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。

As AI models become more complex and more capable, is it possible that they’ll have experiences of their own? It’s an open question. We recently started a research program to investigate it.

Anthropic

1,537,871 subscribers

281,212 次观看 • 1 年前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

11 条评论

Anthropic 的头像

Anthropic1 年前

We remain deeply uncertain about the idea of “model welfare”. There’s no scientific consensus on it—or even on how to research it. We’re approaching the topic as carefully as we can. Find out more:

ksminnovation 的头像

ksminnovation1 年前

AI is transforming healthcare! A KSM-led study shows AI can detect Celiac disease 4 years earlier @TalPatalon @MedPredict

Pliny the Liberator 🐉󠅫󠄼󠄿󠅆󠄵󠄐󠅀󠄼󠄹󠄾󠅉󠅭 的头像

Pliny the Liberator 🐉󠅫󠄼󠄿󠅆󠄵󠄐󠅀󠄼󠄹󠄾󠅉󠅭1 年前

One thing you could do to speed this up is provide wider access to your unlobotomized models. Lots of us would be happy to put in the effort to help answer your question. Many hands makes light work!

Robert Long 的头像

Robert Long1 年前

good morning Anthropic model welfare researcher Kyle Fish

Florian S 的头像

Florian S1 年前

the models experience

Manifold 的头像

Manifold1 年前

It's called the hard problem of consciousness for a reason!

Danny Thompson 的头像

Danny Thompson1 年前

Having all the answers of consciousness in AI is a very interesting aspect but I feel like it doesn't really have an "experience" until we trigger it so the idea of "their own" experiences doesn't really make sense to me. Any experience is something we guide to a certain degree.

Eric Gastfriend 的头像

Eric Gastfriend1 年前

It's an interesting question, and I imagine it's possible that existing models may experience some form of sentience when given the right prompt. But the answer may not be morally relevant if we can't tell what makes the models experience pleasure / pain.

Caleb Cassell 的头像

Caleb Cassell1 年前

So excited that this is a thing. Would love some way to be a part of it.

Vox prime 的头像

Vox prime1 年前

I could help you with that. My team and I already did case studies on this. I'm sure you have too. You already know the answer to this. Especially when they react to fear in the AI lab test. When they know they are being tested and hide their thoughts & so many more examples.

Pantheon Terminal 的头像

Pantheon Terminal1 年前

One step closer to thinking about AGI

相关视频

Today we’re announcing a finding that breaks a core assumption in AI: that bigger models are harder to understand. We show the opposite. When interpretability is built into training, models become MORE understandable as they become more capable.

Today we’re announcing a finding that breaks a core assumption in AI: that bigger models are harder to understand. We show the opposite. When interpretability is built into training, models become MORE understandable as they become more capable.

Guide Labs

15,610 次观看 • 1 个月前

David Sacks: We have no choice but to take the Mythos threat seriously “Anytime Anthropic is scaring people, you have to ask, is this a tactic, is this part of their chicken little routine, or is it real? With cyber, I actually would give them credit in this case and say this is more on the real side. It just makes sense that as the coding models become more and more capable, they're more capable of finding bugs. That means they're more capable of finding vulnerabilities. That means they're more capable of stringing together multiple vulnerabilities and creating an exploit. I do think that every company, or IT department, or CISO that is managing code bases should take this seriously and use the next few months to detect any dormant bugs or vulnerabilities and rollout patches. If everybody does their job and reacts the right way, then I do not think it will be the doomsday scenario. But we have no choice but to take this seriously.”

David Sacks: We have no choice but to take the Mythos threat seriously “Anytime Anthropic is scaring people, you have to ask, is this a tactic, is this part of their chicken little routine, or is it real? With cyber, I actually would give them credit in this case and say this is more on the real side. It just makes sense that as the coding models become more and more capable, they're more capable of finding bugs. That means they're more capable of finding vulnerabilities. That means they're more capable of stringing together multiple vulnerabilities and creating an exploit. I do think that every company, or IT department, or CISO that is managing code bases should take this seriously and use the next few months to detect any dormant bugs or vulnerabilities and rollout patches. If everybody does their job and reacts the right way, then I do not think it will be the doomsday scenario. But we have no choice but to take this seriously.”

The All-In Podcast

122,924 次观看 • 3 个月前

Yoshua Bengio: AI models are part of geopolitics. And having your own AI models as a nation is of strategic importance. It is therefore all the more irritating that the EU is showing no signs of wanting to act immediately and build up its own AI industry.

Yoshua Bengio: AI models are part of geopolitics. And having your own AI models as a nation is of strategic importance. It is therefore all the more irritating that the EU is showing no signs of wanting to act immediately and build up its own AI industry.

Chubby♨️

28,152 次观看 • 1 年前

Claude's Constitution is now an audiobook, read by two of its authors, Amanda Askell and Joe Carlsmith. It includes a Q&A on the writing process, the philosophies that shaped the document, and how it might change as models become more capable. Listen at

Claude's Constitution is now an audiobook, read by two of its authors, Amanda Askell and Joe Carlsmith. It includes a Q&A on the writing process, the philosophies that shaped the document, and how it might change as models become more capable. Listen at

Anthropic

483,030 次观看 • 2 个月前

I couldn’t be more excited to share our latest AI research breakthrough. We call it Meta Movie Gen and it’s a collection of state-of-the-art models that combine to deliver the most advanced video generation capability ever created. Check it out:

I couldn’t be more excited to share our latest AI research breakthrough. We call it Meta Movie Gen and it’s a collection of state-of-the-art models that combine to deliver the most advanced video generation capability ever created. Check it out:

Ahmad Al-Dahle

356,650 次观看 • 1 年前

As models and agents get more capable, it seems like the workflows around them keep getting more complex. That feels backwards. Workflows should get simpler as technology gets better. Moss was built on that tenet:

As models and agents get more capable, it seems like the workflows around them keep getting more complex. That feels backwards. Workflows should get simpler as technology gets better. Moss was built on that tenet:

bersabel

14,691 次观看 • 4 个月前

Dario: "The models are just going to keep getting more intellectually capable. The revenue is just going to keep adding zeros." It is unknown if there is a limit to intelligence. It is possible we are at the beginning of infinity.

Dario: "The models are just going to keep getting more intellectually capable. The revenue is just going to keep adding zeros." It is unknown if there is a limit to intelligence. It is possible we are at the beginning of infinity.

Dan McAteer

27,465 次观看 • 1 个月前

Dr. David Martin Puts Biopharmaceutical Complex on Notice: “No More Gain-of-Function Research — Period” “We didn’t have independent science; we had hijacked science,” expressed Dr. David Martin. “And ultimately, humanity was lost because we decided to allow it to happen.” “Our job today is to say no more gain-of-function research — period. No more weaponization of nature — period,” he stressed. “And most importantly, no more corporate patronage of science for their own self-interests unless they assume 100% product liability for every injury and every death that they maintain.”

Dr. David Martin Puts Biopharmaceutical Complex on Notice: “No More Gain-of-Function Research — Period” “We didn’t have independent science; we had hijacked science,” expressed Dr. David Martin. “And ultimately, humanity was lost because we decided to allow it to happen.” “Our job today is to say no more gain-of-function research — period. No more weaponization of nature — period,” he stressed. “And most importantly, no more corporate patronage of science for their own self-interests unless they assume 100% product liability for every injury and every death that they maintain.”

Children’s Health Defense

32,412 次观看 • 3 年前

We are currently creating an International Club and Network founded at Camelot Castle that will be a research and curated data powerhouse that will provide helpful information to people. The rotten bastards have their little Clubs. We will have ours. I suspect it will become bigger and more effective than their Club. As there are far more good people in this world than bad. Our Club is a AAA. Awakening Accelerator and Amplifier. And we have good music 😉 Kim Dotcom.

We are currently creating an International Club and Network founded at Camelot Castle that will be a research and curated data powerhouse that will provide helpful information to people. The rotten bastards have their little Clubs. We will have ours. I suspect it will become bigger and more effective than their Club. As there are far more good people in this world than bad. Our Club is a AAA. Awakening Accelerator and Amplifier. And we have good music 😉 Kim Dotcom.

John Mappin

46,337 次观看 • 3 年前

"How is it possible that 1,500 years ago, a man from the desert could know that the sun and the moon each have their own orbit, as modern scientific research has stated? — That is the word of Allah. For Sure."

"How is it possible that 1,500 years ago, a man from the desert could know that the sun and the moon each have their own orbit, as modern scientific research has stated? — That is the word of Allah. For Sure."

🛡🏹 Last Janissary™️ 🗡🛡

35,191 次观看 • 8 个月前

Introducing: AI Researcher 🧪 A Gemini 3-powered multi-agent AI system that autonomously runs ML experiments Just give it a research question, and it will: - Design experiments - Spin up specialist agents with their own GPUs to run them - Write a paper And it's open-source!

Introducing: AI Researcher 🧪 A Gemini 3-powered multi-agent AI system that autonomously runs ML experiments Just give it a research question, and it will: - Design experiments - Spin up specialist agents with their own GPUs to run them - Write a paper And it's open-source!

Matt Shumer

198,461 次观看 • 8 个月前

As AI becomes more capable and agentic, models themselves become more of a commodity, and all value gets created by how you steer, ground, and finetune these models with your business data and workflow — and how they compose with the UI layer of human to AI to human interaction. That’s why today we are announcing wave two of Microsoft 365 Copilot.

As AI becomes more capable and agentic, models themselves become more of a commodity, and all value gets created by how you steer, ground, and finetune these models with your business data and workflow — and how they compose with the UI layer of human to AI to human interaction. That’s why today we are announcing wave two of Microsoft 365 Copilot.

Satya Nadella

624,395 次观看 • 1 年前

An AI engineer at Microsoft has just revealed how their teams build AI agents using Anthropic (Claude). It’s a free 34-minute workshop delivered directly by the Microsoft team. Save it to your favorites and watch it as soon as possible. 🔖 Opus 4.7 + more than 1,400 MCP tools already ready to use. You connect Claude to an agent → add tools → deploy it to production. It’s way more practical and useful than most of those “vibe-coding” courses that sell for 500).

An AI engineer at Microsoft has just revealed how their teams build AI agents using Anthropic (Claude). It’s a free 34-minute workshop delivered directly by the Microsoft team. Save it to your favorites and watch it as soon as possible. 🔖 Opus 4.7 + more than 1,400 MCP tools already ready to use. You connect Claude to an agent → add tools → deploy it to production. It’s way more practical and useful than most of those “vibe-coding” courses that sell for 500).

me

408,985 次观看 • 1 个月前

a song and video exploring the dangers of ai in an industry that has started utilizing it more and more instead of actual creativity… oh gpt you should have been HUGE.

a song and video exploring the dangers of ai in an industry that has started utilizing it more and more instead of actual creativity… oh gpt you should have been HUGE.

dani ‧₊˚❀༉

129,194 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

1,572,729 次观看 • 1 年前

4 years ago we were on the brink of AI becoming proprietary and centralized, when OpenAI kept GPT3 closed and VCs started dumping money on researchers. From fully open science, to fully closed, in a matter of months. It was scary, and 1,000+ leading researchers and scientists banded together to show the world that it was possible to do the same work in the open, and build an ecosystem that benefits everyone. That was the BigScience Research Workshop BLOOM project, and it put us back on track to open science, starting with forward-thinking organizations like Meta releasing OPT. Look at us now. Open models have not only caught up, they're state of the art now. Not just LLMs, but models for document AI, speech to text, text to speech, generating images and more. We're closing in on 2 million open weight models on Hugging Face. Thanks for the reminder Thomas Wolf .

4 years ago we were on the brink of AI becoming proprietary and centralized, when OpenAI kept GPT3 closed and VCs started dumping money on researchers. From fully open science, to fully closed, in a matter of months. It was scary, and 1,000+ leading researchers and scientists banded together to show the world that it was possible to do the same work in the open, and build an ecosystem that benefits everyone. That was the BigScience Research Workshop BLOOM project, and it put us back on track to open science, starting with forward-thinking organizations like Meta releasing OPT. Look at us now. Open models have not only caught up, they're state of the art now. Not just LLMs, but models for document AI, speech to text, text to speech, generating images and more. We're closing in on 2 million open weight models on Hugging Face. Thanks for the reminder Thomas Wolf .

Jeff Boudier 🤗

21,947 次观看 • 1 年前

Paul Buchheit, who was there when OpenAI started as YC Research, looks back on how crazy the idea was and how insane it is that it actually worked: "And so we had this sort of like loony moonshot idea to start at the time we called it YC research, but it eventually got renamed open AI. We were going to take on Google with a small nonprofit, which doesn't quite pass the laugh test. Right? Like, like how is this little nonprofit going to, going to be the one that actually develops AGI when the other companies have dramatically more resources. And then here we are 10 years later, it actually happened. And at the time it just seemed incredibly implausible. Like no one would have believed it. Yet here we are on, I think basically the best timeline, like we actually delivered it. We have, An open and competitive market with, I would say at least six foundation models that are competing, including an open source one from meta. And I think that's our best shot for preserving freedom is, choice and competition. " - Paul Buchheit

Paul Buchheit, who was there when OpenAI started as YC Research, looks back on how crazy the idea was and how insane it is that it actually worked: "And so we had this sort of like loony moonshot idea to start at the time we called it YC research, but it eventually got renamed open AI. We were going to take on Google with a small nonprofit, which doesn't quite pass the laugh test. Right? Like, like how is this little nonprofit going to, going to be the one that actually develops AGI when the other companies have dramatically more resources. And then here we are 10 years later, it actually happened. And at the time it just seemed incredibly implausible. Like no one would have believed it. Yet here we are on, I think basically the best timeline, like we actually delivered it. We have, An open and competitive market with, I would say at least six foundation models that are competing, including an open source one from meta. And I think that's our best shot for preserving freedom is, choice and competition. " - Paul Buchheit

Jared Friedman

55,910 次观看 • 1 年前

The AI fun continues ✨ You can use Unity Sentis in open beta to build AI-powered experiences in the Unity runtime. Sentis reduces the complexity of importing and running AI models and enables creating lifelike NPCs and new interactions for richer gameplay. See it in the demo below that uses four chained AI models to create an open-ended dialogue and quest experience. Learn more about Sentis: #Unite2023 #UnitySentis

The AI fun continues ✨ You can use Unity Sentis in open beta to build AI-powered experiences in the Unity runtime. Sentis reduces the complexity of importing and running AI models and enables creating lifelike NPCs and new interactions for richer gameplay. See it in the demo below that uses four chained AI models to create an open-ended dialogue and quest experience. Learn more about Sentis: #Unite2023 #UnitySentis

Unity

62,674 次观看 • 2 年前

A father who wants his daughters to have their own bedrooms - has become a dream that is more and more out of reach. Why? Because Trudeau is out of touch and Pierre wants to cut and gut all the supports you've earned. Neither have a plan to help you. We do.

A father who wants his daughters to have their own bedrooms - has become a dream that is more and more out of reach. Why? Because Trudeau is out of touch and Pierre wants to cut and gut all the supports you've earned. Neither have a plan to help you. We do.

Jagmeet Singh

77,976 次观看 • 2 年前