal with the sheer shittyness of ai that's fueled with corporate greed. and if I'm being honest the trajectory of our future is more distopian than utopian, it's not really enstilling confidence of the future when the work that's ai is currently replacing is actually the creative ones and not the boring menial work that's it's supposed to.
Several years ago, I naively assumed that large language models would be trained from sources like academic journals and text books and other things that are generally understood to be truthful. But that would just be very illegal! It wouldn't be any more illegal than scraping countless youtube videos and reddit posts for the data, but academic journals and publishing houses actually have the legal representation to defend the work they publish from being blatantly stolen!
Plus they are also often not freely available, which idk i love open access but now i am kinda thankful that generative ai doesnt get its hands on it.
Fun Fact!
They actually *did* train the original generations of machine learning systems on open source literary and academic sources!
And then they ran into the same problem A.I. Bros whine about now - it "wasn't enough data". So companies started buying data from companies like facebook, and stealing art by scraping the Internet and pretending the aggregate mix had no ethical concerns!
And even if it was trained solely on academic resources, it would still churn out false information because it doesn’t have an understanding of truth or being factual. It’s only trying to put words together in a pattern that is similar enough to what it’s been trained on to pass as human writing.
@@Shoulderpads-mcgee yeah. No matter what it seems that LLMs will never have the ability to just be internally consistent
i can’t even escape it at my fast food job. AI MAKES THE SCHEDULE NOW. it’s literally terrible
I still can't believe that Will Smith eating spaghetti is a benchmark for AI video generation. The Turing Test of the future is going to involve a virtual dinner with Will Smith, and spaghetti is on the menu.
I fucking love that eeeeeeveryone uses that same video 😂😂😂
🍝🍝🍝
It drives me nuts that Google used to give me the exact answer I was looking for at the top - and now it gives me an ai response that answers my question in the absolute wrong way.
Congrats on your release - they look beautiful next year when I'm a nurse I will be able to have one myself :)
For some really weird reason I don't have that feature on all the devices that I use Google on. Maybe it's because I haven't updated it yet but now I'm certainly not going to update it now
@eclipsemoon01 I looked up how to get rid of it - and you can't. So don't update anything !!! Its so bizarre that Google AI is worse than Google is.... I guess it's just too much info idk
To be fair, the previous search result was also identified by AI. Just, less GenAI involved
"maybe the problem isn't the lack of data" damn right
there's literally research PROVING that, even, higher quality data (instead of "literally everything") results in WAY higher efficiency, you waste WAY less resources on training, and get very good results at a FRACTION of the data...
the problem is: EFFORT, lol, the companies do not want to put in the EFFORT to make actual high-quality data, they want FREE data without bothering to pay people
there are actual ethical datasets, though, where the data is explicitly agreed on, but y'know...why waste money and time, when you can just steal data?
It's okay, we can just have AI decide what's good data and lousy data, before we feed it to the training AI.
(In case it's not obvious - that's not a serious suggestion)
So this is a symptom of the multi-billion dollar gigacorporations skimping on the singular thing that makes or breaks this tech?
Thus leading their output to get consistently shittier as it keeps getting inbred generation after generation? Amazing. A story as old as time is the richest bastard on earth saving money on one thing he should really spend as much money on as possible, and self-destructing as a result.
Can't wait for this bubble to pop and every rich imbecile involved to NOT learn their lesson because once again, no, the most intelligent people on earth do not get the most money. Please, we've learned this lesson 1288567 times, it's time to actually learn from it
All of the points you make in this video are exactly why I’m going to be majoring in Computer Science and AI. I want to be able to learn and create a better use for mainstream AI that isn’t generative. It’s heartbreaking to see how something I’ve been wanting to do for a large part of my life change into a product that (almost) everyone hates. So far, I’m not quite sure how to change the industry, but I have years to plan and figure it out. So please do not lose hope for the future of ai. There are so many students (me!) and people that are working to improve the situation
I spent 15 years working my way up from homless to director at an FI adjacent company. Then they got a new ceo who decided the company would no longer sell products and instead use their industry connections to become a data broker and sell decades worth of data they'd collected on customers. Then they laid off most of us and now I'm unemployed and close to becoming homeless again. Scraping and selling data on people is pure evil in so many ways.
Wish you the best. Modern capitalism is evil
im so sorry, i hope things work out for you
It's an unfortunate natural conclusion to advertisement, marketing and capitalism in general. People are a data point to be manipulated to keep the money flowing.
Damn man I hope things get better for you, now I’m scared for my own future :,)
i wish ai could do my laundry. or my cooking. not fucking trying to replace my job as an artist or give me the worst answers in google search
even as a research tool its p bad.... ik some people will recommend it for job resumes but it j looks better to write it urself
@@greeny.officialyou are so real for that
AI has made most search features completely useless. When they started pushing SEO content I thought it was bad, now with AI it’s a hellscape. It’s so so frustrating for trying to find cooking recipes, and even worse for trying to find academic research articles. I feel like I’m constantly having to fight the algorithms to get to what I actually want, the thing that I *specificaly* asked for. 🥲😒
I mean...you're just describing washing machines, now watch out for the new "AI-powered" ones.
washing machines are a thing dude? its now just also targetting creatives, why should we specifically not target art with ai?
What's funny is to most people, when you watched terminator or jurassic park, you'd walk away thinking "man, that would suck" but it seems like all the people pushing ai and genetic cloning watched those movies and thought "that's literally the world i want to live in"
Yeah, the Torment Nexus meme strikes again... Malcolm's quote about "scientist preoccupied with whether or not they could without thinking if they should" feels very relevant nowadays.
@@Adagamante i swear i saw a video years back about a company that makes clone dogs with glow in the dark toenails, and their end goal was to eventually revive the mammoth species and reintroduce it to the wild lol. As if that wouldn't be the most catastrophic thing to introduce to an ecosystem
@@Countdownsmiles what a reasonable goal, right? As if the current day megafauna isn't endangered across a myriad of species... Focus on cloning those first, you buffoons!
It's almost as if, we learned from the past when technophobes warned us against "playing God", and we went ahead anyway, and developed life-saving medical technology, and world communications networks, and international transportation.
@@AlexReynard Nobody said anything about medicine bro. We're talking about bringing animals from 200 million years ago back to life lol. Learning from the past is not something i would credit our world leaders for lol
I'd say the thing that sets photography and blender type things is that, you take the photo, unlike ai. And the built in models for blender is made for you to use, not stolen from others. It can be changed by you. It can be created by you
Hey gabi! About Spotify's curated "mood music" playlists: The songs there are put together by just a handful of ghost producers under a special contract which pretty much guarantees them a spot on the playlist in exchange for a much lower (about half) royalty payment per stream. Spotify heavily promotes these playlists because if people listen to this music instead of literally anything else, they don't have to pay out as much in royalties. And I guess they prefer it to look like there are diverse artists from all over the world contributing to these playlists rather than just a few dudes in Sweden, so each one of them are pretending to be hundreds of artists. The reason it all sounds so similar is 1. readily available sample loop packs and 2. because they want it to be like that I guess??? (This info is a couple of months old, so maybe some of these ghosts producers have started using AI generated music as a base by now)
I love that at the beginning you tried to use the AI "agent" like a search engine by entering search terms, the exact thing people have stopped doing in search engines, entering questions instead.
And now the AI actually REQUIRES questions to be able to work at all.
Wait wait, is that why google search is so useless? Do I have to enter questions when I want to find things ?
@@Call-me-Al I couldn't tell you, but at some point people just started doing it and I'm assuming Google is being tailored to that.
not really, the AI chatbots will still lock on keywords. You don't need to create a sentence for them to work. See for example the prompts of text or image generators, it's just a long string of words separated by commas
@@Call-me-Al from my personal experience, yeah! I've been finding more and more luck finding results when I ask... which is something that I was actually told NOT to do when I was young, and rather use keywords. Silly.
@@1BergerVongSchlauigkeitHer I blame voice assistants on phones for it. As soon as people could start searching for things by talking to their phone, they seemed to treat it like a person and ask questions instead of just treating it like a search engine.
I just worry that there is no solution. Asking humanity to slow what it thinks is “progress” basically never happens until a catastrophe
Most people won't listen to me because they are so ubiquitous today (especially in America), but the same thing happened with cars. Look up the "Motor vehicle fatality rate in U.S. by year" on Wikipedia, and see the solid red graph. Now, imagine a similar "technological advancement" being rammed through by people that stand to make life changing amounts of money from it, faster than the public is able to really consider the repercussions and come up with ways to make the tech more safe (but unfortunately slightly less convenient/fun, like speed governors for the example of cars), that are then able to sweep a curve like that under the rug of operator error and after long enough, "oh but can you imagine life without it?!" as they actively worked behind the scenes to dismantle all the ways we previously lived without it.
Now, death is an extreme example, but it's also the most likely to get something done about it and *that* didn't even work in that example. In fact, most recently, the numbers are steadily ticking back up after a long period of decline... but I'll save that discussion for another time and get off my soap box for now.
People never considered what we were progressing into, and now look what's happened. AI everywhere
I’m desperately hoping it can either be heavily regulated with laws that protect stolen copyright, environmental impact, or transparency. AI users should have to be required transparently label things as AI the same way UA-camrs have to clearly state that they’re being sponsored or how A&W got in trouble for putting “aged vanilla” on their cans when they don’t use anything of the sort in their drinks.
That, or people find better ways to outsmart AI like that program that makes your art impossible for AI to decipher.
@@Shoulderpads-mcgeeI think AI is very quickly going to run into the problem that it’s burning a ton of money for something that’s just not profitable so they’re going to basically speedrun the enshittification that’s plagued the other major tech platforms. Like, ramping prices way up, degrading models to be cheaper, firing off tons of staff and demanding the remainder replace their workload.
While watching the AI hype men squirm will probably be funny, it’s going to be rough for the average employee of these companies.
@@bree77749 Cars weren't a technology "rammed through." A working EV had already been built before the 20th century. Model T (the first real mass market car) didn't enter production until 1908. It took decades of lobbying and shady business practices to make cars a necessity for so much of American life and it's not like there was no public pushback lol. You can still legally put your horse and buggy on the road right now. Nothing stopping you. Why would these car companies not have lobbied against this? Forcing the Amish to integrate or walk everywhere is a small but untapped market for revenue. I lived down the road from an Amish family for 20 years. Just gotta have safety components like reflectors, something you'd legally be required to have on your bicycle if you went on the highway. It's no secret older cars were more dangerous. And the unfortunate silver bullet to all the arguments against such a car centric society is that enough people just...like their car or rely on it too much to survive now. Most of our issues with cars are not cars themselves but the infrastructure we built for them. The likelihood of dying in any random low-mid speed car accident in a modern vehicle is actually quite low. There's just so many cars on the roads every day that it's a numbers game.
Alternative scale for the energy consumption: 1300 Megawatts is slightly more than the 1.21 Gigawatts Marty McFly needed to travel through time.
I think the standard is to say I don't know when you don't know something. If you are making up an answer rather than saying I don't know then that is called a lie not an hallucination. I wonder if the CTO of Open AI is projecting?
BIG TIME. Like, in what world is that considered acceptable/normal practice? If I gave a customer a made up direction, I would be FIRED.
22:10 the funniest part of companies pretending that they own whatever is posted in their platforms is that this implies they're the proud owners of a whole lot of highly illegal material like you know what involving children that gets posted on their platform
As an artist and privacy advocate i stopped fighting for ai laws when i brought this issue up like 2 years ago. When creeps were using it for this exact reason , and people were still defending it .
like it was no big deal or a joke. Online users will still take it as a joke or meme until it happen to their sister or daughter.
10:04
What will you do if you find out they still don't care even when it does happen to their sister or daughter??
@@seanbell69 from what ive experienced, people usually wake up once the issue reaches their front door, when they're no longer a spectator but another person having to deal with it.
for those who still don't care about the repercussions , what can i do? I cannot change a persons deep rooted morals, it is up to them to develop some form of empathy at that point.
what i had done and will keep doing is reporting instances of people using it for csam and targeted deep faked revenge etc, but when it comes to speaking to people about the Encompassing side effects, i lost the energy
🤢🤮
I think the issue is that people don't understand there are existing defamation laws if something fake is presented as real for one, so it is not like there is no legal recourse that exists at all, bc there is.
The issue however that many don't like is you really don't have any right to have your likeness not ever being used by somebody else without your consent or approval.
Consent was never the legal issue in such cases.
And a lot of freeom of speech exists expressly to give people a right to say or depict things regardless if others approve or consent.
Take a case like Flynt v Farwell where courts decided that you have certain rights to use a person's likeness without their consent provided it cannot be reasonably argued that it is defamation or libel bc it is being presented as factual or real.
I think the difficult thing would be writing laws that do not overly infringe on freedom of speech, even if there are valid cases to be made about non-consensual use of likeness.
It is not that it is a joke, it is that regulations with good intent to target valid objectional things, could wind up being vague and wind up shutting down legitmate forms of freedom of speech.
So it is kinda like how you can't really outlaw photoshop, even though you can use that tool to remove one persons head and paste it onto another persons body.
The ability to do that might give creeps an oprotunity to make images others do not consent to, but the consent is not issue that one should be concerned about in terms of legality imo.
@@memegazer You do have a right to your likeness. It's commonly referred to as a "right of portrait". In the US this is limited to your likeness being used to advertise a product, but in many countries a broader version of this right exists. Here in the Netherlands the right to portrait applies by default, with exceptions being made for example for the purposes of journalism. That is pretty much the same in all of Europe. Americans are often unaware of this, but in much of the world you cannot just film someone without consent, even in public. Also, regarding this specific issue: there are many other laws where a distinction is made when something is "for purposes of sexual gratification". That could easily be used for this issue as well.
"look at my feet" directly after child predators was crazy whiplash
Not to mention the fact that ANYTHING you look up online has ten thousand AI responses up front. Including reference or scientific images that people use for IDing things like animals and plants. Not only is it irritating, but it can be downright DANGEROUS for those that trust those resources for accurate information.
100% and as an artist, makes my job so harder, and makes me waste a shit load of time that would be spent Learning styles, and how, say a a person might look under water. Or just anything to do with the ocean and like 90% WERE FUCKING AI UGh
Yeah, trying to find legitimate historical art or pictures to figure out costume or architectural or prop details is starting to become a complete crapshoot.
@@animatrix1490 i can only imagine! I'm an artist myself, so I tend to Google a bunch of random things for reference material. I can't even begin to think about the kind of horrors you'll find when trying to do homework or do replica pieces. Sounds downright AWFUL
@@EranRaye I was working on a fantasy comic from late 2020 to late 2022 and I wanted to base the visuals on the Republic of Georgia's historical dress, architecture, and culture (ended up with some Turkey, Azerbaijan, etc. in there). During that time, I was able to find quite a bit of material through Google images--some of it also led me to some very cool articles and Reddit threads.
This year in like May and June I wanted to research some historical dress, hairstyles, country manors, etc., from late 1700s or early 1800s France and it was almost impossible to find stuff. I had to narrow and narrow my search terms and ended up finding out it was way easier to just not even bother with the Images tab. Finding articles from actual academic websites and archives (which would often include images with some actual citations or sources) was so much more effective than trying to dig through and differentiate real stuff from all the slop and garbage in the Google image results. *Wikipedia* gave me better material than Google. It really didn't take very long for it to get VERY bad.
good ai: they accidentally made an ai that could recognize cancer cells (was supposed to distinguish different kinds of donuts from each other or something)
bad ai: type words and out pops an image (is trying to replace creatives)
Creatives can't compete with AI, why are people trying to hold it down so rich whiners can continue making more money than all of us?
In the 1800s, painters made arguments against cameras, saying that they were trying to replace creatives. 'Photography is only for people too lazy to learn how to paint'.
Did creativity go extinct with cameras, now that people had increased access to reference images for their art?
@@AlexReynardexcept with cameras you still have to put in effort to get a good angle, lighting, and subject to photograph. you don’t have to do anything with ai except write a sentence and wait for it to spit a picture back at you.
@@seanbell69
What do you mean?, The big people behind AI is also Rich people making more money than us, they're not trying to better the world, they're just trying to replace as many people they can with AI, EXCEPT the ones at the very top, just so they have more money to line in their pockets.
@@AlexReynardbut no one was entering photographs into painting competitions and tricking people into thinking the photo was anything besides a photo. It also wasn’t trying to replace artists. Maybe we didn’t need painters for portraits or for illustrating war any more because photography is a better historical record but it wasn’t being used to replace painting or drawing as an artistic expression. Meanwhile ai seeks to take jobs away from artists like the animated show on Netflix that used ai backgrounds rather than paying an artist to do their job. The camera also wasn’t built by stealing paintings the way ai is built on stealing people’s art.
i love how they created one big plagiarism machine.. and then instantly were shocked that all it can do is just recreate all the stuff it stole
just... worse
There was definitely the idea in mind that humans create art based on things we've seen, so why can't an AI? But human creativity is taking the things we've seen and making them different somehow. AI can't do that, it only knows how to mimick. You can't code creativity.
@@courtykat AI actually does it all the time? You just need a good prompt
@@fresch4395 Yeah, a prompt invented by a human. Tell an AI to draw something that doesn't exist. It'll just give you an amalgamation of other human-made artwork depicting things which don't exist. AI doesn't have the capability to imagine.
@@fresch4395 it doesn't though, if you think that you have been fooled
when it comes to LLM's its just really good predictive text
its the same with music and images
i have been using it every day for the last month in my software engineering job and man does it tell people very confidently to do very dangerous, insecure things and its because its been taught off a bunch of terrible answers from stackover flow... absolutely no checks on whether they are good answers or not
@@fresch4395if the AI didnt have 1000’s of _human made_ data points to draw from it would not be able to create anything. Unlike humans, it doesnt take the time to look at various artworks and perceive them through its own life experiences, preferences and abilities. It has none of those. It can create, but it is by no means creative.
gabi just so you know, the one of will smith eating spaghetti, the “ai now” video is much literally just him recording himself eating spaghetti
As an artist in feature animation, I'm more worried that AI art will be "enough" for most people because they don't know it could be better than if an artist worked on it. If you don't know what is possible, you can't compare and you will be impressed by what AI gives you even though it is, by definition, just a 3-dimensional database giving you results (or pixel colour) by sheer statistics. Even if you have the best works in the dataset, it will just return the top of the bell curve, which, although it can be shifted for slightly better results in the algorithm, is still worse than the best work in the dataset and always, by definition, mediocre.
Even an all-knowing AI will never have lived one perspective or one life, so it will never be able to tell stories or tales to humans. If you have "lived" multiple paths or multiple lives with all perspectives of the internet, books, etc., do you have a perspective at all?
So, if most people are ok with that, I'm more sad that I will not be able to consume good art than I fear of losing my job.
Which is why I'm glad I've started commissioning artists to do background and character art for my stories. It certainly made me appreciate the effort and skill going into each picture.
I've been hiring artists for a game recently and at least from my perspective AI is just not a viable replacement.
Specifically if we talk animation my bet is that after hype is done it will do half the work in realistic 3d animation (and maybe 20% stylized or 2d), but it will do the boring stuff. Retargering clips, cleaning mocap, rigging, decent but absolutely not shippable first pass on text-to-face, etc. It will be one more tool that sometimes can really deliver, sometimes not, but one of the 100 tools needed to do sth right. In my case it provides better translations than we got from humans (yes, humans could do it even better, but our budget was enough for very cheap rushed translation and AI does mediocre for nearly free and if you get native speaker testers to go over it, result is passable for our quality target).
As for mediocre slop, the blackpill is - it's not the fault of AI, or even market economy, but its been unleashed by the customers themselves (or maybe, blind "consumers"). If ppl stop playing or watching slop spigot will be closed down and more budget goes to quality productions. There was not 1 year in this century without quality games/videos, we can collectively keep buying that and paying the ppl making it. Unfortunately as a % the "consumer" spending on slop is increasing, tolerance of "consumers" for shi%ty practices is increasing, this is why it goes the wrong way.
@@Peteruspl Yes, I agree completely. That's exactly my fear that consumers will be okay with mediocre quality. The highest-grossing animated film is still the live-action Lion King, followed by the Inside Out sequel and Frozen 2, so the financial incentive will be pushing sequels and live-action remakes. I see many people unhappy about sequels and live-action remakes (at least in my bubble), but the box office doesn't reflect that, so I'd imagine it will continue... I fear the same could happen with AI art, that it will just be enough for most people.
I have been writing AI tools for years for my own workflow but I find generative AI generally useless for my workflow (except for looking for inspiration). The AI rigging tools also seem to have an actual benefit to animators in terms of performance and quality of deformations but that's also not considered generative AI.
Translation/Transcription seems to be a great benefit to generative AI, though. I have also seen demos where lip-syncs are adjusted by the language which could be quite cool (especially for deaf people - if a movie is well animated deaf people can actually lip-read - though it depends on the stylisation)
I only fear of a future where generative AI will be used as final or just quickly comped or adjusted to make it slightly better for a final product.
This reminds me of a video about it that discussed 1984 for like the quadrillionth time. This time about "newspeak" and how it can be used to limit worldview.
Quite frankly. The idea that people in the future won't be able to conceive anything better than AI is a terrifying thought.
Anyone else feel like search engines used to be able to give way more accurate results with way more vague information compared to today? I feel like i could search a picture or meme with 3 to 4 words and nowadays i have to write an entire sentence and as it gets more descriptive the results just vanish completely.
Yep. And that's obvious: A lot of the same information is thrown on the pile in the course of 4 years, so more shit to sift through.
And then. Companies just take bribes (SEO fees/advertisement dollars) to prioritize that info source, without validation.
Contains: no
Must contain: -meme-
@@scmbgoogle7790😂😂😂
I used to use Google image search in my job as a language teacher. But when I try and search something nowadays, the results are pretty random at times and might include raunchy images for stuff that is not raunchy by default.
Google is useless now. It ignores your keywords (or sometimes even surreptitiously replaces them with more lucrative ones in order to direct you to shopping sites), and the web is so thoroughly polluted with human- and AI-generated slop that exists for no other reason than to game the AdSense program.
I absolutely love Yuumei's art, she keep getting harassed about her art being ai despite her art blogs existing for forever AND despite how she uploads a video of her making just about every piece
When in doubt, whip out the old hand cam. Being able to sync up your IRL footage with your art process is so much harder for AI to replicate.
Yes her art isn’t AI generated, but she has proven to be using AI art generators in her free time. People aren’t mad at her because her art is AI generated, which it isn’t, they are mad at her for blatantly supporting the use of AI art despite advocating for environmental protection. She equates the energy used for generating art as the same as buying supplies and drawing on a computer. As an artist, I find her takes to be incredibly hypocritical.
@@pineapplepizza7489 hmm yeah if that's true, then that is pretty hypocritical
Not even showing speedpaints is enough to prove your art isn't ai anymore sadly - I've seen stuff showing off a new ai tool that can create a fake speedpaint of a piece of art you feed it
@@pineapplepizza7489 Yeah cos my cheap laptop that can run SD is using SO MUCH ENERGY....
This is a very important video. Thanks Gabi for compiling all this research and making it easy to understand
For one of my previous full time jobs as a graphic designer, I made logos for their multiple websites and advertisements, stuff like that. I also took product photography. Then little by little my boss was like “this isn’t going fast enough, let’s use this ai program to filter out all of the backgrounds of these photos” which ended up having pieces of the product cut off, random sections having blurred edges-it looked awful. And then they were like “let’s use this ai to make the logos instead” and “let’s use this handy new ai program to write our content pages and product instructions.” It was disgusting to be a part of
One of my favorite recent instances of AI messing up was on a podcast where they asked ChatGPT "what time is it?", and instead of the AI explaining that it doesn't have the resources to give a time accurate to their location, the AI just said a random time. And the time it gave didn't even have the correct minute.
It doesn't even have the resources to get the current minute. It's advanced predictive text. So there is no way for it to do that.
wait a second please, i know first hand
kuddos to you hopefully she can add you to credits, or put you as a link IDK I just enjoy her videos
@@scraft1222 its a bot dw
@VilleUTTP your account is only 6 days old and you don't have any content at all, why are you spamming this schizo lie?
Happened in Brazil: Volkswagen used AI to make a commercial using the image of a dead singer. Even though the daughter gave the "ok" for the image usage, the fans argue that she would NEVER do a commercial for a company she does not share her values and vision with.
Imagine having your image used after your death for something you disagree in life?
even more so taking into account Volkswagen's collaboration with the military dictatorship, as the singer was totally against their oppression
Ah, eu lembro disso
Foi recente isso
É quando usaram a Elis Regina pra fazer comercial de carro? Achei bizarro e bem desrespeitoso isso mesmo
It's you repeatedly walking in on screen like a glitch but continuing from where you left off verbally so smoothly lmfao was messing with my brain for a second. Honestly this AI shit is just so ridiculous, and sad for legitimate artist and striving artist. Thank you for sharing your pov ❤
One other argument I've seen brought up is that all of these companies are also using data from dead people, who by definition could not agree to these updated terms of service and legally cannot have their content being used in training models (and I imagine this also works for anyone who's abandoned a platform w/o deleting all their stuff, ie most deviantArt accounts)
Trueeeeee tho
I hate being an artist and feeling like company's are trying to replace us to save money, as if we don't need money to live and eat,
You could always steal money.
@@Cubeytheawesome you're not wrong
Tbf (as an illustrator myself), humans have been trying to dramatically underpay and undervalue (or not pay) artists for decades. This unfortunately just feels like the end point for humans wanting all the things for as cheap as possible. Not exclusive to, but definitely including art and music. It sucks.
or buy their stuff 😭 that’s what i really don’t understand
You sound like conservatives yelling about how unskilled migrants are taking your jobs.
If a machine can make art better than you, maybe the problem isn't the machine.
My graduation thesis was about Gerative/Generative A.I in mid 2023, we can't forget that some of the initial model training for a certain company was made using very very cheap labor from other countries and was prejudicial to the worker's mental health since they were exposed to very disturbing prompts during model training, because humans had to flag and filter prompts/images/results as "ok" or as not allowed/nsfw/breaking TOU...
WHO!!
Isn't a lot of AI also racist and mysoginist...I mean look at the UA-camrs that got their videos "stolen" Gabbi brought up - not a single woman.
@@Unrayorojo Times magazine's arcticles " 150 African Workers for ChatGPT, TikTok and Facebook Vote to
Unionize at Landmark Nairobi Meeting" Perrigo, 2023 and "OpenAI Used Kenyan Workers on Less Than $2 Per Hour
to Make ChatGPT Less Toxic" Perrigo, 2023. These may give you a hint XD (can't post links since comments usually dont allow it!)
@@moongirl8807 At the time of writing my thesis we didn't see that point in specific, but i wouldn't doubt.
One topic my thesis partner and I came across was the sexualization of people using those "make A.I portraits with your pictures" which i'm sure most people used at some point in 2022-2023, some which had a lot of reports of doing that to women, and most of the arcticles i've found were about asian people, either the models made sexualized/fetishized images or got their nationality wrong, like accentuating east asian features on southeast asian people, and so on...
@@moongirl8807 About what i've just posted:
technologyreview's arcticle "Lensa Ai undressed me without consent"
and also analyticsinsight arcticle "Lensas fetish for asian women shows sexist standards", and the last arcticle is catntran at medium, titled "Does lensa ai thinks all asian people look the same"
I wholeheartedly agree, as someone in cs I often see people humanizing AI or getting really wrong infos, but this video is just right.
Corporate greed is always bad, there's no denying that, but I also think that AI is just the "new big thing" that will fizzle out of the spotlight soon enough, see cryptos, nfts, and so on, because like you said, AI has been around for a while now, it's nothing new
The Anti AI movement is taking form. And I’m all here for it
thanks for going over all the grey areas involved in distinguishing "generative" AI "art" from using AI as a tool for the artistic process and other tedious processes.
It really is depressing that whenever I want to search an image it has come to this, I tried to find some references for practicing on drawing anime and like it's one thing if it was even decent, but I saw like loads of just bad AI anime, there was one I was search "anime with braids" and it was so revolting to see the girl's hair had seeped into her body creating hair for hands, it was super scary
You got to limit it to only images before 2020 or so. Earlier is better to ensure it's not ai
generally, for drawing references for anime, you want to look at image repositories that let you add your own tags for filtering (pixiv, safebooru etc) - most of them are ahead of the game and let you blacklist the "ai-generated" tag, which can help a lot in finding actually useful stuff.
@@Rita-kx3yr thanks for the tip, maybe I'll try that next time 😅
THANK YOU as an artist, I really appreciate what you said about ai generated images
PREACH Gen AI is horrible in every way, real AI is something we need for dangerous jobs, not to replace our creativity and enjoyment of life
Bro we had so much potential to have robots make our life easier, but now they make it so much worse
@@minhuang8848 okay but let’s just take your point about color palettes and fashion what happens when a company like H&M or American Eagle decides that using AI is easier and cheaper to make their new lines with AI rather than paying fashion designers to do so it’s not really a problem on a small scale but the problem lies with companies trying to save a buck off the backs of people who have already done the work rather than crediting and paying those people
@@minhuang8848if you use AI in any capacity for creativity, you’re already failing at life and should be shamed
@minhuang8848 Booo, 'let's have machines do all the artwork and games' Real AI actually can help, gen AI only steals and is done for a way to cut jobs and costs, as well as use too much energy to damage the environment. Watch the video.
@@Waldo-in-walls Roombas are great. Whatever companies are trying to push, not so much
The worst part of Windows Recall is that they are basically forcing everyone to upgrade to it. They plan to stop all support services for windows 10 in october of next year, which makes using it dangerous and makes you vulnerable to online threats.
I think Windows Recall only works for those dumb laptops with built in NPUs, no?
Right! I own a windows 10 pc that my mom had gotten from her job for us to use at home. She doesn’t use the pc at all so I have it. Mind you we can’t afford to buy a whole new one, like it took ages js to get this one, and now I keep getting notifications like “oh upgrade ur pc cause it’ll stop working soon” like… oh you wanna make MY pc stop working even tho it’s already ass? Okay, then YOU buy me a new one cause wtf 💀
Plus what’s worse, is that I play the sims, and it is much much easier to play on windows than it is on a Mac so that’s great 🥰
And that's why I moved to Linux for the first time ever.
@@albert2006xpIt is for now, but ppl have been able to get it to work on virtual machines without the processor, so that seemed to be more of a marketing restriction than a hardware limitation.
Knowing microsoft, theyll just push it to all possible computers eventually with little to no notice.
As a future game dev if I have to see one more AI person talking about how they are going to make art or a whole video game with AI I'm going to jump off a cliff. Its supposed to be about the journey not the end result. I've been learning art with blender and coding for unity from scratch to build quality games it's insane
I guess you can delete your whole project and start a new one without any remorse and still ask people to pay for the fact that you WHERE delevoping a game, yes? After all, who cares about the end results, its only about developer's journey, not the end product for people to enjoy...
@@protoney8122 I don’t think you have enough knowledge in what’s being discussed to participate meaningfully.
Congrats on learning art! You could probably also reach out to newer artists who are looking to practice new skills to collaborate or if you have any budget commission
@@clare5688 yeah! I've been thinking of doing so definitely something that may happen in the near future. Also thank you for the kind words