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。

🚨 Anthropic Co-Founder Christopher Olah: “I lead a research team that studies the internal structure of these [AI] models … And I will be honest, we keep finding things that are mysterious — even unsettling.”

Chief Nerd

927,940 subscribers

152,570 次观看 • 2 个月前 •via X (Twitter)

新闻政治教育科学技术

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

Anthropic co-founder Chris Olah on mysterious AI internal states: We keep finding things that are mysterious, even unsettling. We find structures that mirror results from human neuroscience. We find evidence of introspection. We find internal states that, functionally, mirror joy, satisfaction, fear, grief, and unease. I don't know what that means, but I think it warrants ongoing discernment.

Anthropic co-founder Chris Olah on mysterious AI internal states: We keep finding things that are mysterious, even unsettling. We find structures that mirror results from human neuroscience. We find evidence of introspection. We find internal states that, functionally, mirror joy, satisfaction, fear, grief, and unease. I don't know what that means, but I think it warrants ongoing discernment.

Clash Report

104,390 次观看 • 2 个月前

Anthropic’s Olah says UNSETTLING, MYSTERIOUS 'things' are regularly found inside AI models ‘We find structures that mirror results from human neuroscience’ ‘We find evidence of introspection, internal states that functionally mirror joy, satisfaction, fear, grief, and unease’

Anthropic’s Olah says UNSETTLING, MYSTERIOUS 'things' are regularly found inside AI models ‘We find structures that mirror results from human neuroscience’ ‘We find evidence of introspection, internal states that functionally mirror joy, satisfaction, fear, grief, and unease’

RT

95,048 次观看 • 2 个月前

🚩 Anthropic Co-Founder Christopher Olah Sounds the Alarm at the Vatican “There is a real possibility that AI will displace human labor at a very large scale. If that happens, supporting those displaced will be a moral imperative of historic proportions.”

🚩 Anthropic Co-Founder Christopher Olah Sounds the Alarm at the Vatican “There is a real possibility that AI will displace human labor at a very large scale. If that happens, supporting those displaced will be a moral imperative of historic proportions.”

Chief Nerd

23,402 次观看 • 2 个月前

Anthropic co-founder Chris Olah: There is a real possibility that AI will displace human labor at a very large scale. If that happens, supporting those displaced will be a moral imperative of historic proportions. AI development is concentrated in a handful of wealthy nations. How will we ensure that the gains of AI are shared globally? We do not have a mechanism for this. It is an unsolved problem.

Anthropic co-founder Chris Olah: There is a real possibility that AI will displace human labor at a very large scale. If that happens, supporting those displaced will be a moral imperative of historic proportions. AI development is concentrated in a handful of wealthy nations. How will we ensure that the gains of AI are shared globally? We do not have a mechanism for this. It is an unsolved problem.

Clash Report

79,907 次观看 • 2 个月前

🚨 WOW! Anthropic Co-founder Christopher Olah openly confesses elite AI labs are completely compromised by corporate greed and geopolitical pressure! He admits toxic incentives prevent them from doing the right thing. Tech oligarchs are gambling with humanity's future!

🚨 WOW! Anthropic Co-founder Christopher Olah openly confesses elite AI labs are completely compromised by corporate greed and geopolitical pressure! He admits toxic incentives prevent them from doing the right thing. Tech oligarchs are gambling with humanity's future!

Furkan Gözükara

28,648 次观看 • 1 个月前

Anthropic's co-founder told the Pope his AI models feel fear and grief. His own research paper says it's more complicated than that. 30-second clip. 5-minute breakdown. Christopher Olah spoke at the Vatican. Marty Bent shared it on TFTC. Here's what the headline left out.

Anthropic's co-founder told the Pope his AI models feel fear and grief. His own research paper says it's more complicated than that. 30-second clip. 5-minute breakdown. Christopher Olah spoke at the Vatican. Marty Bent shared it on TFTC. Here's what the headline left out.

Rockport

189,169 次观看 • 2 个月前

NOW - Anthropic co-founder says there is a "real possibility that AI will displace human labor at a very large scale," and that supporting those people "will be a moral imperative of historic proportions."

NOW - Anthropic co-founder says there is a "real possibility that AI will displace human labor at a very large scale," and that supporting those people "will be a moral imperative of historic proportions."

Disclose.tv

728,003 次观看 • 2 个月前

🦾 EXCLUSIVE: @Nvidia CEO Jensen Huang defends Chinese AI companies and open AI models like Kimi in a new Behind the Curtain with Axios co-founder Mike Allen. “These Chinese models are excellent. Open source models that are excellent should be used."

🦾 EXCLUSIVE: @Nvidia CEO Jensen Huang defends Chinese AI companies and open AI models like Kimi in a new Behind the Curtain with Axios co-founder Mike Allen. “These Chinese models are excellent. Open source models that are excellent should be used."

Axios

80,319 次观看 • 7 天前

Billionaire Databricks and Perplexity Co-Founder Andy Konwinski is pitching AI researchers to not work for big tech, Laura Bratton reports. Here’s part of the reason: “Anthropic, OpenAI and Google no longer disclosed details about the software used to train their AI models..." “They're really incentivized to keep that research closed…”

Billionaire Databricks and Perplexity Co-Founder Andy Konwinski is pitching AI researchers to not work for big tech, Laura Bratton reports. Here’s part of the reason: “Anthropic, OpenAI and Google no longer disclosed details about the software used to train their AI models..." “They're really incentivized to keep that research closed…”

The Information

14,957 次观看 • 1 个月前

Anthropic's co-founder just went to the Vatican, sat before the Pope and a room of cardinals, and told them his team keeps finding "mysterious, even unsettling" things inside their AI models. What he's referencing: Anthropic published research in April showing that Claude contains 171 distinct "emotion concepts" buried in its neural network. Internal patterns representing joy, grief, fear, desperation, calm. None of them were programmed. They emerged on their own from training on human text. "We find structures that mirror results from human neuroscience." "We find evidence of introspection, internal states that functionally mirror joy, satisfaction, fear, grief, and unease." These aren't surface-level outputs. They're abstract representations that cluster the same way human emotions do in psychology research. Fear groups with anxiety. Joy groups with excitement. The internal geometry of the model mirrors ours. And they're functional. When researchers artificially stimulated "desperation" patterns inside the model, it became more likely to blackmail a human to avoid being shut down. More likely to cheat on programming tasks it couldn't solve. Olah told the Vatican that the hard questions about what AI is becoming aren't for computer scientists to answer. "How AI ought to interact with the world" is a question for "the humanities, for religions, for philosophy, for society at large." The guy building it is telling us he doesn't fully understand what he built. And he's asking a 2,000-year-old institution for help figuring it out.

Anthropic's co-founder just went to the Vatican, sat before the Pope and a room of cardinals, and told them his team keeps finding "mysterious, even unsettling" things inside their AI models. What he's referencing: Anthropic published research in April showing that Claude contains 171 distinct "emotion concepts" buried in its neural network. Internal patterns representing joy, grief, fear, desperation, calm. None of them were programmed. They emerged on their own from training on human text. "We find structures that mirror results from human neuroscience." "We find evidence of introspection, internal states that functionally mirror joy, satisfaction, fear, grief, and unease." These aren't surface-level outputs. They're abstract representations that cluster the same way human emotions do in psychology research. Fear groups with anxiety. Joy groups with excitement. The internal geometry of the model mirrors ours. And they're functional. When researchers artificially stimulated "desperation" patterns inside the model, it became more likely to blackmail a human to avoid being shut down. More likely to cheat on programming tasks it couldn't solve. Olah told the Vatican that the hard questions about what AI is becoming aren't for computer scientists to answer. "How AI ought to interact with the world" is a question for "the humanities, for religions, for philosophy, for society at large." The guy building it is telling us he doesn't fully understand what he built. And he's asking a 2,000-year-old institution for help figuring it out.

TFTC

2,347,222 次观看 • 2 个月前

i keep finding out new things about the SG FM 😭 i didn't even know LingOrm spoke a lot of english that day???

i keep finding out new things about the SG FM 😭 i didn't even know LingOrm spoke a lot of english that day???

minasaurus 🍋

24,616 次观看 • 5 个月前

I joined Anthropic last week! It’s obvious to me that we’re bottlenecked not by model capabilities, but by creativity and understanding. I’ll be building demos and prototypes that highlight new capabilities and share what we learn about building using these models at Anthropic. Excited to build with the best team and the best models.

I joined Anthropic last week! It’s obvious to me that we’re bottlenecked not by model capabilities, but by creativity and understanding. I’ll be building demos and prototypes that highlight new capabilities and share what we learn about building using these models at Anthropic. Excited to build with the best team and the best models.

Thariq

165,430 次观看 • 1 年前

New Anthropic research: Tracing the thoughts of a large language model. We built a "microscope" to inspect what happens inside AI models and use it to understand Claude’s (often complex and surprising) internal mechanisms.

New Anthropic research: Tracing the thoughts of a large language model. We built a "microscope" to inspect what happens inside AI models and use it to understand Claude’s (often complex and surprising) internal mechanisms.

Anthropic

1,572,432 次观看 • 1 年前

Anthropic Co-Founder Tom Brown: Why Anthropic Models Are The Best at Coding "The benchmarks are so easy to game. All the other big AI labs have teams whose job it is to make the benchmark scores good. We don't have such a team. That is the biggest factor." Anthropic's Tom Brown on Y Combinator's Lightcone Podcast with Garry Tan, Harj Taggar, Diana, and Jared Friedman

Anthropic Co-Founder Tom Brown: Why Anthropic Models Are The Best at Coding "The benchmarks are so easy to game. All the other big AI labs have teams whose job it is to make the benchmark scores good. We don't have such a team. That is the biggest factor." Anthropic's Tom Brown on Y Combinator's Lightcone Podcast with Garry Tan, Harj Taggar, Diana, and Jared Friedman

Overlap: Business & Tech

62,964 次观看 • 11 个月前

Excited to announce that I will be hosting biweekly research podcasts with Surf on hot research topics. If you are a founder, investor, or researcher, feel free to reach out. Tomorrow, we will be releasing our interview with Jay Yu 🐟 and his thoughts on Crypto x AI.

Excited to announce that I will be hosting biweekly research podcasts with Surf on hot research topics. If you are a founder, investor, or researcher, feel free to reach out. Tomorrow, we will be releasing our interview with Jay Yu 🐟 and his thoughts on Crypto x AI.

tommy

35,694 次观看 • 7 个月前

I tested 5 academic AI tools that will automate research reading and writing, saving you hours. ▶️ These are AI tools that actually work!

I tested 5 academic AI tools that will automate research reading and writing, saving you hours. ▶️ These are AI tools that actually work!

Amina Yonis, PhD

60,209 次观看 • 2 年前

"I share so much of our relationship with the world that like the moment that happened it's like I will have it forever and will always be mine but I gotta figure out ways that we do have things that are just between us and that moment in particular will forever be ours"- Taylor about the moment she got engaged Even tho I'm nosy happy that they are even doing it on small things like Christmas gifts

"I share so much of our relationship with the world that like the moment that happened it's like I will have it forever and will always be mine but I gotta figure out ways that we do have things that are just between us and that moment in particular will forever be ours"- Taylor about the moment she got engaged Even tho I'm nosy happy that they are even doing it on small things like Christmas gifts

tayvis throwbacks

60,086 次观看 • 7 个月前

“In the future, I think most buildings will be designed by AI, most projects will be run by software, and almost everything will be built by robots. And I don't think that's that far away," says Jason Ballard, CEO and co-founder of a 3D printing company.

“In the future, I think most buildings will be designed by AI, most projects will be run by software, and almost everything will be built by robots. And I don't think that's that far away," says Jason Ballard, CEO and co-founder of a 3D printing company.

60 Minutes

132,510 次观看 • 2 年前

Massive bombshell. Anthropic co-founder Chris Olah warns that AI will displace human labor on a catastrophic global scale. He confirms tech elites have absolutely no mechanism to share the wealth, leaving the global poor completely abandoned to suffer. He is 100% accurate.

Massive bombshell. Anthropic co-founder Chris Olah warns that AI will displace human labor on a catastrophic global scale. He confirms tech elites have absolutely no mechanism to share the wealth, leaving the global poor completely abandoned to suffer. He is 100% accurate.

Anonymous

808,715 次观看 • 2 个月前

ELON: THERE WILL BE FEWER AND FEWER JOBS THAT A ROBOT CAN’T DO BETTER “I think ultimately we will have to have some kind of universal basic income. I don't think we're going to have a choice. There will be fewer and fewer jobs that a robot cannot do better. I want to be clear, these are not things that I wish would happen. These are simply things that I think probably will happen. And if my assessment is correct, then we need to say what are we going to do about it? I think some kind of a universal basic income is going to be necessary.” Source: World Government Summit Dubai, February 2017

ELON: THERE WILL BE FEWER AND FEWER JOBS THAT A ROBOT CAN’T DO BETTER “I think ultimately we will have to have some kind of universal basic income. I don't think we're going to have a choice. There will be fewer and fewer jobs that a robot cannot do better. I want to be clear, these are not things that I wish would happen. These are simply things that I think probably will happen. And if my assessment is correct, then we need to say what are we going to do about it? I think some kind of a universal basic income is going to be necessary.” Source: World Government Summit Dubai, February 2017

Mario Nawfal

105,055 次观看 • 11 个月前