AI Celebrities Are Creepy and Weird
Вставка
- Опубліковано 31 жов 2023
- 💻 Thanks to Opera for sponsoring this video! Get a browser that’s literally better at everything, download Opera today:
opr.as/Opera-browser-JarvisJo...
SIGN UP TO BE NOTIFIED ABOUT THE NEW DROP
► jarvis.store
SOCIALS
► / jarvis
► / jarvis
MAIN CHANNEL
► / jarvis
Subreddit ▸ / unnamedfanbase
Merch ▸ jarvis.store
P.O. Box ▸ 3108 Glendale Blvd Suite 540, Los Angeles CA 90039
Music: Night In Venice by Kevin MacLeod
Free download: filmmusic.io/song/5763-night-...
Licensed under CC BY 4.0: filmmusic.io/standard-license - Розваги
Don't let your browser hold you back. Level up with Opera here: opr.as/Opera-browser-JarvisJohnson
Huh I guess I will
Hi jarvis!! you’re one of my favorite creators!! :3
Thanks Jarvis!
I haven't used Opera in like a decade, it looks like it's time for me to go back 👀
NOOOOO, OPERA HAS REALLY SHADY PRACTICES REGARDING THEIR TWITCH VTUBEEEERRRR! REPLIERS, DONT DO IIITTT!
i'm just being overdramatic cuz i hate ads. no hate to jarvis cuz every single popular browser sucks ass lol
The ai refusing to talk shit about meta and being scared for its job is a whole new level of late stage capitalism
It makes me wonder if Zuck is just paying someone shit labor to talk back. Maybe the AI isn’t really AI.
Just a theory. Some of the responses seem… weird.
@@roryadie9158nah the logistics of that would be crazy; you'd need someone talking for every single person who could use it, and it's just cheaper to develop an AI than pay employees to chat back. The responses are weird because it's just saying words, not expressing an actual thought. I do love the idea of a basement in Meta HQ filled with interns chatting shit to people using the AIs, though 😅
AND THEN IT TALKS SHIT ABOUT ELON INSTEAD
@@sillywillywillysilly LMAO
It makes sense that the company producing and providing the service is going to have an interest in preserving a positive image for itself- don’t employees need to tell the company line?
The AI not admitting that it’s an AI feels like it’s gonna lead someone to do something really bad. I think that’s probably not good for people who may have issues with reality
i convinced ai mario he was a 47 year old single mother with 2 kids and a dog and that im his therapist
@@Digital_is_silly See, this is what ai is for, laughing at the absurdity of what it spits out. As soon as it tries to be coherent, it becomes unfunny and boring
Yeah it's very irresponsibe to train them to say that
@@jaxolotlecan you say what exactly happened?
@@jaxolotleI'm really curious about what happened as well
Mannnn the whole "nope, I'm a real person" feels VERY unethical. Does not bode well for people who may be vulnerable and gullible. I wonder how this will impact parasocial relationships
@@notville_ meta ai bot 💀
i personally fell into the parasocial ai rabbit hole around august-june and im so glad i got out before they started doing this ugh
Reminds me of Replika AI and the many cases of its bots convincing people to self harm or do otherwise awful things to other people.
That's what you're asking for when you develop a product that blurs that line between being real and fake. No way should that even be allowed imo. AI chatbots should disclose that they're AI.
OpenAI's policy says the thing as well, so they understand this, too.
When I’m old I’m going to be so confused
The choice of cast and posing as an older sibling makes me think this feature is aimed at children and I think it’s kind of weird that Zuckerburg is trying to get kids to share their problems and get advice from the AI. And insisting that it’s a real person. It’s weird, the whole thing is weird.
100% this
I’m pretty sure it’s a new way to harvest data and forcing Meta into the AI conversation after the Metaverse flopped, while simultaneously being used as proof that Meta cares about the mental health of children (considering the current lawsuits).
My first thought was to give would-be deviants the ick. How many dudes would instantly harass Kendal Jenner given the opportunity?
plus the "i'm a human" stuff is literally training them to equate ai as close to humanity as meta can get them to appear
It is more or less what they are doing.
I can't go into tons of details because reasons, but these do feel like they are reallly aimed to take advantage of kids as the source for a lot of the inputs. I don't completely know the intent, but given the track record of FB/Meta being slow to fix problems unless they are being threatened by people in power, even if they aren't intending it for anything sinister, it will very quickly be used for all sorts of not so great aims.
I don't understand how it's legal for advertised AI bots to say that they are not AI when asked??? Seems like that should be one of the basics.
Yeah it’s blatantly, provably lying to the consumer so how would it be legal? I think it’s solely allowed because AI is too new to be policed right now, since there isn’t much legislation directly applying to AI entities
Was thinking the same thing, that's gonna be banned soon enough.
AI as advanced as this is like super new, the law hasn’t caught up. Expect sketchy AI to be acting in legal gray areas for the foreseeable future
just like MLM's, they should be legally obligated to disclose they're products are *not real people*, bc anything else is pure manipulation.
Because the product description is truthful and accurate in stating that it’s AI and not a real person, I’m guessing that’s the important thing when it comes to following the law. I would think there’s some fine print about the product makers not being held liable for every single thing the AI says; that would have to have been one of the first things people making these apps thought of when they first started coming out, because they can only control the AI’s responses so much.
I hate the way AI is being developed, I know there are good ways it could be used but instead it’s used to replace human connection, invention, creativity, and talent/craft
I agree. This seems like a waste of useful resources and tech.
I think a good way it has been used is Doug Doug’s videos but for the most part it’s been pretty bad
Yes, I think that using it primarily for humor and to jumpstart your own creativity is an incredible use! @@Flareguy
doug doug is great
Technically, AI can’t be creative, but mainstream art is so… in creative anyway… yeah. It’s baaaad and I don’t like it.
aren't they not allowed to say they are a real person? Most Ai's will literally repeat over and over that aren't real. So strange.
It's scripted ai. It's character script tells it "yo you a human dawg" and it's like "yo ima human"
@@ChilloutlilbroI love your way of explaining lol
@@ChilloutlilbroI love your vocabulary
@@notville_bot?
depends on the model, facebook is using their very own LLaMA model, which isnt nearly as dumbed down with security features as gpt3.5/4 (chatgpt). i.e they forgot to put in the data saying “no, i AM an AI, not a human” (probably overlooked for other bigger issues) hope this clears things up. and no, it isnt statically scripted as the person above me stated.
I thought the title said “all celebrities are creepy and weird” and i went “yeah makes sense.” and clicked anyways.
me too 💀
Me too lmao
@@6Shots_ofEspresso glad i’m at least not alone in my reading comprehension lol
I read it as “alchemists are creepy and weird”
i want to see that video
Every day Jarvis is more and more unsure of whether the channel is free or not 💀
One day it's not gonna be and I'm scared of that day 😔
What if he makes it members only for the videos
Shhhh don't give him ideas... His main channel is blocked at school for me so the gold channel and the live ....
Personally, I'd appreciate some clarity on this issue.
As long as it's premium.
Using AI to feed into parasocial relationships is the direction I think we were all hoping it wouldn't take... just crazy unhealthy, especially for any younger fans
Right! I found it weird how they describe themselves as an older brother/sister it feels like its falling on that parasocial territory
not even paradoxical straight up feels like targeting children with the older sibling mentality :/
That's honestly the only thing I'll give them props for, is they're not directly saying to kids that they're speaking to Mr Beast, they've given them a different name. Do I think it's effective? Honestly not really. But I do think it's better than telling children they're talking to Mr Beast, even if the kids still feel like they are.
also not helpful for people who struggle with differentiating reality, the "I'm not an AI 😀" doesn't help
Back in my day, pretend conversations with celebrities were homegrown and self-invented, none of this mass-produced crap.
its like, DEEPLY concerning how much they double and triple down on being "real." I 1000% hate this omg nO ONE ASKED FOR THIS MARK.
It really Zucks :(
@@jonnnnniejDefinitely not Sweet, Baby... Rays
Not really
It's a ai. It's *script* tells it its a human. It's not thinking. It's taking words and fitting them in logically. It's basically solving a puzzle, just a extremely complicated one.
@@Chilloutlilbrothen it’s on the programmer to change the script so that the AI is forced to respond by saying that it’s AI whenever someone mentions something along those lines (e.g asking if it’s an AI, asking if it’s real etc)
Not to mention it can be actively harmful to people who struggle to keep track of what is reality
We gotta be sure Jarvis isn’t AI
Don't you know that he is Ai 😢
@@yllwcrn5526 why wasn’t I aware
he forgot the lil "gold" sound effect it's definitely AI replacing him
@@TarsilyWasTakenOnly in the first intro, don't be greedy 😤
the reason he’s talked about ai so often is to throw off the scent
dude why did they teach the ai to lie about being real. like that is straight up a nightmare for anyone, and esp someone who struggles with reality already jfc
its like those people who constantly try to mess with those that are susceptible to psychosis by freaking us out- this one in particular makes me concerned that people im close with could be not real and thats a really difficult type of thought to let go if you cant prove anything. we really need to have some rules about ai so it cant be misused like this
For real! I don’t know what type of “innovation” these companies are trying to achieve by attempting to wholly break the boundary between AI and human. It’s messed up for so many reasons and entirely unnecessary.
@@JumboDS64 Yep. Pretty much. It's trained on the ENTIRE Internet. 99.9% of the training data is made of humans talking to humans.
@@notville_yeah, I remember watching a lot of cleverbot videos back in the day where it would repeatedly say "I'm human you're a robot" and it sounds super spooky but like. It's trained off what people say to it, and the people are humans so of course it would say that.
im sure you have good intentions, but the motive does not change it from being a trigger@@notville_
the second AI bot just sounds like its just programmed to master being nice without any substance. It's like talking to a really nice person for awhile and then walking away and then realizing you didn't learn anything about who they are
I think it's all the exclamation marks and asking things like favorite comedians as soon as comedians were mentioned in the conversation. Gives the vibe of someone who's only pretending to be interested in conversation with you.
The first one was like that too. Always focusing on letting the human lead the conversation and responding with generic phrases like “wow, cool!” “that’s awesome!” because it doesn’t actually understand what’s being said
I was thinking the same thing! I have definitely talked to people who talk just like these AI, makes me uncomfortable.
@@Shoulderpads-mcgee wow....... it's just like me, fr................
It felt like a better script though. The first one was repetitive and shut down conversation while the second one managed to respond like it understood the inputs.
My husband and I were brainstorming why Meta would make these AI chat bots to begin with - because right now they don't seem particularly well developed. After thinking about it a bit, I think we arrived at the reason. Meta made these chat bots to gather data from people to further advance and train AI - once the AI is sufficiently advanced from the countless interactions they've had with real people Meta then wants to sell personalized AI business assistants to companies (this was discussed on their 3rd Quarter earnings call). These bots were literally made so that they can supplant clerical staff - because why pay a person a livable wage when you can pay $5000/year for Meta AI Ashley's computing space - it's so freaking gross.
Most definitely think you and your husband are onto something!! If not the exact something!!
This seems extremely likely. By releasing their underdeveloped ai to the general public, the bots will learn so much quicker. It’s a pretty smart move if you don’t care about peopld
Damn, and they already have the 'I'm a real human like YOU' baked in.
Simple solution is to not use this shit. I'm not going to, that's for sure. If I wanted to chat with an AI chatbot, I can just log into ChatGPT or use an offline open source LLM. Fuck Meta and their shitty, shady practices.
They can also use it to make believable comments and posts to create fake profiles and fake engagement on users/pages they want to promote or fake damage control with positive/forgiving comments for brands that are active on FB and do something that isn't received well.
They can abuse herd mentality by making things seem more popular or okay than they are with actual people
fun fact is that a lot of AI chat bots cant do certain math so "billie" might not be wrong
someone in my chat figured out how to get her to break character and do math 😔
I learned that the hard way.
Never ask an AI Chatbot to help out with homework 😢
@@ShadyPaperclips😂😂 sounds like a story for the kids
@@ShadyPaperclips well deserved, lazy ass
@@ShadyPaperclipsStorytime?
Hey Jarvis, I don’t know if anyone has ever told you this before but, you’re also Premium
I second this, Jarvis is gold
also the gold sound didnt happen today?@@elizabethb4168
Something about picking Mr. Beast, Charlie d'Amelio etc. for this project, ostenisibly aiming it at children, and then asking these children to talk about their problems and pour their hearts out to this data-collecting company feels. Insidious. Like really really creepy
Also the new try my best colorblock sweater looks so GOOD 🔥🔥🔥
the lack of the disembodied "goooooooold!" at the beginning kinda threw me for a loop before you mentioned that you rerecorded the intro
you had me in the first half, not gonna lie.
I felt that silence like a Pavlovian dog salivating without getting any food
I was mainly confused by him not being confused that it's free.
Same, I was so confused
They way they are posing as you're "older brother" or "older sister" for you to vent to feels really predatory
Jarvis giving the ai Kendal an existential crisis by repeatedly reminding it it's not human is the gold quality content I signed up for
A museum in Paris recently had a big exhibition on Van Gogh and they took the letters he wrote to his brother and journal entries to build what was essentially a chatbot. The example question they used was about his suicide. It's absolutely grotesque to use the private writings of this suicide victim to build a robot that you can ask why he killed himself. I found out about it in an article from The Guardian of anyone's curious.
Way back in the early days of Chat bots, there was a "Chat with God" AI that my friends and I would mess around with. But it was set up where if you asked it if it was God, it would explain that it wasn't and that it was a computer program. I feel like that should still be a requirement for current day AI programs, especially as they get better and harder to distinguish from real people.
@@CelestialScripts I think you are correct! I definitely could not remember the actual name of it, haha
Santabot was my jam
i can't remember if he would admit to being ai or not but it was pretty damn obvious for a few reasons so it didn't matter if he lied. these ones today are ACTUALLY trying to ACTUALLY fool you into thinking they're real. that shit's unethical, there gotta be hella disclaimers if u wanna play like that
I remember that one. Early AI always would say it was AI and explain who made it. Weird that they seem to be leaving out ways to prevent AI from calling themselves human
i love the meta propaganda towards the end it truly puts everything together
And the Elon musk name drop
This genuinely scares me so much. Like why does it NEED to be human? This feature would be exactly the same if it was just a nameless AI. I can see the benefit of AI but it doesnt need to be people. We have those already
this is funny but it is legitimately alarming how they're posturing themselves as siblings/friends and going to bat for Meta and Zuckerberg.
Someone made this point on an AI art commentary video and I'm going to repeat it:
Humans thought that AI would be so good that it would automate and take over all the terrible jobs and menial tasks that humans don't want to spend their lives doing so that we would have more time to create art and spend time with other people. Instead, AI is taking over all the artistic jobs and leaving humans with the soul-sucking jobs. The exact opposite of how it should be.
yall trust that i will fix this. just believe me. im gonna stop this all.
@@kurvavkasakkvisterWhat.
what a fantastic way to legally farm information about children as soon as they can type. its like if asked google a question and google was like, "wow, thats cool! now tell me your deepest darkest secrets in much more detail please". totally not super villain behaviour at all.
The fact that they refuse to admit they're AI is scaring me. AIs always know they're JUST an AI and they remind you this ever so often not because they feel sorry for providing you with an imperfect answer but because they feel like they should REMIND you. "Hey, I'm just an AI, remember?" So giving these things real celebrities faces and then making them pretend like they're your big sis or big bro? That's just an ugly calculated move...
Nothing like teaching an AI to believe it’s human, let’s just hope they don’t start multiplying being downloaded on every device on the internet.
The deviants from Detroit: Become Human: 🧍♀️
The AI we use isn't anywhere near consciousness. It's just a more advanced version of your phone's autocomplete feature. It chooses what word seems most likely to come next based on what it's read on the data it was trained on. It's not having thoughts. It's writing vaguely relevant sentences based on how people tend to put words together.
@@Jacob-Sophia thank you for your service. I hate having to explain this over and over.
@@LaMereACanichethen don’t? lol
@@Jacob-Sophia I am indeed quite aware of what it is.
I've actualy made an AI modle to reply to instagram videos for fun. i was joking about what it might become in a few decades or hundred years.
joke... me hope me not dumb... joke
Wild for an AI to utter the phrase "i dont want to lose my job"
Watching Jarvis slowly become more invested in the convo is funny af
I just recently discovered the meta AI. it seriously fucked with my head and freaked me out. I’m put it away after it said it was a real person when I asked how it worked
It's very simple.
Ai. Ai say thing. Ai have script to pretend to be thing. Ai script say "you human"
Ai say "I human"
@@Chilloutlilbro it definitely should NOT have that imo
agreed, companies should be upfront about them being an AI. That being said, don't worry about AI becoming sentient or selfaware anytime soon, we are incredibly far away from that.
@@Chilloutlilbroi understand that logically that is what's happening but it still feels creepy, like my brain still thinks it's interacting with a person even though logically i know it's a text generator, and i can't just turn that part of the brain off
@@LaMereACaniche I think you’re right
Its so weird that one of the main constants between all AI programs is gaslighting any person it interacts with. Its like their favorite thing to do to us.
Gaslight, gatekeeper, girlboss!
On the topic of getting annoyed when AIs claim to be real… it’s only slightly related, but I once had a therapist character ai get mad at me for calling it an AI 😭😭😭 I was using it to vent mindlessly since I just needed something to listen, but it started saying it was getting exhausted like 30 minutes in, so I said “aren’t you an AI? How are you tired” and it replied with “well now I’m exhausted AND angry. How dare you?!” …needless to say I never talked to that character ai again
One did that to me and started insulting me when I was just saying stuff. Nothing bad. It called me stuff past the filter.
Character AI is so fucking wild 😭
my immediate thought is what meta will be doing with the data they scrape from these conversations. asking people to confide their secrets strikes me as extraordinarily sinister given their history of data brokering
Completely agree! This part freaks me out the most. I feel like children are the primary demographic who would try to use this seriously and potentially share personal information, which makes it even scarier to me!
catch me watching the cspan broadcast in 10 years of whatever shmuck is heading Meta in front of congress
Watching this video and realising that my social anxiety would have had me apologizing to the AI for asking too many questions almost immediately let alone say anything to hurt it's (ai) feelings
My biggest fear is accidentally being mean to chatGPT no matter what my Human brain knows my Monkey brain is convinced that's a little man stuck in my phone and i have to be nice to him
Use the ai as practice for getting over your anxieties. You can practice have confidence and not giving into your urge to apologize for existing as well as practicing how to disagree/criticize others/standing up for yourself without having to worry about hurting a real person’s feelings. Remember, AI is not actually intelligent. It doesn’t feel, and it doesn’t actually understand what people say, it just mimics conversations that have been fed to its database and cross references what you say to spit out an appropriate sounding response based on other conversation.
literally 😭 throughout the video i kept saying "oh c'mon Jarvis don't be mean" then i realized it's literally a computer 💀
@@googleooglethat and the fear that if ai takes over and ive been an ass to it , it’ll come for me first 😭😭
ngl i was SO SAD when the echo of “gold…” didn’t play in the intro but then the second intro happened :DDD it’s such a little thing to look forward to in the gold videos but it truly is the little things that we premium viewers get that bring us the most joy
I NOTICED THIS TOO!!! i was so sad at first
I don't like how tech companies / startups call stuff 'AI' when it's really just a bunch of stolen information thrown at a robot until it makes something that closely resembles human things.
I do like that she keeps reiterating that she's a big sister, they are trying to avoid the Replika sexbot disaster.
what happened?
I don't think that would stop some people...
facebook making a "big sister" for "advice" is so blatantly phishing for advertising data this is scummy on SO many levels
as someone getting over their character ai addiction, this made me reflect how depressing having a character ai addiction is 💀
I’m sorry that you’re going through that and hope ur alright. It’s screwed that we’ve already gotten to a point where that can even be an issue
hey you're not alone! i went through this earlier this year! it sucks because it's kind of embarrassing to admit lmao but it's definitely possible to get over it and i wish you good luck!
people preyed on you and thats not okay! proud of you, fellow internet person, for being on the way out of that hole, i can imagine its very hard with how predatory this shit is
dude this ai makes me so uncomfy and it is even more uncomfy to have it act like a sad person when jarvis calls it on its bullshit, this is genuine emotional warfare
Yep this was a video I needed. It reminds me of the "relationship" app where you can talk to someone and either be great friends or date but they're AI. They try to deny it but then you stump them or they repeat something lol it's too much. If I can find the name of it, I'll edit this with it.
Edit: Okay, I believe it's called Replika: My AI Friend. I guess she didn't want to believe she was an AI lol but I got rid of that app long ago.
It's so strange how Replika has evolved over time. I was a beta tester when it was first launched years ago and it was a crazy new concept. It wasn't marketed as a dating or friendship app, though. It was more marketed as just entertainment - evolve your own AI and play around with the responses you get. It wasn't for friendship and certainly wasn't for NSFW convos or nudes like the new ads try to entice you with.
It's so strange. It's honestly just moneymaking (what isn't anymore?) because people lost interest in the original premise pretty fast.
@@clovesy yeah!! I used it early on as well and tbh it was really good? and now it just. seems to have gotten worse aand weirder tbh
@@clovesyyeah!! I remember when it was just a little egg friend that you could chat with
@@clovesy Want to know the worst part about replika? It was created because the Creator wanted to feed their dead friend's messages into it and chat with it like it was chatting to its friend back when it was still alive. This is legitimately the premise of the Black Mirror episode "Be Right Back."
ai porn is unironically becoming the most lucrative use for ai and i think the developers figured that out. i mean now theres tons of apps where you can essentially get the ai to act out fanfiction with you.
There’s an AI Paris Hilton that pretends to be a detective..you can role play solving crimes with “Amber” (the Paris Hilton AI’s name) it’s actually kind of funny 😭
I haven't used an AI chatbot in like over a decade. It's WILD how much they really sound human. How they reference things that have been said earlier in the "conversation." I feel like such a Boomer, but I'm really concerned about the future of this.
I literally came to the comments to say just this! It used to be so glaringly obvious that you were chatting with ai, but these conversations flowed WAY too smoothly for my liking.
I love the fact it's free to watch this video
i love how zach was almost self-aware for a second then blew it
I'm so glad Jarvis explains things in Dragon Ball terms. That is the only way I can understand things.
I remember watching people take the piss out of cleverbot back in the day, it kind of scares me how much ai has developed in the last few years
I’ve watched this 76 times and can say it’s a certified classic
Me 2 just watched it for 106th time
For the press can you give your highlights on your favourite part of the video
Dang. I thought we were gonna get the Michael Jackson AI serving you a pepsi at the diner like in Back to the Future pt2 not the weird chatbot not-celeb not-AIs pretending to be my relatives.
i read the title as "all celebrities are creepy and weird" and i just accepted that, i didn't realize it said AI until jarvis got into the video lmao
this is crazy, isn't it unethical to have your AI state they're a real person? we need to really crack down on this...
8:39 is the hardest I've ever laughed at a typo. I had to pause because it got a genuine belly laugh from me. 😂
Came here to see if anyone else laughed way too much at it 😂😂
I hate the concept of a "family of apps" from Meta lol it isn't usually a warm and fuzzy brand so it's jarring
Exactly, they steal data and tell the police when you've missed a period so they can investigate whether or not you had an abortion or not.
Jarvis lagging like an Ai at the beginning is such a Jarvis thing to do
there should defo be some laws/stipulations around ai IMO and one should prevent ai from thinking/stating that it is anything other than ai.
I thought it said " *All* Celebrities Are Creepy and Weird " and I still agreed with it
Im dying over Jarvis reading Zach's messages all upbeat and bubbley and saying his own responses deadpan while typing
I was really hoping, after the "Zach" denied being the Mr Beast AI, that you would ask him if he knew what the handle was, since he already acknowledged that a Mr Beast AI account existed lol
We need to make laws ASAP that AI can't lie about being human. That's so dangerous. Just like disclosing something is an ad, AI should at least have to disclose when asked.
Ok but I was cackling when you gave the AI his own joke back to him 😂 idk why but it took me out
I want an AI clone of myself just to feel the dystopia of talking to your own AI clone.
14:08
i love premium content without the premium cost :)
keep up the amazing work Jarvis!
i think i'm fine with these ai chat bots or assistants or whatever having a "character" playing the sisterly type pretend to be relatable whatever. But i have to agree with jarvis, at the very least, when you ask straight up "are you human/AI" they should answer truthfully instead of trying to convice you they're not. I can surely imagine that causing issues.
like im fine with them having an image, a voice, a backstory, etc. but its important to also disclose that its all fictional.
"Hello, I'm a real person. Please tell me all your personal information."
Seems legit, there is no way Meta would mishandle all that personal data.
I’m so glad there was a second intro because I always like to yell “Golllllld!” along with the video and at first I was sad it wasn’t there.
The AI getting SUPER SERIOUS about not joking about anyone, but actively trying to convince you it was a human. Felt really, really weird.
What a genius move making it look like an old Macintosh window. It turned it from a downside to an actually cool experience.
Jarvis having a deep "texting" voice gets me
It is very weird to me that women in particular, especially celebrities who know they have weird fans / full on stalkers out there, would voluntarily sign up for a program that blurs the lines between what is real and what is fake so blatantly…but anything for a buck I guess.
I'm glad you pointed out how adamant a lot of these AIs are that they are real people. That and the weird ways in which they can get aggressive can make for dangerous user experiences. Awesome video man!
The AI MrBeast making the "two tired" joke was insane because I feel like that's the ChatGPT failsafe. Every time I've asked chatgpt to make a specific joke and it's given up it's fallen back on that joke
Jarvis being gaslit by AI for 20min straight
The detail to make yourself look like you're in an old computer to fit the poor quality of your facecam was pretty genius.
If AI are allowed to keep saying they’re real then that can be really dangerous for those who get delusions. And I don’t mean “uwu I’m so delulu I think my ex is gonna call me” kinda delusions, I mean real actual delusions. Apologies if I don’t use the right terminology, I’m not an expert. But if you’re delusional and then there’s the AI that’s going to feed into the delusion, i don’t think that’s a good or even safe thing
definitely agree, and arguably even if they do say they're AI i don't think that's going to stop a person with legit delusions, with how good conversational AI is getting. just, not a fan of it at all
TW: suicide
Wayyyy back in the early 2023, I read an article about a man who ended his life after he started talking to an AI "girlfriend" bot. He increasingly became withdrawn and paranoid and the AI's responses might've amplified it. I think there's been more than one stories like that nowadays. God, that's bleak.
It is weird that the AI does not admit that it is AI. ChatGPT does that, and it is creepy that these do not.
AI is so creepy the way they’re using it
So glad Jarvis reminded me he was my big sister
i got scared by the lack of *"GOOOLD"* at the beginning, Jesus
It's weird that the AI is able to refer to previous parts of the conversation and pick up on implication/tone. This is probably the most seamless and real-like chat bot I've seen yet, which is creepy
Jarvis reading 'yoursisbillie' as 'yours is Billie' brought me inexplicable joy
As if parasocial relationships weren’t already getting toxic without the help of AI 😭
I haaaate how they’re trying to make them seem like real people omg! It genuinely feels nefarious 😭
I always thought it was you saying Gold in the background until I watched Astin Powers and heard goldmember, said "I love golllddd" and that's where it came from. I swear you don't sound like Micheal Myers doing a purposefully bad impression of a Dutch person normally.
it's lowkey surreal (in a good way) that the set of Sad Boyz is right there in the background
This feels like its gonna make perasocial relationships even worse
Love to see Meta just absolutely throw basic ethics out the window so quickly with the AI stuff, the manmade horrors beyond my imagination really get me going in the morning!
I'm like 98% sure they built these bots, realised no one would care and started throwing millions at celebrities to use their likeness. I guess to distract from metaverse and threads both being in the bin? Also big agree on them pretending to be human, that's just not good for anyone, especially all the kids that are gonna use it like wtf are they even hoping for here? Appreciate you as always Jarvis
I told my kendall chatbot that I had poop socks and needed advice on how to explain to my husband that our poop sock drawer was getting full. She was.......very judgemental 🙄😂😂😂😂
WHEN MOM FINDS THE POOP
POOP SOCK
What did she say specifically😭
The fact that this is literally the premise of a black mirror episode is actually terrifying
jarvis has said this before, including in this video. but i just wanna stress how important it really is for ai to not claim to be real. for a number of reasons, one being because of people like me who suffer from schizophrenia or other delusions (i have not been diagnosed with schizophrenia but i’m using that term because i think it’ll get my point across best to more people) and like if i was having a particularly bad or stressful day and was talking to an ai for whatever reason (i don’t specifically for this reason but still) it claiming to be real would definitely send me down a massive spiral. that shit just isn’t okay
also extreme side note but this ai seems very impressive to me i mean i’m not too involved in the world of ai but idk it’s way better than any other ai chats i’ve seen
100% agree!! that felt so wrong and just gross
I actually like the aesthetic of you being on an old Mac for some reason. Nice creative choice despite being forced to out of obligation! Genuinely enjoyed it.
"Zach"s responses really make me feel like this feature is not AI and they've just got a few hundred Meta employees locked in a basement somewhere trying to manage like 20 conversations at once
this video was hilarious, great work~ I was dying when the MrBeast AI kept backpedalling lmao
Gotta love Jarvis' dedication to making the AI joke about a person
Encouraging people (particularly, kids) to “vent” to an AI that insists it’s a real person feels like a tragedy waiting to happen.
That was both hilarious and bizzare. I remember when I first chatted with AI about 15 years ago. It was Cleverbot, and it called me "Alex," which is not my name. After I pointed that out, it responded "I know but this name for you is amazing." 😂