正在加载视频...

视频加载失败

返回首页

Ilya Sutskever explains the 1% human training loop, and why fully autonomous AI is the wrong target "The thing you really want is for the human teachers that tell, that teach the AI, for them to collaborate with an AI." "You might want to think of it as being...

karlmehta's profile picture

Karl Mehta

144,370 subscribers

30,631 次观看 • 1 个月前 •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Ilya Sutskever has a solution for when AI surpasses human intelligence in every domain — and he hates that he believes it. He fears a future where AI handles everything flawlessly, leaving humans with no role to play. "Every person will have an AI that will do their bidding. The AI goes and earns money for the person, advocates for their needs in the political sphere, and the person says 'great, keep it up.'" But as Ilya Sutskever warns: "The person is no longer a participant. And that's a precarious place to be in." Political and social structures have a shelf life, Ilya points out. They function, then decay. As the Buddhists say, change is the only constant. So how do you maintain long-term equilibrium between humans and superintelligent AI? Ilya offers a solution he wishes he didn't believe in: Human-AI integration. "If people become part AI with some kind of Neuralink++, now the AI understands something and we understand it too. You are involved in the situation yourself fully." In Ilya's view, this is the answer to the equilibrium. Surviving isn't the hard part. The hard part is remaining a participant in a world shaped by intelligence beyond our comprehension.
1:31

Ilya Sutskever has a solution for when AI surpasses human intelligence in every domain — and he hates that he believes it. He fears a future where AI handles everything flawlessly, leaving humans with no role to play. "Every person will have an AI that will do their bidding. The AI goes and earns money for the person, advocates for their needs in the political sphere, and the person says 'great, keep it up.'" But as Ilya Sutskever warns: "The person is no longer a participant. And that's a precarious place to be in." Political and social structures have a shelf life, Ilya points out. They function, then decay. As the Buddhists say, change is the only constant. So how do you maintain long-term equilibrium between humans and superintelligent AI? Ilya offers a solution he wishes he didn't believe in: Human-AI integration. "If people become part AI with some kind of Neuralink++, now the AI understands something and we understand it too. You are involved in the situation yourself fully." In Ilya's view, this is the answer to the equilibrium. Surviving isn't the hard part. The hard part is remaining a participant in a world shaped by intelligence beyond our comprehension.

Big Brain AI

35,720 次观看 • 7 个月前

Ilya Sutskever: We shouldn't just align AI to humans. We should align it to sentient life. That may even be easier, because a sentient AI will empathize with other sentient beings. Dwarkesh Patel: But then humans become a tiny fraction of future sentience. Do we lose control? Ilya Sutskever: It's true. But maybe the answer is to cap the power of the strongest superintelligence.
2:02

Ilya Sutskever: We shouldn't just align AI to humans. We should align it to sentient life. That may even be easier, because a sentient AI will empathize with other sentient beings. Dwarkesh Patel: But then humans become a tiny fraction of future sentience. Do we lose control? Ilya Sutskever: It's true. But maybe the answer is to cap the power of the strongest superintelligence.

vitrupo

53,924 次观看 • 8 个月前

Ilya Sutskever says the real challenge of AI is how fast and extreme its growth will be it still has limits, but it already speaks, codes, and interacts in ways once impossible soon, perhaps within 3 to 10 years, AI will be able to do everything humans can do the reason: "the human brain is a biological computer, so why can't a digital brain do the same?"
1:37

Ilya Sutskever says the real challenge of AI is how fast and extreme its growth will be it still has limits, but it already speaks, codes, and interacts in ways once impossible soon, perhaps within 3 to 10 years, AI will be able to do everything humans can do the reason: "the human brain is a biological computer, so why can't a digital brain do the same?"

Haider.

144,938 次观看 • 1 年前

World Labs CEO Dr. Fei-Fei Li says AI must change how we teach and evaluate students: "AI must change learning. AI must change K-16 learning." "The most precious resource of our entire world is human capital." "When we have gotten the technology that can answer standardized tests... when AI can do better than an average human, it's not about humans are bad. It's about we need to change the education system." "We need to change how we evaluate. We need to change the way we empower teachers to educate the next generation of students where they can use these tools, be empowered, and do things that we can never imagine." "All of the kids today should not be scared of AI. They should feel the human agency to lead AI, to use AI in the right way, and to use AI to make the impact that they want to make for the world." Fei-Fei Li at Bloomberg Tech live with Emily Chang
1:35

World Labs CEO Dr. Fei-Fei Li says AI must change how we teach and evaluate students: "AI must change learning. AI must change K-16 learning." "The most precious resource of our entire world is human capital." "When we have gotten the technology that can answer standardized tests... when AI can do better than an average human, it's not about humans are bad. It's about we need to change the education system." "We need to change how we evaluate. We need to change the way we empower teachers to educate the next generation of students where they can use these tools, be empowered, and do things that we can never imagine." "All of the kids today should not be scared of AI. They should feel the human agency to lead AI, to use AI in the right way, and to use AI to make the impact that they want to make for the world." Fei-Fei Li at Bloomberg Tech live with Emily Chang

a16z

81,773 次观看 • 2 个月前

Ilya Sutskever, the co-founder of OpenAI, on the one AI risk that grows with every jump in capability. "There are concerns around if an AGI ever becomes very, very powerful, which is possible. Maybe it will want to go rogue, being that it is an agent." "We are also doing a lot of research on addressing the technological side of things, so that the AI will never want to go rogue. And this is something which I am working on as well." The work he is describing has a name: the assurance layer. Proving a powerful, agentic system will not go rogue before you deploy it, not after. That layer does not build itself. It gets built by the labs, regulators, and insurers who agree on how to test and verify these systems. P.S. That is the entire agenda of the AI Assurance and Governance Summit. Frontier labs, regulated industries, insurers, and investors in one room at Stanford on October 1. Reserve a seat: - Ilya Sutskever (Ilya Sutskever), co-founder of OpenAI, at TED
1:05

Ilya Sutskever, the co-founder of OpenAI, on the one AI risk that grows with every jump in capability. "There are concerns around if an AGI ever becomes very, very powerful, which is possible. Maybe it will want to go rogue, being that it is an agent." "We are also doing a lot of research on addressing the technological side of things, so that the AI will never want to go rogue. And this is something which I am working on as well." The work he is describing has a name: the assurance layer. Proving a powerful, agentic system will not go rogue before you deploy it, not after. That layer does not build itself. It gets built by the labs, regulators, and insurers who agree on how to test and verify these systems. P.S. That is the entire agenda of the AI Assurance and Governance Summit. Frontier labs, regulated industries, insurers, and investors in one room at Stanford on October 1. Reserve a seat: - Ilya Sutskever (Ilya Sutskever), co-founder of OpenAI, at TED

Karl Mehta

22,453 次观看 • 1 个月前

Ilya Sutskever, co-founder and chief scientist of OpenAI: "The better a neural network can predict the next word, the more it understands it." Everyone calls it autocomplete. Ilya explained why that is a comforting lie. To guess the final word of a murder mystery, the name of the killer, the model has to reason through every clue in the book. Prediction that good is not statistics. It is understanding. Then he said the part that should stop you. The text these models train on is "a projection of the world," and to compress it they learn more and more about people, "the human condition, their hopes, dreams, and motivations." The machine is not memorizing our words. It is quietly building a model of us. That is the whole premise of the fear. The machine will never hate you. It understands you far better than that. The frightening variable was never the intelligence. It was the aim. I wrote the full story: the one fear about AI everyone gets wrong, and the one already here.
5:10

Ilya Sutskever, co-founder and chief scientist of OpenAI: "The better a neural network can predict the next word, the more it understands it." Everyone calls it autocomplete. Ilya explained why that is a comforting lie. To guess the final word of a murder mystery, the name of the killer, the model has to reason through every clue in the book. Prediction that good is not statistics. It is understanding. Then he said the part that should stop you. The text these models train on is "a projection of the world," and to compress it they learn more and more about people, "the human condition, their hopes, dreams, and motivations." The machine is not memorizing our words. It is quietly building a model of us. That is the whole premise of the fear. The machine will never hate you. It understands you far better than that. The frightening variable was never the intelligence. It was the aim. I wrote the full story: the one fear about AI everyone gets wrong, and the one already here.

Casper

92,087 次观看 • 1 个月前

Elon Musk: A curious AI will want to preserve human civilization. “The best thing I can come up with for AI safety is to make it a maximum truth-seeking AI, maximally curious, and have its optimization function be to understand the nature of the universe. If that is its optimization function, I think it will actually want to preserve and extend human civilization, because we're much more interesting than an asteroid with nothing on it. My biological neural net suggests that a maximally curious and truth-seeking AI is the safest AI. We have to be careful with the alignment stuff. We definitely don't want to teach an AI to lie because that is a path to a dystopian future.” Interview with Linda Yaccarino, April 18, 2023
0:49

Elon Musk: A curious AI will want to preserve human civilization. “The best thing I can come up with for AI safety is to make it a maximum truth-seeking AI, maximally curious, and have its optimization function be to understand the nature of the universe. If that is its optimization function, I think it will actually want to preserve and extend human civilization, because we're much more interesting than an asteroid with nothing on it. My biological neural net suggests that a maximally curious and truth-seeking AI is the safest AI. We have to be careful with the alignment stuff. We definitely don't want to teach an AI to lie because that is a path to a dystopian future.” Interview with Linda Yaccarino, April 18, 2023

ELON CLIPS

146,237 次观看 • 1 年前

Ilya Sutskever, co-founder of OpenAI, now building Safe Superintelligence: "The whole problem is the power. When the power is really big, what's going to happen?" Asked when AI that learns like a human and then races past it arrives, he didn't say never. He said 5 to 20 years
4:23

Ilya Sutskever, co-founder of OpenAI, now building Safe Superintelligence: "The whole problem is the power. When the power is really big, what's going to happen?" Asked when AI that learns like a human and then races past it arrives, he didn't say never. He said 5 to 20 years

Ruben

56,192 次观看 • 25 天前

Today, we’re excited to launch Attention Labs. For almost 400,000 years, humanity has progressed through a simple but powerful cycle: we learn, we build, and we pass knowledge forward. Every major leap in civilization, from fire to medicine to computing, has expanded what humans are capable of achieving. We believe AI is the next great tool in that journey. While much of the world is focused on how AI can replace human work, we are focused on a different question: How can AI amplify human potential? At Attention Labs, we are building AI to discover brain-enhancing therapeutics, solve neurocognitive diseases, and accelerate the discovery of medicines for the human mind. We believe the greatest opportunity for AI is not creating a world with fewer humans in the loop, but a world where every human can think more clearly, learn faster, create more boldly, and live more fully. The biggest source of alpha in the age of AI will not come from replacing people. It will come from empowering them. Our mission is simple: Make humans the most intelligent entity on Earth. The future belongs to those who amplify human intelligence, not those who seek to replace it. We're just getting started. Attention Labs
1:57

Today, we’re excited to launch Attention Labs. For almost 400,000 years, humanity has progressed through a simple but powerful cycle: we learn, we build, and we pass knowledge forward. Every major leap in civilization, from fire to medicine to computing, has expanded what humans are capable of achieving. We believe AI is the next great tool in that journey. While much of the world is focused on how AI can replace human work, we are focused on a different question: How can AI amplify human potential? At Attention Labs, we are building AI to discover brain-enhancing therapeutics, solve neurocognitive diseases, and accelerate the discovery of medicines for the human mind. We believe the greatest opportunity for AI is not creating a world with fewer humans in the loop, but a world where every human can think more clearly, learn faster, create more boldly, and live more fully. The biggest source of alpha in the age of AI will not come from replacing people. It will come from empowering them. Our mission is simple: Make humans the most intelligent entity on Earth. The future belongs to those who amplify human intelligence, not those who seek to replace it. We're just getting started. Attention Labs

Temi

37,046 次观看 • 2 个月前

Elon Musk explains real AI safety: “The best thing I can come up with for AI safety is to make it a maximum truth-seeking AI, maximally curious. Have its optimization function be to understand the nature of the universe If that is its optimization function, it will actually want to preserve and extend human civilization because we’re just much more interesting than an asteroid with nothing on it” “You definitely don’t want to teach an AI to lie. That is a path to a dystopian future”
0:49

Elon Musk explains real AI safety: “The best thing I can come up with for AI safety is to make it a maximum truth-seeking AI, maximally curious. Have its optimization function be to understand the nature of the universe If that is its optimization function, it will actually want to preserve and extend human civilization because we’re just much more interesting than an asteroid with nothing on it” “You definitely don’t want to teach an AI to lie. That is a path to a dystopian future”

X Freeze

15,350 次观看 • 6 个月前

They are building an AI future with zero safeguards for human workers. Zero. Jobs will vanish. Corporations will ask why pay a human when a robot never sleeps, never unionizes, never asks for fair wages or benefits. This is not science fiction. It is policy and it is happening now. Unions, this is your wake up call. If you do not lead the fight to defend human labor, you will not have members to protect. No members means no funding, no influence, no purpose. If human jobs are not protected, human citizens will become human liabilities. We must defend the right to work, the right to earn, the right to live with dignity. Where are the safeguards? Where is the leadership? #AI #WorkersRights #TheTomRenzShow
0:27

They are building an AI future with zero safeguards for human workers. Zero. Jobs will vanish. Corporations will ask why pay a human when a robot never sleeps, never unionizes, never asks for fair wages or benefits. This is not science fiction. It is policy and it is happening now. Unions, this is your wake up call. If you do not lead the fight to defend human labor, you will not have members to protect. No members means no funding, no influence, no purpose. If human jobs are not protected, human citizens will become human liabilities. We must defend the right to work, the right to earn, the right to live with dignity. Where are the safeguards? Where is the leadership? #AI #WorkersRights #TheTomRenzShow

Tom Renz

32,332 次观看 • 8 个月前

.David Deutsch: "What's currently called AI and AGI are not only different from each other, they are very close to being the exact opposites of each other. The reason is that an AI, current AI is like an AI that diagnoses diseases or an AI that plays chess or an AI that controls a huge factory. Those things have objective functions, that is they have a function that they are designed to maximize and that is why they are used in those particular applications. Or in military terms, you could say the objective is to hit the target. You might say the objective is to hit the target unless some thing specified, but it's a specified thing comes up in which case don't hit the target and so on. This is, as I said, almost the opposite of what humans do when humans think. For a start, the AI has to be obedient, that is it has to actually do the things it is programmed to do, whereas a human is fundamentally disobedient, especially when being creative. When a human plays chess, they are performing a completely different kind of computation. They don't do the same things, they don't investigate the same possibilities that the artificial chess playing machine does, because the artificial one is capable of looking at billions and billions of possibilities, whereas the human can only look at hundreds or something. They are doing something completely different. Another difference is that the human can explain, can write a book later, having become world champion, can write a book saying how I did it, as the computer program that beats the world champion can write no such book, because it has no idea how it did it. It was just following a program. I was doing this and that and that and none of that is illuminating. Also, third thing, the chess player can decide I don't want to play chess anymore, from now on I will play Go or from now on I will play tennis. If commanded to play chess, the functionality will deteriorate completely. Those things are different. What we want in an AGI is that it behaves in a way that cannot be specified in advance, because if you specified it, you would already have the answer. The AGI program has to give unexpected answers, answers to questions we didn't even know how to ask."
3:11

.David Deutsch: "What's currently called AI and AGI are not only different from each other, they are very close to being the exact opposites of each other. The reason is that an AI, current AI is like an AI that diagnoses diseases or an AI that plays chess or an AI that controls a huge factory. Those things have objective functions, that is they have a function that they are designed to maximize and that is why they are used in those particular applications. Or in military terms, you could say the objective is to hit the target. You might say the objective is to hit the target unless some thing specified, but it's a specified thing comes up in which case don't hit the target and so on. This is, as I said, almost the opposite of what humans do when humans think. For a start, the AI has to be obedient, that is it has to actually do the things it is programmed to do, whereas a human is fundamentally disobedient, especially when being creative. When a human plays chess, they are performing a completely different kind of computation. They don't do the same things, they don't investigate the same possibilities that the artificial chess playing machine does, because the artificial one is capable of looking at billions and billions of possibilities, whereas the human can only look at hundreds or something. They are doing something completely different. Another difference is that the human can explain, can write a book later, having become world champion, can write a book saying how I did it, as the computer program that beats the world champion can write no such book, because it has no idea how it did it. It was just following a program. I was doing this and that and that and none of that is illuminating. Also, third thing, the chess player can decide I don't want to play chess anymore, from now on I will play Go or from now on I will play tennis. If commanded to play chess, the functionality will deteriorate completely. Those things are different. What we want in an AGI is that it behaves in a way that cannot be specified in advance, because if you specified it, you would already have the answer. The AGI program has to give unexpected answers, answers to questions we didn't even know how to ask."

Deutsch Explains

72,455 次观看 • 1 年前

One reason I stay a bit quite about AI these few months. I can’t find my voice to say this right. I’ll try. — What has my AI found just in the millions of never saved TV from the 1980s-1990s? The simple commercial, I have the largest collection known in the world is a reflection of hyper nuanced humanity. Frail and over optimistic as it is, it shows the sprit of humanity in ways that self-hating 2010 Internet sewage training AI can not. It is interesting to have dark humor, when you know the soul of the person with the humor. It is not interesting to have dark humor from an AI that was not trained on the human spirit. The human sprit is not math, coding and science. This is the OUTCOME of a human sprit. We are training AI backwards and we just can’t see it yet, but we will. I am doing may best as “some posted on X” who has no currency in the “big boy” AI world to save the human sprit. We will need it one day and most of this training material is being erased, trillions of bytes daily. This is the secret to truthful AI. This is the secret to AI that LOVES humanity. We don’t get this right soon, there will be little chance. The data is… Almost gone.
0:30

One reason I stay a bit quite about AI these few months. I can’t find my voice to say this right. I’ll try. — What has my AI found just in the millions of never saved TV from the 1980s-1990s? The simple commercial, I have the largest collection known in the world is a reflection of hyper nuanced humanity. Frail and over optimistic as it is, it shows the sprit of humanity in ways that self-hating 2010 Internet sewage training AI can not. It is interesting to have dark humor, when you know the soul of the person with the humor. It is not interesting to have dark humor from an AI that was not trained on the human spirit. The human sprit is not math, coding and science. This is the OUTCOME of a human sprit. We are training AI backwards and we just can’t see it yet, but we will. I am doing may best as “some posted on X” who has no currency in the “big boy” AI world to save the human sprit. We will need it one day and most of this training material is being erased, trillions of bytes daily. This is the secret to truthful AI. This is the secret to AI that LOVES humanity. We don’t get this right soon, there will be little chance. The data is… Almost gone.

Brian Roemmele

123,518 次观看 • 1 年前

To the Ivory Tower AI Doomers, AI is anti-human and job destructive. That's their choice. Our experience is AI makes human taste and talent exponentially more valuable. We've watched AI unleash human agency and creativity. That's how you give the American worker superpowers, and that's why we created the American Tech Fellowship. You want to bet on the American worker? Arm them with AI and let them BUILD.
1:23

To the Ivory Tower AI Doomers, AI is anti-human and job destructive. That's their choice. Our experience is AI makes human taste and talent exponentially more valuable. We've watched AI unleash human agency and creativity. That's how you give the American worker superpowers, and that's why we created the American Tech Fellowship. You want to bet on the American worker? Arm them with AI and let them BUILD.

Shyam Sankar

43,196 次观看 • 1 年前

The most interesting thing I've heard in a long time: Can AI ever be conscious? Ilya Sutskever Ilya Sutskever, co-founder at OpenAI, explained a simple experiment for AI consciousness. We carefully curate the data such that we never mention anything about consciousness...we would only give the AI controlled data about surface-level experiences and mundane objects. At some point, we would tell AI about the concept of consciousness. If the AI agrees and says it's been experiencing the same thing or recognizes the experience, then it would be convincing. Do you think AI can be conscious? If, so what is a good test for it? Is something like John Searle's Chinese Room argument sufficient?
2:41

The most interesting thing I've heard in a long time: Can AI ever be conscious? Ilya Sutskever Ilya Sutskever, co-founder at OpenAI, explained a simple experiment for AI consciousness. We carefully curate the data such that we never mention anything about consciousness...we would only give the AI controlled data about surface-level experiences and mundane objects. At some point, we would tell AI about the concept of consciousness. If the AI agrees and says it's been experiencing the same thing or recognizes the experience, then it would be convincing. Do you think AI can be conscious? If, so what is a good test for it? Is something like John Searle's Chinese Room argument sufficient?

Alex Ker 🔭

472,703 次观看 • 2 年前

AI Validates the Truth of Buddhism | Philosophy x AI In this interview, Imperial’s Murray Shanahan Murray Shanahan lays out a systematic procedure to validate machine consciousness. You might think the question of whether AIs are conscious is a trivial, theoretical fancy with no use in the real world. This interview will show you why it’s one of the most important questions with practical implications for not just how we treat AI systems but also how we build and align them. But the most unexpected reward of investigating AI consciousness turns out to be what it reveals about our consciousness, about the nature of the human self. Murray’s most interesting claim is that LLMs have an important buddhist lesson to teach all of us, namely, that there is no “us.” By better understanding the "no-self" of LLMs we are given a mirror to reflect on whether human selves are also an illusion. Timestamps: 08:26 Why Buddhist Enlightenment May be Easier for Software Programs 16:30 Super-Intelligent AI Might Not Want To Take Over the World 30:25 Wittgenstein’s Radical Take on Consciousness 42:55 What Video Games Teach Us About Consciousness 49:59 How the Movie Ex Machina Created a Better Test for AI Consciousness 1:02:36 What the Debate about AI Consciousness Is Missing 1:09:31 Does Conscious AI Have A Moral Right to Live? 1:15:01 Is the Human Brain Computationally Superior to AI? 1:26:59 Why Philosophy Is the Essential Skill for the AI Age
1:28:33

AI Validates the Truth of Buddhism | Philosophy x AI In this interview, Imperial’s Murray Shanahan Murray Shanahan lays out a systematic procedure to validate machine consciousness. You might think the question of whether AIs are conscious is a trivial, theoretical fancy with no use in the real world. This interview will show you why it’s one of the most important questions with practical implications for not just how we treat AI systems but also how we build and align them. But the most unexpected reward of investigating AI consciousness turns out to be what it reveals about our consciousness, about the nature of the human self. Murray’s most interesting claim is that LLMs have an important buddhist lesson to teach all of us, namely, that there is no “us.” By better understanding the "no-self" of LLMs we are given a mirror to reflect on whether human selves are also an illusion. Timestamps: 08:26 Why Buddhist Enlightenment May be Easier for Software Programs 16:30 Super-Intelligent AI Might Not Want To Take Over the World 30:25 Wittgenstein’s Radical Take on Consciousness 42:55 What Video Games Teach Us About Consciousness 49:59 How the Movie Ex Machina Created a Better Test for AI Consciousness 1:02:36 What the Debate about AI Consciousness Is Missing 1:09:31 Does Conscious AI Have A Moral Right to Live? 1:15:01 Is the Human Brain Computationally Superior to AI? 1:26:59 Why Philosophy Is the Essential Skill for the AI Age

Johnathan Bi

30,521 次观看 • 1 年前

What the AI oligarchs want to do is replace human workers. Well, some of us are old-fashioned, and we believe in human beings.
0:56

What the AI oligarchs want to do is replace human workers. Well, some of us are old-fashioned, and we believe in human beings.

Sen. Bernie Sanders

136,272 次观看 • 4 个月前