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。

Sam Altman says GPT-5 is superhuman at knowledge, pattern recognition, and recall -- but still struggles with long-term thinking it can now solve Olympiad-level math problems that take 90 minutes, but proving a new Math theorem, which takes 1,000 hours? "we're not close"

Haider.

66,948 subscribers

345,910 次观看 • 11 个月前 •via X (Twitter)

科学技术教育

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

📁 Demis Hassabis says that one of the main reasons we are not yet at AGI is the lack of consistency. He explains that current systems can solve International Math Olympiad level problems but still fail at basic logic tasks or even chess. They are uneven intelligences, exceptional in some areas and surprisingly weak in others, which falls short of true AGI.

📁 Demis Hassabis says that one of the main reasons we are not yet at AGI is the lack of consistency. He explains that current systems can solve International Math Olympiad level problems but still fail at basic logic tasks or even chess. They are uneven intelligences, exceptional in some areas and surprisingly weak in others, which falls short of true AGI.

Jon Hernandez

10,541 次观看 • 7 个月前

Sam Altman says GPT-5 is not finished yet, but he expects it to be a "significant leap forward" and better at a wider variety of tasks than GPT-4

Sam Altman says GPT-5 is not finished yet, but he expects it to be a "significant leap forward" and better at a wider variety of tasks than GPT-4

Tsarathustra

111,151 次观看 • 2 年前

Terence Tao thinks AI is already very good at using existing, well-understood math techniques to solve problems. An important question is how many open problems in math could be solved this way, without developing any new ideas. An extreme case of a proof like this is the four-color theorem, which was proved by checking a thousand examples one-by-one. If there are lots of problems that are vulnerable to just mechanically applying existing techniques, AI could start to produce new math results incredibly rapidly. But these proofs might not help advance our civilization’s knowledge the way math has in the past.

Terence Tao thinks AI is already very good at using existing, well-understood math techniques to solve problems. An important question is how many open problems in math could be solved this way, without developing any new ideas. An extreme case of a proof like this is the four-color theorem, which was proved by checking a thousand examples one-by-one. If there are lots of problems that are vulnerable to just mechanically applying existing techniques, AI could start to produce new math results incredibly rapidly. But these proofs might not help advance our civilization’s knowledge the way math has in the past.

Dwarkesh Patel

64,547 次观看 • 4 个月前

Sam Altman on GPT 6 • The GDPv eval will shift some of the way we do post training • overall we are keeping the same strategy going into GPT 6 Sam Altman on AGI: Sam Altman about AGI: "we are finally in the moment": "As the closer we get to it, the fuzzier the Agi concept becomes. But the one thing that I care the most about and, to my great surprise, we're finally in the moment; where it's starting to happen is, when it can do novel discovery, when It can expand the total human knowledge base.“ I have seen some examples from credible professors, that novel math problems have been solved. Such as GPT 5 Pro reportedly solving Yu Tsumura’s 554th problem, an IMO level group theory challenge that an August preprint said no LLM had solved, so the new claim is being actively contested. Sébastien Bubeck also said GPT 5 Pro produced a new proof that tightened a bound in smooth convex optimization.

Sam Altman on GPT 6 • The GDPv eval will shift some of the way we do post training • overall we are keeping the same strategy going into GPT 6 Sam Altman on AGI: Sam Altman about AGI: "we are finally in the moment": "As the closer we get to it, the fuzzier the Agi concept becomes. But the one thing that I care the most about and, to my great surprise, we're finally in the moment; where it's starting to happen is, when it can do novel discovery, when It can expand the total human knowledge base.“ I have seen some examples from credible professors, that novel math problems have been solved. Such as GPT 5 Pro reportedly solving Yu Tsumura’s 554th problem, an IMO level group theory challenge that an August preprint said no LLM had solved, so the new claim is being actively contested. Sébastien Bubeck also said GPT 5 Pro produced a new proof that tightened a bound in smooth convex optimization.

Chris

221,718 次观看 • 10 个月前

Sam Altman says AI reasoning is still at the GPT-2 stage but the improvement curve is steep and the new o1 model represents a new paradigm of AI development which will enable rapid progress in capabilities

Sam Altman says AI reasoning is still at the GPT-2 stage but the improvement curve is steep and the new o1 model represents a new paradigm of AI development which will enable rapid progress in capabilities

Tsarathustra

419,003 次观看 • 1 年前

📁 Sam Altman says that even though building software is now dramatically easier, building a real business is still just as hard. AI removes technical barriers, but it does not solve the scarcest resource, human attention. Even in a world of radical abundance, distribution and focus remain the bottleneck. The edge will not be speed, but creativity and building things that truly matter.

📁 Sam Altman says that even though building software is now dramatically easier, building a real business is still just as hard. AI removes technical barriers, but it does not solve the scarcest resource, human attention. Even in a world of radical abundance, distribution and focus remain the bottleneck. The edge will not be speed, but creativity and building things that truly matter.

Jon Hernandez

44,084 次观看 • 6 个月前

📁 Sam Altman says that a truly intelligent AI will filter information, understand your context and manage your life with clarity and calm. He explains that with trust and long term use, it can create a peaceful and distraction free experience. He adds that this level of digital serenity is impossible with pre AI technology.

📁 Sam Altman says that a truly intelligent AI will filter information, understand your context and manage your life with clarity and calm. He explains that with trust and long term use, it can create a peaceful and distraction free experience. He adds that this level of digital serenity is impossible with pre AI technology.

Jon Hernandez

33,285 次观看 • 8 个月前

Demis Hassabis explains that tasks like coding and math are easier to automate because they offer clear, verifiable outputs. However, scientific discovery like creating new hypotheses or theories is much harder for AI. It’s not just about solving known problems, but inventing new questions, which requires a higher level of creativity. He believes we're still missing key capabilities needed for that level of AI. While progress is rapid, reaching true scientific innovation without human input remains uncertain.

Demis Hassabis explains that tasks like coding and math are easier to automate because they offer clear, verifiable outputs. However, scientific discovery like creating new hypotheses or theories is much harder for AI. It’s not just about solving known problems, but inventing new questions, which requires a higher level of creativity. He believes we're still missing key capabilities needed for that level of AI. While progress is rapid, reaching true scientific innovation without human input remains uncertain.

Wes Roth

15,090 次观看 • 6 个月前

Sam Altman says OpenAI strategy is to solve AI first, then connect it with robotics "imagines a world where premium ChatGPT users get a humanoid robot included" Mechanical and cognitive hurdles remain, but they're within reach. A billion robots will take time -- but maybe a million can automate the whole supply chain

Sam Altman says OpenAI strategy is to solve AI first, then connect it with robotics "imagines a world where premium ChatGPT users get a humanoid robot included" Mechanical and cognitive hurdles remain, but they're within reach. A billion robots will take time -- but maybe a million can automate the whole supply chain

Haider.

101,999 次观看 • 1 年前

Jordi: It seems pretty obvious that the models just getting smarter is not by default going to make adoption accelerate massively again. If you're a business that's not doing code-gen, not generating a bunch of assets, etc. - and somebody says, "For $20 a month, you get someone that's PhD level in math, but extremely low agency." A laundromat would be like, "Actually, I don't need somebody who's PhD level at math. Sir this is a laundromat."

Jordi: It seems pretty obvious that the models just getting smarter is not by default going to make adoption accelerate massively again. If you're a business that's not doing code-gen, not generating a bunch of assets, etc. - and somebody says, "For $20 a month, you get someone that's PhD level in math, but extremely low agency." A laundromat would be like, "Actually, I don't need somebody who's PhD level at math. Sir this is a laundromat."

TBPN

76,844 次观看 • 7 个月前

$TSLA Tesla still has a shot at $500+ and new ATHs. But it needs to happen now and not stall out like previous attempts that led to breakdowns and corrections. Remember, I’m still looking to time Tesla’s next major trading range that eventually takes it toward $1,000+.

$TSLA Tesla still has a shot at $500+ and new ATHs. But it needs to happen now and not stall out like previous attempts that led to breakdowns and corrections. Remember, I’m still looking to time Tesla’s next major trading range that eventually takes it toward $1,000+.

The Swing Trader

64,462 次观看 • 6 个月前

📁 Kevin Weil, Head of Product at OpenAI, says models are already solving frontier math problems humans had not cracked. What starts at 5% capability becomes 80% in a year. We are in that phase now with science. AGI may not feel real inside a chat box. It will feel real when it discovers new materials, cures disease and accelerates the science of 2050 into 2030.

📁 Kevin Weil, Head of Product at OpenAI, says models are already solving frontier math problems humans had not cracked. What starts at 5% capability becomes 80% in a year. We are in that phase now with science. AGI may not feel real inside a chat box. It will feel real when it discovers new materials, cures disease and accelerates the science of 2050 into 2030.

Jon Hernandez

13,702 次观看 • 5 个月前

This thread reminded me of Thiel's take on biology and the excuses of biologists for why we aren't solving hard problems (I have a biology background and work as a scientist and still agree with him). My take is that AI won't solve biology, but it will help us understand it at a deeper level. Saying that it is impossible is a lazy take. Saying that AI will solve all problems is a lazy take too. There is an in-between where AI will support and help scientists move things forward. I am really tired of the excuses of the gatekeeper boomer scientists and the hype of the AI bros.

This thread reminded me of Thiel's take on biology and the excuses of biologists for why we aren't solving hard problems (I have a biology background and work as a scientist and still agree with him). My take is that AI won't solve biology, but it will help us understand it at a deeper level. Saying that it is impossible is a lazy take. Saying that AI will solve all problems is a lazy take too. There is an in-between where AI will support and help scientists move things forward. I am really tired of the excuses of the gatekeeper boomer scientists and the hype of the AI bros.

Linden

67,241 次观看 • 1 个月前

A NOBEL WINNING PHYSICIST ARGUED THAT NO AI, NO MATTER HOW POWERFUL, WILL EVER TRULY UNDERSTAND A SINGLE THING IT SAYS. HIS REASON IS NOT COMPUTE OR DATA -- IT IS A MATH THEOREM FROM THE 1930s THAT SAYS SOME TRUTHS CAN BE SEEN BUT NEVER COMPUTED 81 minutes with Roger Penrose -- the Oxford physicist who won a Nobel for his work on black holes and general relativity. -> His claim: whatever consciousness is, it is not a computation. A machine following rules can imitate understanding, but never actually have it. He builds it on Gödel: a human can just "see" that certain statements are true, even though no algorithm can ever prove them. That seeing, he argues, is non-computational. If he is right, understanding is not something you scale into. You can stack a trillion parameters and still have zero awareness underneath. Which cuts straight through the AI moment. Everyone assumes bigger models will eventually "wake up". Penrose says that is a category error -- more computation is still just computation. You thought intelligence and understanding were the same thing. This is the conversation that pulls them apart. Save this. It is the sharpest case against the hype ↓

A NOBEL WINNING PHYSICIST ARGUED THAT NO AI, NO MATTER HOW POWERFUL, WILL EVER TRULY UNDERSTAND A SINGLE THING IT SAYS. HIS REASON IS NOT COMPUTE OR DATA -- IT IS A MATH THEOREM FROM THE 1930s THAT SAYS SOME TRUTHS CAN BE SEEN BUT NEVER COMPUTED 81 minutes with Roger Penrose -- the Oxford physicist who won a Nobel for his work on black holes and general relativity. -> His claim: whatever consciousness is, it is not a computation. A machine following rules can imitate understanding, but never actually have it. He builds it on Gödel: a human can just "see" that certain statements are true, even though no algorithm can ever prove them. That seeing, he argues, is non-computational. If he is right, understanding is not something you scale into. You can stack a trillion parameters and still have zero awareness underneath. Which cuts straight through the AI moment. Everyone assumes bigger models will eventually "wake up". Penrose says that is a category error -- more computation is still just computation. You thought intelligence and understanding were the same thing. This is the conversation that pulls them apart. Save this. It is the sharpest case against the hype ↓

slash1s

275,880 次观看 • 1 个月前

Baylor HC Dave Aranda - Defensive Philosophy - Selection Not Compression - Change the Math - Shrink the Run Area "Keep it simple enough so the player can grow. The player can take over the game." "It's a game of 11v11, but the person with the most at the point of attack is going to win. We can do that with fundamentals, but we can also do that with scheme."

Baylor HC Dave Aranda - Defensive Philosophy - Selection Not Compression - Change the Math - Shrink the Run Area "Keep it simple enough so the player can grow. The player can take over the game." "It's a game of 11v11, but the person with the most at the point of attack is going to win. We can do that with fundamentals, but we can also do that with scheme."

James Light

28,931 次观看 • 1 年前

Never bet against Cursor. Subagents solve so many issues between different models it's kind of insane, especially with Composer 1. Composer 1 can quickly gather context for a model like GPT-5.2 Codex, which otherwise could take on the order of 10 mins just to search files. Definitely one of the most underrated models right now. Now you get the benefits of it's speed, but can handoff to a smarter model at implementation. The video I attached is just a simple example, but you can see how long GPT 5.2 takes to gather context on a very simple problem. For harder problems, the search time grows significantly.

Never bet against Cursor. Subagents solve so many issues between different models it's kind of insane, especially with Composer 1. Composer 1 can quickly gather context for a model like GPT-5.2 Codex, which otherwise could take on the order of 10 mins just to search files. Definitely one of the most underrated models right now. Now you get the benefits of it's speed, but can handoff to a smarter model at implementation. The video I attached is just a simple example, but you can see how long GPT 5.2 takes to gather context on a very simple problem. For harder problems, the search time grows significantly.

ostyn

19,631 次观看 • 6 个月前

A.J. Brown says he is still in awe about being a Patriot and that he caught himself not paying attention during practice at one point because he was thinking about his new situation: “Damn, I’m a Patriot… I know this ain’t heaven, but it’s close to it.”

A.J. Brown says he is still in awe about being a Patriot and that he caught himself not paying attention during practice at one point because he was thinking about his new situation: “Damn, I’m a Patriot… I know this ain’t heaven, but it’s close to it.”

Justin Turpin

49,140 次观看 • 2 个月前

The way math is presented can be really misleading about the actual nature of math. In theory, someone could have discovered quantum information theory decades before quantum mechanics even existed. It's just about combining linear algebra with probability. But nobody would have known it was worth combining these fields - or how, exactly, to combine them - before QM. Math papers and textbooks go rigourously step-by-step, starting with axioms and definitions, then deducing theorems. It's great for making sure proofs can be verified and ensuring the subject is based on sound logic. But for learners, it hides the really important thing: where did the axioms come from? Why these definitions and not others? The thing that's missing is the sense that there were concrete problems that these ideas helped solve.

The way math is presented can be really misleading about the actual nature of math. In theory, someone could have discovered quantum information theory decades before quantum mechanics even existed. It's just about combining linear algebra with probability. But nobody would have known it was worth combining these fields - or how, exactly, to combine them - before QM. Math papers and textbooks go rigourously step-by-step, starting with axioms and definitions, then deducing theorems. It's great for making sure proofs can be verified and ensuring the subject is based on sound logic. But for learners, it hides the really important thing: where did the axioms come from? Why these definitions and not others? The thing that's missing is the sense that there were concrete problems that these ideas helped solve.

Dwarkesh Patel

58,236 次观看 • 1 个月前

Why do we do math in school? When I asked this question as a kid, my teachers always told me, "You'll use it one day." That's almost always false. Most adults are never going to be in a sticky situation with two binomials, thinking, thank goodness I can FOIL my way out of this mess. Most of us get by just fine without ever solving for x again. And especially in an Al age, fewer and fewer people will "use math" in the narrow, practical sense. But the origins of math have never been primarily about utility. Math is formative. It trains the mind to love what is true, to recognize what is orderly, and to be drawn toward what is beautiful. It teaches us that the universe is not chaos, but something intelligible, something structured, something that can be known. And you can see that truth made visible in the world's most breathtaking churches and cathedrals: arches, vaults, domes, proportions, symmetry, harmony, light-geometry turned into glory. The mathematics that shapes a cathedral is not cold or sterile. It's the language of wonder, carved into stone. We don't teach math because everyone will use it. We teach math because it forms the kind of person who can see that reality has meaning.

Why do we do math in school? When I asked this question as a kid, my teachers always told me, "You'll use it one day." That's almost always false. Most adults are never going to be in a sticky situation with two binomials, thinking, thank goodness I can FOIL my way out of this mess. Most of us get by just fine without ever solving for x again. And especially in an Al age, fewer and fewer people will "use math" in the narrow, practical sense. But the origins of math have never been primarily about utility. Math is formative. It trains the mind to love what is true, to recognize what is orderly, and to be drawn toward what is beautiful. It teaches us that the universe is not chaos, but something intelligible, something structured, something that can be known. And you can see that truth made visible in the world's most breathtaking churches and cathedrals: arches, vaults, domes, proportions, symmetry, harmony, light-geometry turned into glory. The mathematics that shapes a cathedral is not cold or sterile. It's the language of wonder, carved into stone. We don't teach math because everyone will use it. We teach math because it forms the kind of person who can see that reality has meaning.

Jeremy Wayne Tate

95,173 次观看 • 6 个月前