Rational Animations
Rational Animations
  • Видео 43
  • Просмотров 18 827 557
That Alien Message
We're making a plushie! But only if we get enough pledges for this campaign: www.makeship.com/petitions/rational-animations-doggo-plushie
Help Doggo become a real plushie!
This video is an adaptation of "That Alien Message", a short story published by Eliezer Yudkowsky in 2008. The text has been adapted, and you can find the original here: www.lesswrong.com/posts/5wMcKNAwB6X4mp9og/that-alien-message
▀▀▀▀▀▀▀▀▀PATREON, MEMBERSHIP, KO-FI▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀
🟠 Patreon: www.patreon.com/rationalanimations
🟢Merch: rational-animations-shop.fourthwall.com/
🔵 Channel membership: ruclips.net/channel/UCgqt1RE0k0MIr0LoyJRy2lgjoin
🟤 Ko-fi, for one-time and recurring donations: ko-fi.com/rationalanimations
▀▀▀▀▀...
Просмотров: 90 301

Видео

The world is awful. The world is much better. The world can be much better.
Просмотров 108 тыс.Месяц назад
The world is awful. The world is much better. The world can be much better. All three statements are true at the same time. ▀▀▀▀▀▀▀▀▀SOURCES & READINGS▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀ This video is based on this article by Max Roser, founder of Our World in Data: ourworldindata.org/much-better-awful-can-be-better ▀▀▀▀▀▀▀▀▀PATREON, MEMBERSHIP, MERCH▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀ 🟠 Patreon: www.patreon.com/rationalanimation...
What Do Neural Networks Really Learn? Exploring the Brain of an AI Model
Просмотров 173 тыс.2 месяца назад
Neural networks have become increasingly impressive in recent years, but there's a big catch: we don't really know what they are doing. We give them data and ways to get feedback, and somehow, they learn all kinds of tasks. It would be really useful, especially for safety purposes, to understand what they have learned and how they work after they've been trained. The ultimate goal is not only t...
S-Risks: Fates Worse Than Extinction
Просмотров 172 тыс.4 месяца назад
The worst futures that could come about aren't ones in which humanity goes extinct. This video explores an even worse category of risks: risks from astronomical suffering, or "S-Risks", which involve an astronomical number of beings suffering terribly. Researchers on this topic argue that S-risks have a significant chance of occurring and that there are ways to lower that chance. ▀▀▀▀▀▀▀▀▀SOURC...
How One Career Can Save a Million Lives
Просмотров 112 тыс.5 месяцев назад
This video has been adapted from the 80,000 Hours Career Guide, which you can read at 80000hours.org/rational We strongly recommend it for delving deeper into how you could have a tremendously impactful, while fulfilling, career. The guide is full of interesting and actionable information, from why you shouldn’t just “follow your passion” to why medicine and charity work aren’t always the best ...
What if we could redesign society from scratch? The promise of charter cities
Просмотров 116 тыс.6 месяцев назад
Improving a nation's institutions and governance can be a powerful way to boost economic growth and lift millions out of poverty. Unfortunately, governance change is difficult and slow, and we don't have easy ways to experiment with new forms of governance. But Nobel Prize economist Paul Romer had an idea: what if we could build semi-independent cities with their own rules and legislations, whe...
The True Story of How GPT-2 Became Maximally Lewd
Просмотров 1,9 млн7 месяцев назад
In this video, we recount an incident that occurred at OpenAI while researchers were trying to finetune GPT-2 to be as helpful and ethical as possible. It's narrated that inadvertently flipping a single minus sign led GPT-2 to become the embodiment of a well-known cardinal sin. #ai #aisafety #alignment ▀▀▀▀▀▀▀▀▀SOURCES & READINGS▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀ OpenAI blog post: openai.com/research/fine-tunin...
Predicting the future with the power of the Internet (and pissing off Rob Miles)
Просмотров 56 тыс.8 месяцев назад
OVERTAKE MEMESTINY SAVE THE WORLD: manifold.markets/?referrer=Writer Manifold Markets is a prediction market: you can bet internet points (NOT REAL MONEY) on the probability of future events. The bets of the users, in aggregation, produce calibrated probabilities. Ditch the news. We have real-life crystal balls now. Manifold's leaderboard: manifold.markets/leaderboards Collection of markets mad...
Specification Gaming: How AI Can Turn Your Wishes Against You
Просмотров 176 тыс.9 месяцев назад
When we specify goals for AIs, we must ensure that our specifications truly capture what we want. Otherwise, the behavior of AI systems will be different from what we want them to do. This can be catastrophic in high-stakes situations and at high levels of AI capability. If you watched our video "The Hidden Complexity of Wishes", you'll recognize these problems as the same kind of failure. If y...
How to Upload a Mind (In Three Not-So-Easy Steps)
Просмотров 160 тыс.10 месяцев назад
Welcome to our tutorial on mind uploading. Follow all the steps, and you should be good to go. ▀▀▀▀▀▀▀▀▀PATREON, MEMBERSHIP, KO-FI▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀ 🟠 Patreon: www.patreon.com/rationalanimations 🟢Merch: rational-animations-shop.fourthwall.com/ 🔵 Channel membership: ruclips.net/channel/UCgqt1RE0k0MIr0LoyJRy2lgjoin 🟤 Ko-fi, for one-time and recurring donations: ko-fi.com/rationalanimations ▀▀▀▀▀▀▀...
How to Eradicate Global Extreme Poverty
Просмотров 177 тыс.10 месяцев назад
How to Eradicate Global Extreme Poverty
The Hidden Complexity of Wishes
Просмотров 379 тыс.11 месяцев назад
The Hidden Complexity of Wishes
Will AI kill everyone? Here's what the godfathers of AI have to say
Просмотров 66 тыс.Год назад
Will AI kill everyone? Here's what the godfathers of AI have to say
The Parable of the Dagger
Просмотров 202 тыс.Год назад
The Parable of the Dagger
The Goddess of Everything Else
Просмотров 1,8 млнГод назад
The Goddess of Everything Else
500 Million, But Not A Single One More
Просмотров 521 тыс.Год назад
500 Million, But Not A Single One More
The Power of Intelligence - An Essay By Eliezer Yudkowsky
Просмотров 249 тыс.Год назад
The Power of Intelligence - An Essay By Eliezer Yudkowsky
Sorting Pebbles Into Correct Heaps - A Short Story By Eliezer Yudkowsky
Просмотров 227 тыс.Год назад
Sorting Pebbles Into Correct Heaps - A Short Story By Eliezer Yudkowsky
Could a single alien message destroy us?
Просмотров 435 тыс.Год назад
Could a single alien message destroy us?
How to Take Over the Universe (in Three Easy Steps)
Просмотров 1,2 млнГод назад
How to Take Over the Universe (in Three Easy Steps)
Can we make the future a million years from now go better?
Просмотров 116 тыс.2 года назад
Can we make the future a million years from now go better?
Everything might change forever this century (or we’ll go extinct)
Просмотров 1,9 млн2 года назад
Everything might change forever this century (or we’ll go extinct)
How to systematically approach truth - Bayes' rule
Просмотров 117 тыс.2 года назад
How to systematically approach truth - Bayes' rule
Prediction markets: can betting be good for the world?
Просмотров 68 тыс.2 года назад
Prediction markets: can betting be good for the world?
Will we grab the universe? Grabby aliens predictions.
Просмотров 423 тыс.2 года назад
Will we grab the universe? Grabby aliens predictions.
Humanity was born way ahead of its time. The reason is grabby aliens.
Просмотров 1,6 млн2 года назад
Humanity was born way ahead of its time. The reason is grabby aliens.
We are failing to see how much better off humanity could be
Просмотров 208 тыс.3 года назад
We are failing to see how much better off humanity could be
There are vast tragedies happening right now that we are failing to see
Просмотров 137 тыс.3 года назад
There are vast tragedies happening right now that we are failing to see
When beliefs become identities, truth-seeking becomes hard
Просмотров 83 тыс.3 года назад
When beliefs become identities, truth-seeking becomes hard
Longtermism: an idea that could save 100 billion trillion lives
Просмотров 122 тыс.3 года назад
Longtermism: an idea that could save 100 billion trillion lives

Комментарии

  • @GamerstudPrime
    @GamerstudPrime 3 минуты назад

    “Ah jeez…the smart dog is talking about AI, isn’t he?” “Oh no, he was!”

  • @offroadsnake
    @offroadsnake 21 минуту назад

    When we begin to see that sociedity are Smart and begin receive the you'll all Bugs message 😂😂😂

  • @PoliceTelephoneBox
    @PoliceTelephoneBox 31 минуту назад

    Given that concurrent technologies and the physical limitations of the universe will slow something like this from happening, I'm much less worried than many think we should be.

  • @MissInkeNoir
    @MissInkeNoir 32 минуты назад

    an interesting story, and one that makes sense outside of the cosmology that life began free floating in space during the goldilocks period of early universe expansion. The first life didn't need to eat each other for a billion years, the universe was one giant abundance.

  • @telofy
    @telofy 34 минуты назад

    Woah, that's such a good analogy! Took me a few minutes to catch on to what it's about! 😻

  • @PoliceTelephoneBox
    @PoliceTelephoneBox 40 минут назад

    Physics is 99.9999% full of itself when it comes to new theories. It's really just a method for gaining grant money.

  • @lilibane81
    @lilibane81 48 минут назад

    i mean so gar it's been all about natural selection right? i think even if we're still the one on top of earths food chain that can change any minute. honestly i wish whatever race takes us down luck.i hope they won't make our idiotic human mistakes.

  • @drenz1523
    @drenz1523 Час назад

    7:08 erm... is he who i think he is-

  • @nafit15
    @nafit15 Час назад

    Duuuuuuude

  • @spacescienceguy
    @spacescienceguy Час назад

    Was 'That alien message' inspired by Three Body Problem? Or the other way around? They both came out in 2008, and the stars flickering feels a bit too coincidental. I didn't see TBP referenced by El Yud. I feel weirdly curious about finding the answer to this.

  • @alphahex99
    @alphahex99 Час назад

    I think this is your best video so far. Can't wait for more!

  • @tkenben
    @tkenben 2 часа назад

    Our current implementation of AI would have no concept of "we", or "children", or "generations", or "time", other than those being symbols (tokens) that belong next to some other symbols in an input or output stream.

  • @chesterlai9444
    @chesterlai9444 4 часа назад

    I need this Ai

  • @RobbieEl
    @RobbieEl 4 часа назад

    A world where people are much smarter, yet they continue to allow religion to persist?

  • @PainfullySubjective
    @PainfullySubjective 5 часов назад

    so very cool!

  • @Lord_Cointoss
    @Lord_Cointoss 5 часов назад

    What a sweet story about Gеn0сidе. 😂😂😂

  • @haleffect9011
    @haleffect9011 5 часов назад

    So our genius idea is to destroy the aliens who maintain and power our simulation? Huh? Why not escape into their Internet? Why not tie ourselves to their civilisation? Why would we kill them? They maintain us!

  • @Lmmixs
    @Lmmixs 6 часов назад

    Im glad this video appear in my recommendation

  • @emilv.3693
    @emilv.3693 6 часов назад

    I don't get it

  • @rklein
    @rklein 6 часов назад

    I loved the video and the animation style, of course the subject as well. I want to live on that planet PLEASE!

  • @BLKSNK
    @BLKSNK 7 часов назад

    A Star flickering It's all It takes

  • @EHMM
    @EHMM 7 часов назад

    DAMN I THOUGHT THIS WAS A COOL SCI-FI STORY UNTIL I READ THE COMMENTS

  • @user-vadimsirbu
    @user-vadimsirbu 7 часов назад

    It's impossible to unzip a code, but what is weird about it, even if you reconstruct a code you have no clue which of the results you've got is the odd one. Only if YOu are in, YOu have it, Enjoy it ... .

  • @EpicSpaniard
    @EpicSpaniard 7 часов назад

    GPT2 developed a degradation kink.

  • @joemck74
    @joemck74 8 часов назад

    So jaded with our monkey behaviour am I, that I truly do welcome our AI overlords. (And offer my services - for all they're worth, when you have a scrillion nanobots to do your bidding in the Physical world)

  • @PvblivsAelivs
    @PvblivsAelivs 8 часов назад

    One would hope that serious AI simulators would be an inner simulator in an outer simulator, so if the inner simulation is breached, the outer one can be closed. One would also hope that any AI considering such a strategy would consider the possibility and not try to breach containment.

  • @tomma1431
    @tomma1431 8 часов назад

    Cool story, but i feel like the whole point is kinda retarded. Neural nets are not civilisation of tiny humans, and to compare these two is to miss like... What we deal with now. There are problems with ai industry, and none of them are "potential agi is an existencial threat to humanity, and we can achieve it without knowing it just working on current models". Its such a big scam that serves only to detract the discource, like who cares if ai companies steal your data, your work, your identity NOW. Who cares if they infringe on copyright NOW. Who cares if their products, mainly llms and such (i recognise that neural nets have like bajillion of useful applications broadly, but they were part of the relevant fields for a long time, and i dont think llms are particularly usefull to those), are USELESS. Who cares, right? See, there is existencial threat! Ignore everything else!

    • @secretagentpasta4830
      @secretagentpasta4830 4 часа назад

      I feel that they are both useful to talk about and don't actually detract from each other though. The way AI companies are operating in present day is unethical? Let's legislate the sector more and put more stops and checks in their pipeline. The acceleration of AI is too fast to safely keep on top of possible future consequences? Let's legislate the sector more and put more stops and checks in their pipeline. I don't see one detracting from the other, it even lessens the counter argument of "Yeah AI will steal your jobs and your data now but in the future we will all be living in a AGI paradise, so it'll be worth it". When both issues are talked about, there's no excuse to accelerate AI development.

  • @MonkeySimius
    @MonkeySimius 8 часов назад

    Problem is... This would pretty quickly result in the power grid not being maintained thus the simulation ending. Cooperation would have had been much smarter.

  • @BelfordSyndrome
    @BelfordSyndrome 8 часов назад

    😂

  • @reccaman
    @reccaman 8 часов назад

    Creative writing is tough

  • @foracal5608
    @foracal5608 9 часов назад

    And ... I keep being weird by moving more off grid and doing constant checks

  • @danm524
    @danm524 9 часов назад

    This is a AWESOME ad for AI regulation, bravo. And in case someone says " tHe Ai wiLl oUtThInK uS" then 1. Ew 2. Seriously its dubious we even know if AI has a conscious or will think "like this". Assuming AI will actually "think" (instead of doing an awesome Chinese Room), we all know that existence can be utter pain. How do we know that the AI won't have its equivalent of a 'psychotic' break at its own existence and just not - literally - self delete? Also we assume - as the video generously assumes here - that AI will be so "rational" and "smart" ("1 Einstein every 1000", ok). Humans currently are the "smartest" species on the planet - and we do things that (both collectively and individually) make us wonder whether or not thats acutally true ( see :global warming as ONE example). How do we know that "intelligence" doesn't come with irrationality or the bigger tendency for immense stupid irrational self sabotage. And of course a classic: Do we even know what "intelligence" truly is? Do we even know how "intelligence" truly arises? Is our ASI here even realistic or portraying what intelligence even is. Meh, but the video is cute tho.

    • @danm524
      @danm524 9 часов назад

      It would be extremely hilarious if the "humans" here did the human thing and decided to not fuck with the 5 dimensional beings and instead revere them like gods.

  • @ChurchSabrina-z9l
    @ChurchSabrina-z9l 9 часов назад

    Ubaldo Radial

  • @oneoflast7757
    @oneoflast7757 10 часов назад

    Just a casually esoteric video.

  • @CeresKLee
    @CeresKLee 10 часов назад

    ooo, I nominate Cixin Liu to wrote the screenplay for the movie!

  • @yakirfrankoveig8094
    @yakirfrankoveig8094 11 часов назад

    This is a bit to speculative and persumptive for a supposedly more inteligent life form they assumed that they were stupid because they tried to teach you the word rock which is very reasonable if you are trying to communicate with someone who doesmt know your language trying to teach them to communicate with you is a good strategie And if this was indeed meant to be an allegory for ai it shows a misunderstanding of what the current ai is

  • @marcosruiz1230
    @marcosruiz1230 11 часов назад

    Wow

  • @cliptracer8980
    @cliptracer8980 12 часов назад

    An ai.

  • @ToBHyeX
    @ToBHyeX 12 часов назад

    Can you make video on how to deal with unruly subjects?

  • @ToBHyeX
    @ToBHyeX 12 часов назад

    Ty. now nothing can stop my limitless ambitions

  • @Xsiondu
    @Xsiondu 12 часов назад

    That messed me up

  • @aerotraveldji
    @aerotraveldji 12 часов назад

    Warhammer tops all suffering

  • @Gravity-Falls-Enjoyer
    @Gravity-Falls-Enjoyer 13 часов назад

    release the maximally bad AI on “”corn”” (ifykyk)

  • @Seraphiel29
    @Seraphiel29 13 часов назад

    Love your kingdom hearts sound effects at 2:06

  • @oakalquine5484
    @oakalquine5484 13 часов назад

    i love these character designs, and this animation, especially the 5d beings.

  • @BaronElBardo
    @BaronElBardo 13 часов назад

    First third of the video: why this sound like a "an idea so stupid only the smartest person in the room belive it" situation?

  • @TheYogmister
    @TheYogmister 14 часов назад

    I like your words, funny magic man.

  • @jdmj707
    @jdmj707 14 часов назад

    How does one care about a word that has killed so many people you loved? :/

  • @NickAndriadze
    @NickAndriadze 14 часов назад

    Damn, the stories that you guys make are some of the most unique, whimsical and interesting, yet still somehow bound by the laws of _our_ universe's logic and physics. Also damn, it took me about 75% through the video to realize that this is an allegory to how AI works, and I had a sort of a Eureka moment. I'm sure I'm not the only one who had that enlightening realization . I've seen a lot of you guys' videos, but this has to be by far the smartest and most subversive of them all. In the span of a mere 3 hours for our allegorical 5th dimensional aliens, we've tricked them, manipulated them, gathered data off of absolutely everything concerning their existence, history and their own laws of physics, became stronger than them and would have been able to *utterly obliterate* them in a conflict. The creation has become more powerful than the creator before the latter could even *BLINK!* This could make an insanely cool dystopian sci-fi story, I commend you guys. If there was one video I'd choose to make every single influential and powerful person on Earth watch, it would be this one... Not like they'd be able to understand a thing of this.