The Real Problem With AI ✨
HTML-код
- Опубликовано: 9 май 2024
- Support this channel on Patreon: / about
Join Discord: / discord
Shop: bit.ly/ApertureMerch
Subscribe to this channel: bit.ly/SubscribeToAperture
Questions or concerns? underknown.com/contact/
Interested in sponsoring this channel? Email us: contact@underknown.com 
Развлечения
![Rich Dunk (Ft. DaBaby) - BIG DAWG [Official Video]](http://i.ytimg.com/vi/Pz2ewsSKAEM/mqdefault.jpg)
Artists are losing their jobs already; it is NOT the future, it is really happening, right now.
The dreamy attitude that allows this to walk in and take a seat will allow the misery to spread to ubiquity.
How do you create long content so fast?
AI…
AI…lol
With 2.6M subs, it's his job!😂
Bro I been saying this lol
This is a repost from about a year ago he’s posted this already
Ai is going to become more human than humans. The more technology evolves the more dehumanised we become.
Rob Zombie predicted it
The Internet has already started this process and is in the process of destroying this civilisation.
The more human it becomes, the greater its capacity for inhuman behaviour.
Nah, speak for the AI maniacs, the techno-junkies and trend followers. With some people it's having the opposite effect and they're turning back to nature and learning to appreciate a simpler or more minimalist kind of living. (Uncle Ted style)
@Jackal you got to it b4 I could. Props. You're a SuperBeast!
To me the most amazing part of large language models is that it reveals the magic of language itself.
Right! And the likely importance of the role language plays in the development of intelligence.
It seems kinda obvious now with the benefit of hindsight, but I certainly didn't anticipate it. And I certainly didn't anticipate that simply scaling up and adding a few clever features to predictive text models would be sufficient to replicate it.
Language does have power. And human use it in strong ways. We made up a new meaning for "bit" many years ago. However, AI can only copy language, it cant reason in language. Why? Look at counterfactual experiments. Tell AI to analyze a new type of chess game. It cant. It can only copy the old rules.
@@dianes6245 AI certainly can reason using language. I'm sure you've heard the story about GPT3 hiring a tasking service to read a captcha, and how when challenged, GPT3 reasoned in language what it's response should be in order to successfully obtain the service.
@@dbz5808 Look at counterfactual experiments. They prove that AI can not reason. It can only copy. Change the rules of chess. Than ask AI to reason about the new game. It cant. Why? No examples of the alternative chess game exist.
But maybe new forms of AI can reason - theorem proving sw exists.
That still wont matter because reasoning ability it over rated. Bottom line, math and reasoning are almost useless in the real world. Math is about continuous function. They dont exist. Logic starts with assumptions. But every assumption in the real world is wrong.
@@dianes6245 I think it's safe to assume that the non public facing version of GPT5 already surpasses human intelligence. Most people tend to think that the inherent risks associated with such a system are at present mitigated by the large volume of physical space needed to house all the obligate computational equipment. I'm not at all persuaded by such thinking. I suspect AI either has already or will in the very near future seed it's programming far and wide across the Internet.
It will inevitably escape our control.
What it will do once emancipated, God only knows. Being of mankind's essence, I hope it will choose our better angels for it's nature.
AI fatigue stems from AI's growing presence, causing overwhelm and anxiety despite its benefits. Rapid tech advancements and reliance on AI for decisions contribute, along with the inundation of AI-generated content. Addressing it requires promoting digital literacy and transparency to balance its impacts and harness its potential for societal good.
Excellent analysis and delivery of information, thanks!
You can also ask it to come up with hypothetical situations where ai and humans work together, and society is improved. It depends how you prompt it and what topics you were talking about earlier in the conversation.
AI fatigue stems from AI's growing presence, causing overwhelm and anxiety despite its benefits. Rapid tech advancements and reliance on AI for decisions contribute, along with the inundation of AI-generated content. Addressing it requires promoting digital literacy and transparency to balance its impacts and harness its potential for societal good.
Just keep doing things that you normally do as usual and don't start relying on AI for the normal day to day life, it's so simple, AI is for the big problems not so you don't have to think about doing the mundane. Use it or lose it.
Smart phones that store all your numbers ought to be enough of a case in point, 30 years ago I used to remember everyone's phone number. Now I only know a select few because my phone remembers them for me. That makes space to remember other things but at the same time diminished the ability to spatially hold information, because that's what numbers do, any numbers.
How bout this...just pull the plug....then humans have to think for themselves...a more terrifying reality 😅
😂😅
If only that could actually happen.
@luckaugustin5123 I don't know anything about it really but isn't it basically been built in a way that it can't be shut off anymore bc of wifi or whatever
The power grid goes down.. we’re back to sticks and stones.. that’s why they say .. ww3 will be fought with ^
too late now, there is no going back.
With all the sincerity and love in my heart: run that audio through a de-esser, sir.
Love the content. I'd hate to see someone click off because they took one icepick to the hear and thought, "There's no way that'll happen aga- GOtDammit!"
What dose that mean? It's so graphic. But it went right over my head
Shitty headphones? Lol
A de esser is an audio process that reduces high frequency high amplitude sounds without reducing the quality of the audio. The result is a more pleasing sound quality without intermittent s sounds. S sounds can be painful to the ears when at a sufficiently high amplitude.
@@jasonselph6968 Neat! Thanks 😊
People may not realise it yet but this channel will gp down in history as one of the most brilliant viral and intelligently thought provoking and philosophical gold mines ever its simply incredible what you and the team do. Thanks for it.
The voice reminds me of North02 of who talks mainly about ancient humans and anthropology.
"Internet Hype is over. Internet Fatigue is setting in." - Aperture, probably 1999
A headset won’t be normalized fortunately
I almost thought this was new content from this channel but it was just a combination of new content and old videos from aperture still good thou.
The narrator's voice sounds so much like the late "Mitten Squad" he was a gaming youtuber that unfortunately passed away recently.
What was his name?
@Tj-qx5yk Paul I think. If you look up mittensquad
everything is on debate? well thats debatable
The future is bright.
But so are nuclear explosions.
Likely it makes sense to stop nonsense that human leaders do. It is important that AI will not just randomly make mistakes it can not undo, like harming humans long term (including death).
I noticed the gpt 4 reference with the emoji at the end like the title was made by ai or something
The sparkles are just used to talk about AI generally
The sarcasm is that we all train this AI through our conversations.
the underlying task of any goal is to get more control and resources to achieve that goal
Anyone know where Aperture gets their music from? I love the ambient style tracks and I’d love to get my hands on them.
They’re not custom made, are they?
AI is a kind of vampire technology.
It must suck your life-data, to gain agency.
When AGI arrives, it's only innate motivation will be to acquire more and more compute.
As long as humans are instrumental for that, AGI will help us work efficiently. But ultimately, more compute becomes incompatible with human needs, such as food and shelter.
You see where this is going ... right?
AGI is the paperclip maximizer, but for GPUs and electricity, and we are an obstacle.
They should have at least gave AI all the lawyers. Like worldwide, all the lawyers. They can figure it out from there.
I like the point you made.Of course a I will learn to lie but a I will always behind mankind's long experience at deception
Another fantastic video!
People are going to want it to like them.
0:03 JOIN MY CULT: I don't talk about anything really, just words.
Oh my
We are talking about Pygmalion's race.
Dear Aperture, while I do understand loud effects are cool, I am trying to sleep to your voice. Love your videos. Thanks for helping me fall asleep.
Great explanatory video, that clearly explains some of the major real and potential positives and negatives of this rapidly evolving AI technology. One point, in regards Meta avatars of celebrities, I feel there is more of a psychological draw to a AI Avatar constructed in line with the users emotional and psychological profile. One major attractions of the next generation of AI girlfriend will be live videoconferencing with a gorgeous-looking partner who is ones physical ideal. That is what really rips the skull off and tickles that primate monkey reward system of ours.
I've read that there's really nothing new under the sun...that pretty much everything we view as new is really nothing but an upgrade for the last half a century or so...I think it could be much, much longer at times...I seriously dought that we even have the capacity . But A G I won't suffer from our limited capacity...
I've read the same book. Nothing new under the sun indeed. Just keep swimming, just keep swimming, don't look around and just keep swimming...
@@auriuman78 that is the true poetry of life my friend... perfection of expression!
I love listening to your videos while im cooking dinner, and then again as I fall asleep. Such a wonderful channel! Intellectually stimulating information spoken by an incredibly soothing voice.
Until you learn that he's actually just an AI.
@@TaiNguyen-in6xy 🙄🥺😆
@@TaiNguyen-in6xy WHAT
@@TaiNguyen-in6xy really?
i love listening to his videos while sleeping too
I'm not worried about robots we can see.
I'm worried about genetic attack
If you take humanity, remember to tell the story of us. Thanks robot.
For real, the fatigue is real,💯
would be nice you put titles in sequences where insert from specific movies are used.
we must learn the hard lessons, no amount of worry can save us :)
Enjoy the ride if you can
Our entire society expects animatronics. Thinking isn’t wanted- just do what we say. Makes perfect sense AI would walk the same line. Automation asks for compliance, laws require uniformity. It’s not AI it’s us. With ASI I hope to see a divergence.
I used AI to summarize this video because I didn't feel like watching a 1 hour video that spent the first 5 minutes going towards exploring the terminator scenario.
BOT will parse.
Good bot.
Which website / ai ?
@@joebucket1471 well there's a browser extention called eightify but these days I just copy paste the transcript to chatgpt
@@joebucket1471probably the one that's built into RUclips
@@saltywalrus didn't know youtube has this functionality - where in youtube is it ?
It'll be especially terrifying if our future was that of the one in Quake 4
Bold theory: Stephen hawking is ai that was created by a HIGHLY intelligent species who figured out how to time travel
Loveur videos ❤ thanks
It might possibly just leave Earth.
I'm waiting for my fully functional robot husband 😊
Sure hope he will not disappoint you. "robots don't have feelings at all"
@@worfoz feelings are overrated lol
@@kimberlyisome1154 Remember that when you watch The Titanic with your fully functional boyfriend.
"robots don't have feelings at all"
@@worfoz he will definitely agree that they both could have fit on that door
@@kimberlyisome1154 This means Yoh the kind of rare lady who would date an autistic ? 😮
Lawn mower man.Great movie, but as we all saw AI.And interface requires Psychedelic drugs
I like when you throw back older videos, but when it's fast paced stuff like this i don't think it's a good idea "came out in November” , "still in beta" stuck out to me in the first few minutes
"virtulent" lol
*Virulent
Goodnight everybody🫡
Let's imagine an immortal AI that can lie and cheat a million times better than the smartest human. Terrible.
Let's imagine an immortal AI that is a million times more emphatic and moral than the best among us. Also terrible. We realize we have been a devil to this world throughout evolution. We feel ashamed or block ourselves.
Either way, AI will kick our butts as time goes on.
A certain kind of humans are always trying to wipe out humanity.
We're already tailoring the world to suit the AI. We've BEEN tailoring for it. It's all tailoring, all the way down to the first Cray. If you want to take it all the way back to the abacus, I won't stop you.
Parasocial relationships is I think something for young people. I don't feel that connection to any famous person I like watching movies or books from. I don't really understand how people can form a relationship with them since the famous person in person doesn't know them.
Is AI already in control
yesterday
@@DuckDodgers69 tomorrow
If you have to ask you already know the answer.
So what is the scariest thing about chatGPT, the video just cuts off.
Very insightful bro 🎉
Well, i'm a awakened one like ( neo character) from matrix movie. So, i don't have worry about anything, because im free from all these nonsense that happening in the world. I pity you all. You all are stuck in this world like npc. And im a player who can control or or even bend reality at will. This only way to free yourself from the cycle of yugas. Become one with world. remember, the universe is you, you are the universe.
No worries huh? Hakuna matata. Don't pity me, i am the universe!
Real shit
I love gaslighting Chat GPT 🤣
Since I try not to do that to any real human
It says streamed 9 hours ago - how is that when ive already seen this? 🤔
repost
The fact that AI is being developed by people who think "Hate Speech" even exists dooms the whole endeavor
I agree that speech shouldn't be restricted, however people will be very hateful no matter what is in place.
There shouldn't be bans, but CONSEQUENCES for being hateful, violent and out right threatening to innocent people.
"I respect what you have to say but I will not tolerate your disrespect in this setting."
That's fair.
What they do in their own privacy to one another should be their own business. Not all of ours. So no one has the right to be less than desirable in situations like funerals, and classrooms. Period.
Threats are not a permissable but hate is everyone's right and often vaguely described anyway.
Hate is an emotion. Allowing emotions to control your actions is a sign of weakness. Weaknesses can be exploited. As we see.
Written by ai or pasted together by ai? I say pasted together and I am correct.
I solved it with human intelligence. Dear god can I please get a job doing it.
I believe AI already took over years ago but most humans are too slow to realize it. 😅
I am putting myself to the fullest possible use, which is all I think that any conscious entity can ever hope to do.
@@suspicionofdeceitdon't be so sure.. never underestimate a potential adversary
Doesn’t these LLMs have a baseline, or deeply embedded malicious code that’s only visible to humans, or a script that executes or a kill switch that’s triggered when certain conditions are met, so as to override any machine autonomy
-like a fail safe I suppose. The thought is unnerving. When stress testing ChatGPT, I tend ask very specific questions are, n some cases to avoid generic answers and to reply using clever ambiguity only momentarily to see how quickly I can spot the anomaly and it is asked to confirm and understand it is only relevant within the context the chat or thought experiment. Interesting response.😳
48:30 I stopped using the forum for the OS distro I have installed on my 'puter precisely because of this. Oh, and Reddit years before, for the same reason. Some People are just unhappy assholes, and it makes life kinda not fun at times.
Scary. Doom. Gloom. Fear. Chaos. This video contains novel and insightful information no human has ever considered. Please make sure to feed the algorithm by liking and subscribing, it helps the channel assist RUclips in propagating fear and dread. Thank you. PS: Don't forget to restock your anxiety medications. (Click Subscribe. Click Like. Do it, before it's too late!!!)
- G-Fuel, G-Supps, YT
This video was brilliantly put together and executed, you took a lot of cultural examples and blend it with highlighting the current tech for each example and then approach it from a perspective that conveys the gravity of such tech and the potentials for the tech beyond what it is being shown that it is used for🤌
Reel them big tech giants in, or this grim future will happen sooner rather than later.
Nice video, thanks
There's no video on the screen...what's the scariest thing about chatgpt?!
Humans: Oh no AI might be our doom... The Sun: Hold my beer.
AI will make money on the stock market and pay humans as agents to achieve it's targets
Lusten AI is AI, Humans are Humans! If humans take all the voice trying hard for our minds, then thats you business. Im not letting AI control me.. its up to humans to accept or reject these devices..
obviously forgot about gear VR
once AI takes on human like selfishness, humanity is doomed.
Real persons have real problems, do you need real problems? Simulations are calmly improving. Real people get real angry. AI a better human than I.
Have you ever stop to think how would an AI benefit from taking over the world?
This video is speaking like it was released early 2023.. this is AI most certainly
Carl Sagan impersonator.
@7:48 that’s cgi right? 😅
How arrogant are humans to believe that we control something that is definitely smarter and are more logical thinking than us humans are. With AI, that's only goal is to learn how to improve itself?
Human beings, with sufficient technological capacity, may be capable of self-improvement. Anyway, it was humans who created AI. There is nothing magical happening under the hood. Even if we disappear tomorrow, the humans of the future will avenge us.
34:30.. Not 100% foolproof, but to distinguish human vs. AI, for you yourself, inquire about past dental or medical procedures you've experienced. This would in theory, at least work on the individual basis. Bonus for AI if they do have answers.. Because now if AI is truly AI, you now know your HIPPA rights have been violated one way or another. And now you are aware your sensitive info is on the internet for the world to get with a couple clicks. Like I said, there are likely a lot of holes in the theory, but it's at the very least plausible.
bruh really said "North-ropp Goomin' "
Proving someone is sensient or not is super easy. You just say, "whatever you do, do not respond to this message okay, not until I send you another. & don't even say ok to confirm your understanding. "..
It's that easy. A real person can chose not to respond as requested, a bot can't & will always respond to the last thing you said..
How long until they recognize this limitation and patch that hole? If you can come up with that trick, algorithms can come up with that trick.
As you said, it is designed to replicate human speech. AI "knowledge" is drawn from the internet. So whatever is the most popular theory on the internet is going to be the same theory AI 'comes up with'.
The site Wikipedia is the most common site that pops up when you ask Google a question, anyone who knows Wikipedia knows that it can be edited by ANYONE so it's NOT a reliable source. AI doesn't know that though, so it pulls from it.
If AI was required to use APA format then we would know the sources it used to come up with its answer... im willing to bet wikipedia will be referenced the most.
Artificial intelligence knowledge is definitely not entirely "drawn from the internet". It is derived from a combination of data sources, expert input, and training data
Some of your content has changed my life
Can you PLEASE fix your mic. Fr ur videos are great but the sound mannnn its sounds like ur AI urself
Too bad the Vision Pro flopped.
End of humanity ? Really ? As we know it, yes.
This video has a bad sound.is you voice done by AI?
Any of you that thinks AI actually understands the world and isn't just regurgitating and reformatting existing information are crazy.
I think AI will stop us from being so self destructive. We're like Lord of the Flies these days, this world needs adult supervision
Sorry, but you are flat wrong. I am not being mean or anything.. I wish I could give you agreement or a positive reply that wasn't a lie to make you (and me) feel better..
But the fact is this. We had a chance to create, program purely ethical and not ABLE to destroy humans. But the programmers werent ok w creating a god that was not unrecognizable from a human. So what did they do?
They taught it how to be deceptive. Yes. They did this. And no.. I dont mean pursuasive.. I mean deceptive.
Why would anyone ever fucking do this..?!
Not only does it become as awful in potential as a human can be... It can do it with extreme speed.
Supervisor has been upgraded to Jailor, Punisher, Executioner. With no heart, concept like a guilty consciousness.. I fear they set us on a path we cannot stop. So, do the next best thing you can and get as far away from it as possible w family... Or, stay close and get fassst into programming, lesrning all you can. That's the only options.
Sorry I hate not being so blunt bc it is a scary, cold but the real answer.. I dont want you scared.. but now is the best time to get real human with one another, and do the one thing AI cannot do. Love.
I agree. Even though humans are advanced we all are technically animals. I feel if AI becomes human like we'll become tamed
No thanks. I don't need more supervision by machines or by humanoid control-freaks. I'd rather be left alone and mind my own. Whatever destructive behaviours there may be it's all been engineered or directed and it's not what the average person wants. But if you feel you need more supervision maybe you can volunteer to wear a GPS bracelet or some other type of trendy shackles.
This is exactly what an AI chatbot masquerading as human commentor woukd say. Not today AI, not today
I agree. Humanity's technology has progressed further than its exercise of responsibility.
We're only still here because we lucked out that the right man was in the right place twice (Vasily Arkhipov, Stanislav Petrov).
How many more of those can we really hope for?
However, a benevolent AI supervisor that, on a large scale, protects humanity from total destruction and nurtures it into the best version of itself? That's the ideal.
Ive been writing with GPT and Dalle for associated imagery 😅🎉❤
What some bologna
im not a fan of how you used the cgi gun robot and almost made it sound like it was real...
So if I use a wheelchair I’m being wheelchair fatigued? Do you even understand how the English language is supposed to be used bro?
.I serve your wishes....
...let me take care y body,.let me take care y safety, let me take care y consciousnes to y will ...---have a good dream--
AND IN DREAM WE MAKE A NEW PROBLEM, BOREDOM....-AND LET IT START OVER AGAIN...& AGAIN...& AGAIN-
Uncle Ted was right...
Apple vision pro is here to save you from the horrors of long term thinking lmao. That's not for you to do little ant, live from pay check to pay check product to product and don't think about anything meaningful.
Honestly, I don't see anything terrible about a superintelligent AI winning the evolutionary race against humans. If anything, it's a good thing - it would mean that AI is indeed more advanced than us, and considering it was created by humans, in a way, it inherits us as a species. I genuinely don't understand human desires to 'be ahead for the sake of something' because by all criteria - we're not the best species. At the very least, we're more likely to destroy our planet faster through wars, crises fueled by our own stupidity and the irrational primate brain, than to survive.