正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

EBM are so back! Yann LeCun has been pointing here for years: AI reasoning needs systems that check structure before they answer. Aleph from Logical Intelligence now leads the major formal reasoning benchmarks – let me explain what it is -> 📺

Ksenia Se

5,102 subscribers

58,510 次观看 • 4 个月前 •via X (Twitter)

健康养生教育科学技术

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

Whatever AI sceptics say, LLMs really can reason. They're not just doing an imitation that looks like reasoning, it's the real deal. But even though they are able to reason, sometimes they won't! If you ask an LLM a question it can't answer, sometimes it will just try to imitate reasoning without doing it. The chain of thought looks basically indistinguishable from actual reasoning. But under the hood something very different is going on. Trenton Bricken talked with me about what work on circuits inside LLMs has revealed:

Whatever AI sceptics say, LLMs really can reason. They're not just doing an imitation that looks like reasoning, it's the real deal. But even though they are able to reason, sometimes they won't! If you ask an LLM a question it can't answer, sometimes it will just try to imitate reasoning without doing it. The chain of thought looks basically indistinguishable from actual reasoning. But under the hood something very different is going on. Trenton Bricken talked with me about what work on circuits inside LLMs has revealed:

Dwarkesh Patel

104,954 次观看 • 3 个月前

I seriously wonder why Yann Lecun takes such different positions from Sam Altman and Dario Amodei. Sam and Dario have made it clear in their posts that they believe reasoning will be solved, if not now, then in a few years' time. What's more: in his last post, Dario said that AGI could be realized as early as 2026. And Sam Altman also says that ASI will come in a few thousand days. So why is Yann again talking about assuming decades? Does it have a more complex definition? Has Meta lost more ground after all? So far I don't have a well-founded answer.

I seriously wonder why Yann Lecun takes such different positions from Sam Altman and Dario Amodei. Sam and Dario have made it clear in their posts that they believe reasoning will be solved, if not now, then in a few years' time. What's more: in his last post, Dario said that AGI could be realized as early as 2026. And Sam Altman also says that ASI will come in a few thousand days. So why is Yann again talking about assuming decades? Does it have a more complex definition? Has Meta lost more ground after all? So far I don't have a well-founded answer.

Chubby♨️

457,863 次观看 • 1 年前

"We won't reach AGI with LLMs." Yann LeCun has been saying this for years. Now he's leaving Meta to prove it. LeCun invented convolutional neural networks—the tech behind every smartphone camera and self-driving car today. He won the Turing Award in 2018, AI's Nobel Prize. At 65, the leader of Meta's FAIR research lab is walking away from $600 billion in AI infrastructure, betting against the entire industry: Meta, OpenAI, Anthropic, xAI, Google. Who is Yann LeCun? Why is he leaving, and why does his next move matter? Here's the story:

"We won't reach AGI with LLMs." Yann LeCun has been saying this for years. Now he's leaving Meta to prove it. LeCun invented convolutional neural networks—the tech behind every smartphone camera and self-driving car today. He won the Turing Award in 2018, AI's Nobel Prize. At 65, the leader of Meta's FAIR research lab is walking away from $600 billion in AI infrastructure, betting against the entire industry: Meta, OpenAI, Anthropic, xAI, Google. Who is Yann LeCun? Why is he leaving, and why does his next move matter? Here's the story:

Computer

442,676 次观看 • 10 个月前

Jimmy Corsetti has been relentlessly pushing the Richat Structure as Atlantis for years now, strutting around like he’s single-handedly solved one of history’s greatest mysteries. 🥸 His theory suffers from a glaring, fatal flaw that makes it crumble under the slightest examination, let me explain why. #RealArchaeology

Jimmy Corsetti has been relentlessly pushing the Richat Structure as Atlantis for years now, strutting around like he’s single-handedly solved one of history’s greatest mysteries. 🥸 His theory suffers from a glaring, fatal flaw that makes it crumble under the slightest examination, let me explain why. #RealArchaeology

Flint Drivel🍖ᴾᵃʳᵒᵈʸ

22,888 次观看 • 3 个月前

Jensen Huang, CEO of NVIDIA, laid out the compute math almost nobody priced in: reasoning AI needs 1,000x more compute than generative AI. Then you multiply by 100x more users. "In order for AI to go through understanding, reasoning, planning, using tools to take action, the amount of computation necessary compared to generative AI is like a thousand times more." "In two years time, the number of cars you need in the world grew by a thousand times. In two years time, the number of airplanes you need in the world grew by 2,000 times. Whatever metric you want to use, the thousand times difference is incredible." "But then you multiply that by the number of people who now want to use it by a hundred times. Which is the reason why GPU consumption is going through the roof." "Even GPUs we sold 4 or 5 years ago are now rising in price faster than good wine. It has defied what everybody said was going to happen." The reasoning era did not make intelligence cheaper to run. It made a single answer cost 1,000x more compute, then handed that answer to 100x more people. Anyone still budgeting AI around "inference gets cheap" is planning for a curve that already snapped.

Jensen Huang, CEO of NVIDIA, laid out the compute math almost nobody priced in: reasoning AI needs 1,000x more compute than generative AI. Then you multiply by 100x more users. "In order for AI to go through understanding, reasoning, planning, using tools to take action, the amount of computation necessary compared to generative AI is like a thousand times more." "In two years time, the number of cars you need in the world grew by a thousand times. In two years time, the number of airplanes you need in the world grew by 2,000 times. Whatever metric you want to use, the thousand times difference is incredible." "But then you multiply that by the number of people who now want to use it by a hundred times. Which is the reason why GPU consumption is going through the roof." "Even GPUs we sold 4 or 5 years ago are now rising in price faster than good wine. It has defied what everybody said was going to happen." The reasoning era did not make intelligence cheaper to run. It made a single answer cost 1,000x more compute, then handed that answer to 100x more people. Anyone still budgeting AI around "inference gets cheap" is planning for a curve that already snapped.

Karl Mehta

54,717 次观看 • 2 个月前

Yann LeCun (AMI Labs Founder): "The AI industry is completely LLM-pilled. Everybody is working on the same thing. They're all digging the same trench." LeCun explains why no lab dares break from the pack: "They are stealing each other's engineers. So they can't afford to do something different because if they start going on a tangent, they're going to fall behind the other guys. And so they're all doing the same thing." This groupthink is exactly what drove him out of Meta. "Meta also became LLM-pilled with sort of recent reshuffling. And it's fine, a strategic decision that maybe makes sense for them. It's just not what I'm interested in." For Yann LeCun, the problem runs deeper than strategy. LLMs are missing something essential about how intelligence actually works: "I cannot imagine that we can build agentic systems without those systems having an ability to predict in advance what the consequences of their actions are going to be. The way we act in the world is that we can predict the consequences of our actions and that's what allows us to plan." His broader critique is that the industry has mistaken fluency for intelligence. Language turned out to be the easy part. The hard part is the physical world. It's why we still don't have domestic robots or level-five self-driving cars, even though today's systems can pass the bar exam and write code.

Yann LeCun (AMI Labs Founder): "The AI industry is completely LLM-pilled. Everybody is working on the same thing. They're all digging the same trench." LeCun explains why no lab dares break from the pack: "They are stealing each other's engineers. So they can't afford to do something different because if they start going on a tangent, they're going to fall behind the other guys. And so they're all doing the same thing." This groupthink is exactly what drove him out of Meta. "Meta also became LLM-pilled with sort of recent reshuffling. And it's fine, a strategic decision that maybe makes sense for them. It's just not what I'm interested in." For Yann LeCun, the problem runs deeper than strategy. LLMs are missing something essential about how intelligence actually works: "I cannot imagine that we can build agentic systems without those systems having an ability to predict in advance what the consequences of their actions are going to be. The way we act in the world is that we can predict the consequences of our actions and that's what allows us to plan." His broader critique is that the industry has mistaken fluency for intelligence. Language turned out to be the easy part. The hard part is the physical world. It's why we still don't have domestic robots or level-five self-driving cars, even though today's systems can pass the bar exam and write code.

Big Brain AI

289,053 次观看 • 4 个月前

Frontier Notes from AGI House: Jerry Tworek Jerry Tworek on why “the era of evals is done.” 𝐅𝐮𝐥𝐥 𝐢𝐧𝐭𝐞𝐫𝐯𝐢𝐞𝐰 ↓ Jerry Tworek spent seven years at OpenAI, where he worked on research that helped shape modern coding and reasoning models, including Codex and reinforcement learning for reasoning. He later served as VP of Research, and is now CEO of , building toward automated AI research. At AGI House, Jerry sat down with Rocky Yu for a technical conversation on Codex, HumanEval, Copilot, RL, tool use, test-time compute, and automated AI labs. We cover: › Why code became the first serious LLM vertical › What HumanEval taught the field about verifiable rewards › Why Jerry thinks “the era of evals is done” › How real-world deployment differs from static benchmarks › What GitHub Copilot taught OpenAI about product quality › How reinforcement learning shaped modern reasoning behaviors › Why tool use is “99% systems and 1% algorithms” › Whether test-time compute still has room to scale › What it takes to build automated AI research labs Watch the full interview below.

Frontier Notes from AGI House: Jerry Tworek Jerry Tworek on why “the era of evals is done.” 𝐅𝐮𝐥𝐥 𝐢𝐧𝐭𝐞𝐫𝐯𝐢𝐞𝐰 ↓ Jerry Tworek spent seven years at OpenAI, where he worked on research that helped shape modern coding and reasoning models, including Codex and reinforcement learning for reasoning. He later served as VP of Research, and is now CEO of , building toward automated AI research. At AGI House, Jerry sat down with Rocky Yu for a technical conversation on Codex, HumanEval, Copilot, RL, tool use, test-time compute, and automated AI labs. We cover: › Why code became the first serious LLM vertical › What HumanEval taught the field about verifiable rewards › Why Jerry thinks “the era of evals is done” › How real-world deployment differs from static benchmarks › What GitHub Copilot taught OpenAI about product quality › How reinforcement learning shaped modern reasoning behaviors › Why tool use is “99% systems and 1% algorithms” › Whether test-time compute still has room to scale › What it takes to build automated AI research labs Watch the full interview below.

AGI House

12,497 次观看 • 1 个月前

Godfather of AI Yann LeCun reveals the humanoid robot industry's secret: the companies building the robots have no idea how to make them smart enough to be useful. "There is a lot of expectations that robotics is going to be a revolution, essentially, in a big market, eventually." "But the secret of the industry is that none of the companies at the moment that are building robots, like humanoid robots, none of those companies has any idea how to make those robots smart enough to be useful." "There's a big push for AI to make progress towards... understanding the physical world. That's what is prompting this investment." "There's nothing wrong with investing in LLMs. LLMs are useful. They're just addressing a completely different set of applications, really." LeCun is not describing someone else's problem. He has spent 15 years on world models, raised more than $1 billion for AMI Labs to build that missing intelligence layer, and in the same interview he calls 2026 the year of the world model and puts consumer robots 3 to 5 years out. Every humanoid robot demo you have seen is a body waiting for a brain. The winner in robotics will not be the company that shipped the body; it will be the one that ships the intelligence. - Yann LeCun (Yann LeCun), Turing Award winner and founder of AMI Labs, on Bloomberg Television.

Godfather of AI Yann LeCun reveals the humanoid robot industry's secret: the companies building the robots have no idea how to make them smart enough to be useful. "There is a lot of expectations that robotics is going to be a revolution, essentially, in a big market, eventually." "But the secret of the industry is that none of the companies at the moment that are building robots, like humanoid robots, none of those companies has any idea how to make those robots smart enough to be useful." "There's a big push for AI to make progress towards... understanding the physical world. That's what is prompting this investment." "There's nothing wrong with investing in LLMs. LLMs are useful. They're just addressing a completely different set of applications, really." LeCun is not describing someone else's problem. He has spent 15 years on world models, raised more than $1 billion for AMI Labs to build that missing intelligence layer, and in the same interview he calls 2026 the year of the world model and puts consumer robots 3 to 5 years out. Every humanoid robot demo you have seen is a body waiting for a brain. The winner in robotics will not be the company that shipped the body; it will be the one that ships the intelligence. - Yann LeCun (Yann LeCun), Turing Award winner and founder of AMI Labs, on Bloomberg Television.

Karl Mehta

44,534 次观看 • 2 个月前

Part 1 of 5. Building Systems. For 251 years, the Army has changed to meet the needs of the Nation. Today, that means changing the way we build. We are fixing the systems behind the force, investing where it matters, and moving faster so Soldiers have what they need to win.

Part 1 of 5. Building Systems. For 251 years, the Army has changed to meet the needs of the Nation. Today, that means changing the way we build. We are fixing the systems behind the force, investing where it matters, and moving faster so Soldiers have what they need to win.

Acting Secretary of the Army

20,814 次观看 • 3 个月前

Bill Gates predicts major white-collar disruption within 2 years, and AI's reliability factor is be the last major barrier stopping AI from taking many jobs "There are a lot of jobs that the only thing holding AI back from taking is that organizing the systems to achieve extreme reliability has been difficult. And I see over the next few years both people who do the implementations and the richness of the models themselves completely eliminating that unreliability barrier." ---- From "The Atlantic" YouTube channel, (full video link in comment)

Bill Gates predicts major white-collar disruption within 2 years, and AI's reliability factor is be the last major barrier stopping AI from taking many jobs "There are a lot of jobs that the only thing holding AI back from taking is that organizing the systems to achieve extreme reliability has been difficult. And I see over the next few years both people who do the implementations and the richness of the models themselves completely eliminating that unreliability barrier." ---- From "The Atlantic" YouTube channel, (full video link in comment)

Rohan Paul

31,521 次观看 • 6 天前

Is this real or AI? Even a year ago that wasn't what made you stop the scroll...but now it is. That's why we built Resemble AI Deepfake Detector for Chrome, available here (for free) > Deepfakes are everywhere, but so are we.

Is this real or AI? Even a year ago that wasn't what made you stop the scroll...but now it is. That's why we built Resemble AI Deepfake Detector for Chrome, available here (for free) > Deepfakes are everywhere, but so are we.

Resemble AI

25,451,665 次观看 • 4 个月前

The people building AI say it could destroy humanity… so why are they still racing to build it? To help answer that, I’m joined by Karen Hao. She’s someone who’s spent years speaking to the people at the centre of this race, conducting over 300 interviews, including insiders at OpenAI to understand what’s really going on. Karen is a journalist who has been covering artificial intelligence for years, looking closely at how these systems are built and what’s really driving them. As we get into this conversation, what becomes clear is how much of the story we hear about AI is controlled by a small number of people - and how difficult it is to separate what’s being said publicly from what’s actually happening behind the scenes.

The people building AI say it could destroy humanity… so why are they still racing to build it? To help answer that, I’m joined by Karen Hao. She’s someone who’s spent years speaking to the people at the centre of this race, conducting over 300 interviews, including insiders at OpenAI to understand what’s really going on. Karen is a journalist who has been covering artificial intelligence for years, looking closely at how these systems are built and what’s really driving them. As we get into this conversation, what becomes clear is how much of the story we hear about AI is controlled by a small number of people - and how difficult it is to separate what’s being said publicly from what’s actually happening behind the scenes.

Steven Bartlett

38,676 次观看 • 5 个月前

Before founding Vals AI, Rayan Krishnan was a college radio host. We're thrilled to get him back in his element with Vivek Gopalan, this time talking all things benchmarks and evals. As AI capabilities compound, measuring them is becoming both harder and more important. What started as an academic exercise is quickly evolving into critical infrastructure for understanding what AI systems can actually do, and how they perform in the real world. Rayan and Vivek go deep on how benchmarks are evolving, measuring increasingly agentic systems, the weird and wonderful ways we test intelligence, the global AI race, and why evals sit on the critical path to ASI. 0:44 Importance and evolution of benchmarks 8:16 From academic artifact → real-world value 10:57 The next generation of benchmarks 13:34 Benchmarking models vs. agents 16:20 Building benchmarks for organizations 22:09 Weird and wonderful benchmarks 25:58 The AI arms race: US vs. China 31:02 Why evals are on the critical path to ASI 32:50 Vals Series A

Before founding Vals AI, Rayan Krishnan was a college radio host. We're thrilled to get him back in his element with Vivek Gopalan, this time talking all things benchmarks and evals. As AI capabilities compound, measuring them is becoming both harder and more important. What started as an academic exercise is quickly evolving into critical infrastructure for understanding what AI systems can actually do, and how they perform in the real world. Rayan and Vivek go deep on how benchmarks are evolving, measuring increasingly agentic systems, the weird and wonderful ways we test intelligence, the global AI race, and why evals sit on the critical path to ASI. 0:44 Importance and evolution of benchmarks 8:16 From academic artifact → real-world value 10:57 The next generation of benchmarks 13:34 Benchmarking models vs. agents 16:20 Building benchmarks for organizations 22:09 Weird and wonderful benchmarks 25:58 The AI arms race: US vs. China 31:02 Why evals are on the critical path to ASI 32:50 Vals Series A

8VC

104,829 次观看 • 1 个月前

Colston Loveland has been heavily involved with SpecialOlympicsIdaho, from donating shoes to athletes to coaching them out on the field. His reasoning behind it? Is simple. "I think that just gives me so much motivation knowing that, you know, on a hard day when I think, you know, man, I don't really want to do this, just thinking back to that, like, man, these guys are doing it." KMVT Sports

Colston Loveland has been heavily involved with SpecialOlympicsIdaho, from donating shoes to athletes to coaching them out on the field. His reasoning behind it? Is simple. "I think that just gives me so much motivation knowing that, you know, on a hard day when I think, you know, man, I don't really want to do this, just thinking back to that, like, man, these guys are doing it." KMVT Sports

Debanny Cerda

16,862 次观看 • 4 个月前

My second album Paradise Now! IS OUT NOW! These words feel immense, bigger than me! And I can’t say what it means from here, but what I do know is it that was made from the heart, a true snap shot where my head has been over the last couple years.

My second album Paradise Now! IS OUT NOW! These words feel immense, bigger than me! And I can’t say what it means from here, but what I do know is it that was made from the heart, a true snap shot where my head has been over the last couple years.

OBONGJAYAR, DEVIL SLAYER

21,329 次观看 • 1 年前

"World models" is one of the buzziest yet ambiguous terms in AI right now. I started this video with many questions: - How are they different from video generation? - Can they do more than AI slop? - Can LeCun be trusted given that he wears knee-high white socks? Many thanks to TJ Galda and NVIDIA AI for helping me answer (most) of these questions!

"World models" is one of the buzziest yet ambiguous terms in AI right now. I started this video with many questions: - How are they different from video generation? - Can they do more than AI slop? - Can LeCun be trusted given that he wears knee-high white socks? Many thanks to TJ Galda and NVIDIA AI for helping me answer (most) of these questions!

Julia Turc

248,493 次观看 • 3 个月前

The CEO of Google DeepMind just went on record saying he disagrees with one of the most respected AI researchers in the world. Demis Hassabis, the man behind AlphaFold, AlphaGo, and Google's entire AI operation publicly pushed back against Yann LeCun's claim that large language models are a dead end for artificial intelligence. LeCun, who left Meta earlier this year to start his own AI lab, has been saying for years that LLMs cannot reason, cannot plan, and will never get us to human-level intelligence. Hassabis disagrees, and he said so directly. His position is that scaling laws are still working, foundation models are still getting more capable, and whatever AGI ends up looking like, LLMs will be a central part of it, not something that gets replaced. He does say there is roughly a 50/50 chance that one or two additional breakthroughs will be needed beyond scaling alone, things like better memory, long-term planning, and world models. But the core disagreement with LeCun is clear, Hassabis believes the current architecture is sound and the current path leads somewhere real. Two Nobel-recognized researchers, two founding figures of modern AI, now publicly on opposite sides of the most important technical question in the industry.

The CEO of Google DeepMind just went on record saying he disagrees with one of the most respected AI researchers in the world. Demis Hassabis, the man behind AlphaFold, AlphaGo, and Google's entire AI operation publicly pushed back against Yann LeCun's claim that large language models are a dead end for artificial intelligence. LeCun, who left Meta earlier this year to start his own AI lab, has been saying for years that LLMs cannot reason, cannot plan, and will never get us to human-level intelligence. Hassabis disagrees, and he said so directly. His position is that scaling laws are still working, foundation models are still getting more capable, and whatever AGI ends up looking like, LLMs will be a central part of it, not something that gets replaced. He does say there is roughly a 50/50 chance that one or two additional breakthroughs will be needed beyond scaling alone, things like better memory, long-term planning, and world models. But the core disagreement with LeCun is clear, Hassabis believes the current architecture is sound and the current path leads somewhere real. Two Nobel-recognized researchers, two founding figures of modern AI, now publicly on opposite sides of the most important technical question in the industry.

Milk Road AI

130,937 次观看 • 5 个月前

AI has transformed how video is created. We think the next wave is about understanding it. Over the past few years, we've seen remarkable advances in video generation, editing, avatars, and creative tooling. An increasingly important problem is teaching machines to search, analyze, reason over, and extract insight from video - across massive libraries and live streams alike. We're calling this video intelligence, and we're actively looking to back founders building here. We're most excited about companies pushing on the core capabilities: - Video-native models - multimodal embeddings, temporal reasoning, and retrieval built specifically for video rather than adapted from image or text - Real-time and large-scale pipelines - infrastructure for processing, indexing, and querying video at the speed and scale enterprises actually need - Agentic and reasoning layers - systems that don't just retrieve clips but answer questions, surface anomalies, and take action on what they see The models and infrastructure to make this real are appearing to be crossing a capability threshold right now. Multimodal foundation models are maturing, storage costs have collapsed, and enterprises are sitting on years of unstructured video with no way to use it. That infrastructure unlocks a wide range of applications including media and sports workflows, security and physical operations, enterprise knowledge management, advertising analytics, robotics, and consumer products, where video has historically been dark data. If you're building in video intelligence at the model layer, the platform layer, or in a vertical application, we'd love to talk!

AI has transformed how video is created. We think the next wave is about understanding it. Over the past few years, we've seen remarkable advances in video generation, editing, avatars, and creative tooling. An increasingly important problem is teaching machines to search, analyze, reason over, and extract insight from video - across massive libraries and live streams alike. We're calling this video intelligence, and we're actively looking to back founders building here. We're most excited about companies pushing on the core capabilities: - Video-native models - multimodal embeddings, temporal reasoning, and retrieval built specifically for video rather than adapted from image or text - Real-time and large-scale pipelines - infrastructure for processing, indexing, and querying video at the speed and scale enterprises actually need - Agentic and reasoning layers - systems that don't just retrieve clips but answer questions, surface anomalies, and take action on what they see The models and infrastructure to make this real are appearing to be crossing a capability threshold right now. Multimodal foundation models are maturing, storage costs have collapsed, and enterprises are sitting on years of unstructured video with no way to use it. That infrastructure unlocks a wide range of applications including media and sports workflows, security and physical operations, enterprise knowledge management, advertising analytics, robotics, and consumer products, where video has historically been dark data. If you're building in video intelligence at the model layer, the platform layer, or in a vertical application, we'd love to talk!

Jason Cui

36,205 次观看 • 4 个月前