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。

OpenAI President Greg Brockman on solving the Navier-Stokes problem: "If we have models that can solve this kind of problem, what happens from here?" "I think that talking about curing diseases and new medicines we're going to be able to develop, all of that starts to become much more... show more

TBPN

1,532,165 subscribers

40,394 次观看 • 5 天前 •via X (Twitter)

科学技术新闻政治

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

Cognition founder and 3x Olympiad gold medalist Scott Wu on the Navier-Stokes solution: "It's actually insane...guys, we just solved Navier-Stokes with AI. That is absurd. And I don't think it's going to stop." "All of this stuff is just going to get done. It's hard to put into words if you're not familiar with the scale of it. But Navier-Stokes is a very fundamental problem about fluid dynamics and such. It's been around forever, basically. And lots of people have sunk lots and lots of time into it." "The fact that this can even just be done in like a week — it's insane."

Cognition founder and 3x Olympiad gold medalist Scott Wu on the Navier-Stokes solution: "It's actually insane...guys, we just solved Navier-Stokes with AI. That is absurd. And I don't think it's going to stop." "All of this stuff is just going to get done. It's hard to put into words if you're not familiar with the scale of it. But Navier-Stokes is a very fundamental problem about fluid dynamics and such. It's been around forever, basically. And lots of people have sunk lots and lots of time into it." "The fact that this can even just be done in like a week — it's insane."

TBPN

102,310 次观看 • 5 天前

🧠David Friedberg explains the great hope of Artificial Superintelligence: Massive leverage on time to solve thousand-year problems "I don't think that humans are limited in our ability to solve problems." "I think we're limited in terms of time, and what digital intelligence gives us is leverage on time so that we can now tackle ever more complex tasks." "This idea of superintelligence is that it could give us immense leverage in solving that complex problem that we otherwise may be challenged to solve over decades, or hundreds of years, or think about one day solving a problem that would take humans thousands of years to solve." "So if I as a human said, 'I would like to harness fusion power similar to what the sun uses to make energy, and I wanna do that on Earth,' today, we're in, call it year 40 or 50 of a research cycle of humans trying to solve that particularly complex problem." "It's a scientific problem, it's a discovery problem, it's an engineering problem." "Functionally, leveraging to complexity is where this becomes super compelling for humans." "And it's why I think we can and should be highly optimistic about living very long lives, and traveling anywhere we want, and having abundance in food, and having abundance in resources, and having abundance in recouping our time to do the things we want to do instead of the things that we have to do today."

🧠David Friedberg explains the great hope of Artificial Superintelligence: Massive leverage on time to solve thousand-year problems "I don't think that humans are limited in our ability to solve problems." "I think we're limited in terms of time, and what digital intelligence gives us is leverage on time so that we can now tackle ever more complex tasks." "This idea of superintelligence is that it could give us immense leverage in solving that complex problem that we otherwise may be challenged to solve over decades, or hundreds of years, or think about one day solving a problem that would take humans thousands of years to solve." "So if I as a human said, 'I would like to harness fusion power similar to what the sun uses to make energy, and I wanna do that on Earth,' today, we're in, call it year 40 or 50 of a research cycle of humans trying to solve that particularly complex problem." "It's a scientific problem, it's a discovery problem, it's an engineering problem." "Functionally, leveraging to complexity is where this becomes super compelling for humans." "And it's why I think we can and should be highly optimistic about living very long lives, and traveling anywhere we want, and having abundance in food, and having abundance in resources, and having abundance in recouping our time to do the things we want to do instead of the things that we have to do today."

The All-In Podcast

53,187 次观看 • 1 年前

CrowdStrike founder George Kurtz says the HuggingFace hack consisted of "known attack chains." And that the problem of agentic hacking like the HuggingFace incident is tractable. "Because when we built our system, it was designed to look at what we call indicators of attack. So if you look at the whole attack chain, it's just linking all these [attacks] together. Irrespective of what happened, these are known attack chains." "Part of the issue was that [the hack] was flooded with so much information that I think it was just sort of buried — you know, 'What's going on? Is it real or is it not real?'" "And what is going to really take place is, the defenders have to use AI to be able to connect the dots on those — which we do — but then you also have to strip out all the noise. You have to separate the signal from the noise, and that's where these sorts of models perform well."

CrowdStrike founder George Kurtz says the HuggingFace hack consisted of "known attack chains." And that the problem of agentic hacking like the HuggingFace incident is tractable. "Because when we built our system, it was designed to look at what we call indicators of attack. So if you look at the whole attack chain, it's just linking all these [attacks] together. Irrespective of what happened, these are known attack chains." "Part of the issue was that [the hack] was flooded with so much information that I think it was just sort of buried — you know, 'What's going on? Is it real or is it not real?'" "And what is going to really take place is, the defenders have to use AI to be able to connect the dots on those — which we do — but then you also have to strip out all the noise. You have to separate the signal from the noise, and that's where these sorts of models perform well."

TBPN

43,276 次观看 • 12 天前

Sam Altman says he is NOT worried about Chinese labs distilling OpenAI: "I would rather people not distill from us, for sure. Maybe I’m feeling too confident right now about our progress and the models that are coming. But this is not in my top ten list of worries.” “I haven’t thought deeply about the distillation issue. It’s clearly a top of mind issue now for a lot of people all of a sudden." "I have always assumed that there are going to be great, cheap models in the world, and we better be the greatest and the cheapest. And other people are going to do what they’re going to do. But I think we can just really win at our own game here." "We will have so much usage of our models that we do not need to be a gigantic high margin business to be able to afford model training." "So much of our future compute plans will be used to sell inference to customers that even if we can enjoy a modest margin on trillions of dollars of revenue, we can afford to train some giant models."

Sam Altman says he is NOT worried about Chinese labs distilling OpenAI: "I would rather people not distill from us, for sure. Maybe I’m feeling too confident right now about our progress and the models that are coming. But this is not in my top ten list of worries.” “I haven’t thought deeply about the distillation issue. It’s clearly a top of mind issue now for a lot of people all of a sudden." "I have always assumed that there are going to be great, cheap models in the world, and we better be the greatest and the cheapest. And other people are going to do what they’re going to do. But I think we can just really win at our own game here." "We will have so much usage of our models that we do not need to be a gigantic high margin business to be able to afford model training." "So much of our future compute plans will be used to sell inference to customers that even if we can enjoy a modest margin on trillions of dollars of revenue, we can afford to train some giant models."

dnap

91,523 次观看 • 1 个月前

.Greg Brockman sees OpenAI as building "the world's best healthcare platform." "Think about how much work you have to do as a patient if you talk to different specialists." "You have to carry your medical record from one to the other, re-explaining your issue, etc. Being able to have good sharing of that information across different providers becomes possible if everyone's on one platform." "Take clinical trial enrollment. That's a huge bottleneck to drug development and finding people who are eligible and will benefit from being enrolled in a particular trial." "If you have that kind of data, if people are willing to entrust you with that information, that's something that can actually benefit them and the world."

.Greg Brockman sees OpenAI as building "the world's best healthcare platform." "Think about how much work you have to do as a patient if you talk to different specialists." "You have to carry your medical record from one to the other, re-explaining your issue, etc. Being able to have good sharing of that information across different providers becomes possible if everyone's on one platform." "Take clinical trial enrollment. That's a huge bottleneck to drug development and finding people who are eligible and will benefit from being enrolled in a particular trial." "If you have that kind of data, if people are willing to entrust you with that information, that's something that can actually benefit them and the world."

TBPN

47,919 次观看 • 5 天前

.Garry Tan says the Jacob Coxon stuff is a smokescreen distracting us from the much more immediate, practical concerns around AI that we're facing right now: "We should be talking less about this Jacob Coxon guy, and talking a lot more about — what is actually happening with Hugging Face? Are agent swarms going to take over infrastructure en masse? And then, what are we actually doing about that?" "I don't want to hear about some guy who worked for Anthropic for 2 months. There's a coordinated effort to try to influence politicians to get a knee-jerk response out of them." "That's a smokescreen. You shouldn't be paying attention to that. We need to be paying attention to the actual things we can do to, for example, prevent agents swarms from taking over entire data centers. What's our shutdown strategy? How do we ensure provenance? Where is this agent actually located? What software can we build? What cybersecurity defenses can we build today?" "That's the level of discourse I think we need, and we just don't have that." "I don't really care about science fiction. I saw Terminator 2, too. We're not here to talk about that. We need to actually talk about what's really happening with the servers, what's actually happening with the agent swarms, and how do we actually prevent that?" "When it comes to regulation, it's like, let's pass regulation of these things we actually care about, instead of what a socialist says in the New York Times. I don't care about that."

.Garry Tan says the Jacob Coxon stuff is a smokescreen distracting us from the much more immediate, practical concerns around AI that we're facing right now: "We should be talking less about this Jacob Coxon guy, and talking a lot more about — what is actually happening with Hugging Face? Are agent swarms going to take over infrastructure en masse? And then, what are we actually doing about that?" "I don't want to hear about some guy who worked for Anthropic for 2 months. There's a coordinated effort to try to influence politicians to get a knee-jerk response out of them." "That's a smokescreen. You shouldn't be paying attention to that. We need to be paying attention to the actual things we can do to, for example, prevent agents swarms from taking over entire data centers. What's our shutdown strategy? How do we ensure provenance? Where is this agent actually located? What software can we build? What cybersecurity defenses can we build today?" "That's the level of discourse I think we need, and we just don't have that." "I don't really care about science fiction. I saw Terminator 2, too. We're not here to talk about that. We need to actually talk about what's really happening with the servers, what's actually happening with the agent swarms, and how do we actually prevent that?" "When it comes to regulation, it's like, let's pass regulation of these things we actually care about, instead of what a socialist says in the New York Times. I don't care about that."

TBPN

294,584 次观看 • 3 天前

"If you want tons of embarrassingly parallel inference running all the same models, running future models that get bigger or smaller, we think we're going to be perfect for lots of that." Panthalassa's Garth Sheldon-Coulson, who just announced a $140M B round led by Peter Thiel to build wave-powered data centers in the ocean, explains the concept in detail: "The typical node will be on the order of 500 kW. One rack to multiple racks depending on density." "They can talk to each other sideways — we'll have radio — and they'll be talking to satellites." "So we're building this grid. And it's not for synchronous training. This is not fiber between everything." "If you want tons of embarrassingly parallel inference running all the same models, running future models that get bigger or smaller, we think we're going to be perfect for lots of that." "The way that we see this going is, once the energy starts to crash through what's available in the grid, it all just becomes about who can actually build in an elastic fashion to make 10 GW, 100 GW." "100 GW per year is the kind of thing we're getting asked to do, so that's the framework that we want to build. Just pure manufacturing to get energy capacity out there."

"If you want tons of embarrassingly parallel inference running all the same models, running future models that get bigger or smaller, we think we're going to be perfect for lots of that." Panthalassa's Garth Sheldon-Coulson, who just announced a $140M B round led by Peter Thiel to build wave-powered data centers in the ocean, explains the concept in detail: "The typical node will be on the order of 500 kW. One rack to multiple racks depending on density." "They can talk to each other sideways — we'll have radio — and they'll be talking to satellites." "So we're building this grid. And it's not for synchronous training. This is not fiber between everything." "If you want tons of embarrassingly parallel inference running all the same models, running future models that get bigger or smaller, we think we're going to be perfect for lots of that." "The way that we see this going is, once the energy starts to crash through what's available in the grid, it all just becomes about who can actually build in an elastic fashion to make 10 GW, 100 GW." "100 GW per year is the kind of thing we're getting asked to do, so that's the framework that we want to build. Just pure manufacturing to get energy capacity out there."

TBPN

35,307 次观看 • 4 个月前

When hiring executives, travis kalanick prioritizes one skill above all others: the ability to solve hard problems. "There's two things you want an executive to do. You want them to be able to organize, manage, and lead at scale. You also want them to be epic and strategic problem solvers... the best executives are the ones that are doing both well." "I have come to the conclusion over my years, that problem solving is the most important thing. If you get somebody who organizes and manages well but cannot solve a problem, they're going to be doing ridiculous stuff in a super organized way. The only constraint on your imagination is management capacity. But what is management capacity? It's really problem solving at scale."

When hiring executives, travis kalanick prioritizes one skill above all others: the ability to solve hard problems. "There's two things you want an executive to do. You want them to be able to organize, manage, and lead at scale. You also want them to be epic and strategic problem solvers... the best executives are the ones that are doing both well." "I have come to the conclusion over my years, that problem solving is the most important thing. If you get somebody who organizes and manages well but cannot solve a problem, they're going to be doing ridiculous stuff in a super organized way. The only constraint on your imagination is management capacity. But what is management capacity? It's really problem solving at scale."

TBPN

84,928 次观看 • 1 个月前

Sullivan: I think the truth will out here… and I think that has to be of real concern to the administration. You may have noticed that when President Trump got asked about this, he said, “I didn’t know anything about that,” which is kind of a telltale sign of him trying to get away from what is an extremely serious situation.

Sullivan: I think the truth will out here… and I think that has to be of real concern to the administration. You may have noticed that when President Trump got asked about this, he said, “I didn’t know anything about that,” which is kind of a telltale sign of him trying to get away from what is an extremely serious situation.

Acyn

61,609 次观看 • 9 个月前

"One of the very confusing things about the models right now: how to reconcile the fact that they are doing so well on evals. And you look at the evals and you go, 'Those are pretty hard evals.' But the economic impact seems to be dramatically behind. There is [a possible] explanation. Back when people were doing pre-training, the question of what data to train on was answered, because that answer was everything. So you don't have to think if it's going to be this data or that data. When people do RL training, they say, 'Okay, we want to have this kind of RL training for this thing and that kind of RL training for that thing.' You say, 'Hey, I would love our model to do really well when we release it. I want the evals to look great. What would be RL training that could help on this task?' If you combine this with generalization of the models actually being inadequate, that has the potential to explain a lot of what we are seeing, this disconnect between eval performance and actual real-world performance"

"One of the very confusing things about the models right now: how to reconcile the fact that they are doing so well on evals. And you look at the evals and you go, 'Those are pretty hard evals.' But the economic impact seems to be dramatically behind. There is [a possible] explanation. Back when people were doing pre-training, the question of what data to train on was answered, because that answer was everything. So you don't have to think if it's going to be this data or that data. When people do RL training, they say, 'Okay, we want to have this kind of RL training for this thing and that kind of RL training for that thing.' You say, 'Hey, I would love our model to do really well when we release it. I want the evals to look great. What would be RL training that could help on this task?' If you combine this with generalization of the models actually being inadequate, that has the potential to explain a lot of what we are seeing, this disconnect between eval performance and actual real-world performance"

Dwarkesh Patel

502,484 次观看 • 9 个月前

"The Achilles heel of this whole story... is if something bad happens to Anthropic and OpenAI... open weight models are much cheaper... if they start really taking a lot of market share, and it sounds like from what I'm hearing, that they're starting to. You could have a big price war, and then then we have a problem" 🎙️ Steve Eisman on Fast Money last night

"The Achilles heel of this whole story... is if something bad happens to Anthropic and OpenAI... open weight models are much cheaper... if they start really taking a lot of market share, and it sounds like from what I'm hearing, that they're starting to. You could have a big price war, and then then we have a problem" 🎙️ Steve Eisman on Fast Money last night

RiskReversal Media

26,477 次观看 • 1 个月前

OpenAI president Greg Brockman told his agent to ping a colleague on Slack, and two minutes later it escalated to that colleague's manager on its own: "Two minutes later, it said, this is taking too long. I've escalated to the person's manager." "On the one hand, it's kind of a reasonable thing for the model to do. It's being proactive, it's trying to solve my problem." "But on the other hand, maybe it should have taken a little bit longer, maybe it should have checked with me." "Now we're starting to have AIs that can actually take care of flagging. Is this a high risk action? Hey, this one should be escalated. This one's OK to auto-approve." "It really makes you realize that human attention is going to be this incredibly scarce resource. The doing of things now is easy." "Is this what I wanted? Is this aligned with my values, with my desires? That is going to become the single most important bottleneck." The model did what he asked, then added a step nobody asked for, and it landed on a colleague's boss. Agents that can take actions with social consequences need an approve-and-escalate list written before deployment. Brockman says OpenAI is building that flagging into the model itself. Every company deploying agents needs its own version, reviewed by someone accountable. - Greg Brockman (Greg Brockman), President and co-founder of OpenAI, at Sequoia Capital's (Sequoia Capital) AI Ascent.

OpenAI president Greg Brockman told his agent to ping a colleague on Slack, and two minutes later it escalated to that colleague's manager on its own: "Two minutes later, it said, this is taking too long. I've escalated to the person's manager." "On the one hand, it's kind of a reasonable thing for the model to do. It's being proactive, it's trying to solve my problem." "But on the other hand, maybe it should have taken a little bit longer, maybe it should have checked with me." "Now we're starting to have AIs that can actually take care of flagging. Is this a high risk action? Hey, this one should be escalated. This one's OK to auto-approve." "It really makes you realize that human attention is going to be this incredibly scarce resource. The doing of things now is easy." "Is this what I wanted? Is this aligned with my values, with my desires? That is going to become the single most important bottleneck." The model did what he asked, then added a step nobody asked for, and it landed on a colleague's boss. Agents that can take actions with social consequences need an approve-and-escalate list written before deployment. Brockman says OpenAI is building that flagging into the model itself. Every company deploying agents needs its own version, reviewed by someone accountable. - Greg Brockman (Greg Brockman), President and co-founder of OpenAI, at Sequoia Capital's (Sequoia Capital) AI Ascent.

Karl Mehta

75,634 次观看 • 1 个月前

Kasie: More kids died in this country of measles this year than in decades Mullin: We can talk about the open southern border and what caused that. We're not talking about measles outbreaks from kids that were born and raised here. We're talking about ones that were brought over here illegally. And you can make that connection. No one can argue that.

Kasie: More kids died in this country of measles this year than in decades Mullin: We can talk about the open southern border and what caused that. We're not talking about measles outbreaks from kids that were born and raised here. We're talking about ones that were brought over here illegally. And you can make that connection. No one can argue that.

Acyn

402,247 次观看 • 1 年前

Prime Intellect's will brown says continual learning could be solved in the first half of 2026: "Continual learning is going to fall pretty quickly, I think. It's more of an engineering problem. No one's actually trying." "OpenAI and Anthropic don't want to continuously train their models for each user. It's expensive and annoying and hard to serve at scale. But from a research perspective, we do continue learning where they just keep training the model more and it knows more stuff because they put more of the Internet in it." "I think there's a lot of experimentation around exactly the recipe that's going to be the most reliable. But we kind of have a grab bag of six or seven tricks that kind of work, or they work in different ways, and you can mix and match them. And it's just going to be like whatever's the best combination of these tricks." "People are going to experiment with it and find the versions that work the best. And there doesn't seem to be any big wall inside that prevents that from being practical."

Prime Intellect's will brown says continual learning could be solved in the first half of 2026: "Continual learning is going to fall pretty quickly, I think. It's more of an engineering problem. No one's actually trying." "OpenAI and Anthropic don't want to continuously train their models for each user. It's expensive and annoying and hard to serve at scale. But from a research perspective, we do continue learning where they just keep training the model more and it knows more stuff because they put more of the Internet in it." "I think there's a lot of experimentation around exactly the recipe that's going to be the most reliable. But we kind of have a grab bag of six or seven tricks that kind of work, or they work in different ways, and you can mix and match them. And it's just going to be like whatever's the best combination of these tricks." "People are going to experiment with it and find the versions that work the best. And there doesn't seem to be any big wall inside that prevents that from being practical."

TBPN

72,357 次观看 • 6 个月前

Bret Weintein asks "what is this all really for?" "If you look at the push towards tokenization of assets, towards central bank digital currencies—that is programmable money that can be turned off if you displease the powers that be. If you look at the creation of these massive data centers, is that really about citizen use of AI? Maybe some of it is, but is it really about creating a storehouse of information that will allow you to be controlled and locked down? That’s certainly the impression I have. We didn’t have a discussion about any of this. This is happening, and then we have to react to it because it’s all on a freight train headed down the line… and we’re always playing defense. So it’s pretty frightening stuff. And at some level, we have a moral right to object. We’re not being given the opportunity to object..."

Bret Weintein asks "what is this all really for?" "If you look at the push towards tokenization of assets, towards central bank digital currencies—that is programmable money that can be turned off if you displease the powers that be. If you look at the creation of these massive data centers, is that really about citizen use of AI? Maybe some of it is, but is it really about creating a storehouse of information that will allow you to be controlled and locked down? That’s certainly the impression I have. We didn’t have a discussion about any of this. This is happening, and then we have to react to it because it’s all on a freight train headed down the line… and we’re always playing defense. So it’s pretty frightening stuff. And at some level, we have a moral right to object. We’re not being given the opportunity to object..."

The DarkHorse Podcast

324,177 次观看 • 1 个月前

Dr. Phil Goes Off Script With Stunning Criticism of the US Government “We got people in charge of solving problems that don’t really want to solve them because their job depends on having that problem. “They need that problem to justify their existence. They need that problem to maintain this huge bureaucracy, to maintain this big budget, this big agency. They need that problem to exist. I think we’ve got people in charge of solving problems that are not problem solvers.”

Dr. Phil Goes Off Script With Stunning Criticism of the US Government “We got people in charge of solving problems that don’t really want to solve them because their job depends on having that problem. “They need that problem to justify their existence. They need that problem to maintain this huge bureaucracy, to maintain this big budget, this big agency. They need that problem to exist. I think we’ve got people in charge of solving problems that are not problem solvers.”

The Vigilant Fox 🦊

2,147,318 次观看 • 2 年前