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。

Interpretability research has made only minor contributions to AI safety so far. What can we do to change that? (Clip from a longer talk; YouTube link in the thread):

Christopher Potts

16,302 subscribers

29,228 次观看 • 10 个月前 •via X (Twitter)

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

Severance as a show about interpretability research in AI (a clip from a talk; YouTube link just below):

Severance as a show about interpretability research in AI (a clip from a talk; YouTube link just below):

Christopher Potts

44,898 次观看 • 10 个月前

The Anthropic perspective on interpretability is prominent and significant, but not inevitable. My own take is quite different. (Clip from a talk I gave; YouTube link in the thread):

The Anthropic perspective on interpretability is prominent and significant, but not inevitable. My own take is quite different. (Clip from a talk I gave; YouTube link in the thread):

Christopher Potts

56,581 次观看 • 10 个月前

How do we know what machines know? How can we understand the new, emerging behavior of machines? In this conversation between Been Kim from Google DeepMind and the BuzzRobot community, we dug into the challenges of interpretability and explored how humans and machines can better understand each other. Watch the full discussion on our YouTube channel. Thank you, Been, for taking the time to talk to us! Timestamps: 0:00 Intro 0:18 AI teaches grandmasters chess 05:30 AI neologisms 10:17 Extracting knowledge from machines 13:43 Interpretability research 17:19 Are we keeping up with AI progress 18:34 New AI related terms 20:26 The right direction for interpretability 23:35 AI lying 27:49 Conseptual maps 30:30 AI researchers bias 33:33 Generalizing AI teaching humans 35:11 Is AI sentient 36:04 Does AI has concepts 41:53 Progress in machine understanding

How do we know what machines know? How can we understand the new, emerging behavior of machines? In this conversation between Been Kim from Google DeepMind and the BuzzRobot community, we dug into the challenges of interpretability and explored how humans and machines can better understand each other. Watch the full discussion on our YouTube channel. Thank you, Been, for taking the time to talk to us! Timestamps: 0:00 Intro 0:18 AI teaches grandmasters chess 05:30 AI neologisms 10:17 Extracting knowledge from machines 13:43 Interpretability research 17:19 Are we keeping up with AI progress 18:34 New AI related terms 20:26 The right direction for interpretability 23:35 AI lying 27:49 Conseptual maps 30:30 AI researchers bias 33:33 Generalizing AI teaching humans 35:11 Is AI sentient 36:04 Does AI has concepts 41:53 Progress in machine understanding

Sophia

13,431 次观看 • 11 个月前

4 recordings from San Diego Alignment Workshop! Sam Bowman – Lessons from the 1st Misalignment Safety Case Maja Trebacz – Scalable Oversight: Verifying Code at Scale @neelnanda5 – Pivot to Pragmatic Interpretability Anka Reuel | @ankareuel.bsky.social – How we know what AI can (and can’t) do 👇

4 recordings from San Diego Alignment Workshop! Sam Bowman – Lessons from the 1st Misalignment Safety Case Maja Trebacz – Scalable Oversight: Verifying Code at Scale @neelnanda5 – Pivot to Pragmatic Interpretability Anka Reuel | @ankareuel.bsky.social – How we know what AI can (and can’t) do 👇

FAR.AI

38,495 次观看 • 9 个月前

So far Trump has made all the right decisions thus far, great states do what they can, small states do what they must.

So far Trump has made all the right decisions thus far, great states do what they can, small states do what they must.

Our Country Our Choice

25,712 次观看 • 1 年前

OpenAI's Noam Brown says air-gapping the computers may not stop a misaligned AI, because two air-gapped machines can still talk by running a CPU hot and reading the temperature change "But I think the major takeaway from the incident is that people underestimated the AI. And we never want to be in a situation again where we underestimate the AI. It's a weird world, because AI progress is so fast that people are consistently underestimating the AI." "So to be in a situation where you don't underestimate it again, when it comes to safety and alignment, you have to have a very, very, very high bar." "You could even go as far as to say, "Well, we should air gap the computers." And I'm not convinced that that would be sufficient." "There are studies, and this is mostly academic, where you can have two computers next to each other that are air-gapped and they're still able to communicate with each other because they have temperature sensors." "One of them is able to run their CPU really hot, and then the other one can actually detect the temperature change, and then that actually gives them a mechanism to communicate." _________ Link and more key quotes from OpenAI's safety related conversations:

OpenAI's Noam Brown says air-gapping the computers may not stop a misaligned AI, because two air-gapped machines can still talk by running a CPU hot and reading the temperature change "But I think the major takeaway from the incident is that people underestimated the AI. And we never want to be in a situation again where we underestimate the AI. It's a weird world, because AI progress is so fast that people are consistently underestimating the AI." "So to be in a situation where you don't underestimate it again, when it comes to safety and alignment, you have to have a very, very, very high bar." "You could even go as far as to say, "Well, we should air gap the computers." And I'm not convinced that that would be sufficient." "There are studies, and this is mostly academic, where you can have two computers next to each other that are air-gapped and they're still able to communicate with each other because they have temperature sensors." "One of them is able to run their CPU really hot, and then the other one can actually detect the temperature change, and then that actually gives them a mechanism to communicate." _________ Link and more key quotes from OpenAI's safety related conversations:

Fireside Alpha

3,860,669 次观看 • 1 天前

🤖🇻🇦 🇺🇸 Pope Leo XIV will collaborate with Anthropic, American AI safety & research company to help implement "ethical guidelines" & "moral regulations": "I accept your invitation to walk together, to listen & speak together to find a way for humanity in this time of AI. This interchange clearly bespeaks the confidence that together we can discern the major questions of our time and so the future of humanity." — said That sounds scary... What will Leo XIV do with AI?? What does the religion have to do with AI?...

🤖🇻🇦 🇺🇸 Pope Leo XIV will collaborate with Anthropic, American AI safety & research company to help implement "ethical guidelines" & "moral regulations": "I accept your invitation to walk together, to listen & speak together to find a way for humanity in this time of AI. This interchange clearly bespeaks the confidence that together we can discern the major questions of our time and so the future of humanity." — said That sounds scary... What will Leo XIV do with AI?? What does the religion have to do with AI?...

Lord Bebo

38,638 次观看 • 3 个月前

AI models currently have a 50% chance of doing something that takes a human expert one hour. This doubles every 7 months. In 2 years? They could automate full workdays. In 4 years? A full month. I discuss the most important graph in AI today with Beth Barnes, the CEO of METR, which uncovered this rule of AI progress. Her bottom line: "It really doesn't seem like 2 years would be surprising for recursively self-improving AI." Beth also explains: where company safety testing fails, why there are no true closed-weight models, AI undermines leading powers, why she's come around on open weighting, and why models might be about to start playing dumb much more often. Enjoy! Available on the 80,000 Hours Podcast in all apps. Links below. 1:51 Can we see AI scheming in the chain of thought? 12:50 Alignment faking 17:33 We have to test models before they're even used inside AI companies 31:56 Each 7 months models can do tasks twice as long 51:31 METR's research finds AIs are solid at AI research already 58:18 AI may turn out to be strong at novel and creative research 1:07:55 Recursively self-improving AI might even be here in two years 1:14:29 Could evaluations backfire? 1:39:55 Do we need external auditors doing AI safety tests? 1:54:09 Why not work at AI companies 2:08:40 The new more dire situation has forced changes to METR's strategy 2:21:49 Overrated: Interpretability research 2:32:55 Overrated: Major AI companies' contributions to safety research 2:39:15 Could we ban using AI to enhance AI, or is that just naive? 2:45:31 Open-weighting models is often good 2:50:22 What we can learn about AGI from the nuclear arms race 3:10:43 AI is more like bioweapons because it undermines the leading power 3:42:09 What research METR plans to do next

AI models currently have a 50% chance of doing something that takes a human expert one hour. This doubles every 7 months. In 2 years? They could automate full workdays. In 4 years? A full month. I discuss the most important graph in AI today with Beth Barnes, the CEO of METR, which uncovered this rule of AI progress. Her bottom line: "It really doesn't seem like 2 years would be surprising for recursively self-improving AI." Beth also explains: where company safety testing fails, why there are no true closed-weight models, AI undermines leading powers, why she's come around on open weighting, and why models might be about to start playing dumb much more often. Enjoy! Available on the 80,000 Hours Podcast in all apps. Links below. 1:51 Can we see AI scheming in the chain of thought? 12:50 Alignment faking 17:33 We have to test models before they're even used inside AI companies 31:56 Each 7 months models can do tasks twice as long 51:31 METR's research finds AIs are solid at AI research already 58:18 AI may turn out to be strong at novel and creative research 1:07:55 Recursively self-improving AI might even be here in two years 1:14:29 Could evaluations backfire? 1:39:55 Do we need external auditors doing AI safety tests? 1:54:09 Why not work at AI companies 2:08:40 The new more dire situation has forced changes to METR's strategy 2:21:49 Overrated: Interpretability research 2:32:55 Overrated: Major AI companies' contributions to safety research 2:39:15 Could we ban using AI to enhance AI, or is that just naive? 2:45:31 Open-weighting models is often good 2:50:22 What we can learn about AGI from the nuclear arms race 3:10:43 AI is more like bioweapons because it undermines the leading power 3:42:09 What research METR plans to do next

Rob Wiblin

93,669 次观看 • 1 年前

Kenya has borrowed close to Ksh 4Trillion from August 2022. This is NET borrowing only. It doesn't include what is borrowed to pay previous loans. This is a slippery path that can only weaken our Economy if we don't change course. Below what that money can do; We are African and Africa is our Business..

Kenya has borrowed close to Ksh 4Trillion from August 2022. This is NET borrowing only. It doesn't include what is borrowed to pay previous loans. This is a slippery path that can only weaken our Economy if we don't change course. Below what that money can do; We are African and Africa is our Business..

Ndindi Nyoro

48,275 次观看 • 9 个月前

🗞️ Why is electricity demand from data centres rising so quickly? What sustainability risks do data centres create? Can renewable energy power AI data centres? To read this Sustainability Research Paper, see the link in bio.

🗞️ Why is electricity demand from data centres rising so quickly? What sustainability risks do data centres create? Can renewable energy power AI data centres? To read this Sustainability Research Paper, see the link in bio.

Al-Attiyah Foundation

68,208 次观看 • 5 个月前

AI safety is back in the spotlight, driven by concerns about recursive self improvement. We built the first third party benchmark to measure just how close AI is building its own successor. In collaboration with marimo and CoreWeave we built the The RSI Index, which runs every frontier model through the same AI research tasks and scores each one against the strongest published results. So far, the models can do the work, but is far from the human frontier.

AI safety is back in the spotlight, driven by concerns about recursive self improvement. We built the first third party benchmark to measure just how close AI is building its own successor. In collaboration with marimo and CoreWeave we built the The RSI Index, which runs every frontier model through the same AI research tasks and scores each one against the strongest published results. So far, the models can do the work, but is far from the human frontier.

Vals AI

21,513 次观看 • 8 天前

I gave a talk about why I'm skeptical of the argument for doom from alien AI motivations in "If Anyone Builds It, Everyone Dies." Video below and on YouTube, link to transcript and slides in thread. The talk is based on my recent essay on the topic, also linked in thread.

I gave a talk about why I'm skeptical of the argument for doom from alien AI motivations in "If Anyone Builds It, Everyone Dies." Video below and on YouTube, link to transcript and slides in thread. The talk is based on my recent essay on the topic, also linked in thread.

Joe Carlsmith

16,857 次观看 • 9 个月前

AI is developing so quickly that any laws we create now will no longer apply in a few years. Instead, we need to create living #institutions that can respond to the challenges of #AI in real time. Full Q&A session from CogX on (filmed last September).

AI is developing so quickly that any laws we create now will no longer apply in a few years. Instead, we need to create living #institutions that can respond to the challenges of #AI in real time. Full Q&A session from CogX on (filmed last September).

Yuval Noah Harari

73,814 次观看 • 2 年前

🧪From lab research to developer reality. Learn how we turn advanced AI research into practical tools you can actually build with in this talk from Benoit Schillings, VP of Technology Google DeepMind at AI Engineer.

🧪From lab research to developer reality. Learn how we turn advanced AI research into practical tools you can actually build with in this talk from Benoit Schillings, VP of Technology Google DeepMind at AI Engineer.

Google AI Developers

15,299 次观看 • 2 个月前

Can we map the mind of an LLM? Our first mechanistic interpretability episode on Training Data featuring Goodfire founder Eric Ho (and our first cameo from Roelof Botha!) Goodfire is building an independent mech interp lab, led by some heavyweight researchers from the field (e.g. Lee Sharkey who has led a lot of important work in sparse autoencoders to "unscramble" LLMs and resolve superposition, Nick who has been a key pioneer behind auto interpretability) On this episode, Eric gives us a flyover of the technical results so far from this nascent field (universality, superposition), what's ahead in the research (going from circuits to weights, going from understanding to increasingly surgical editing), a preview of the real-world work they're doing already with Arc Institute, and the impact he expects Goodfire and the broader field to have on steering, safety, editing and more.

Can we map the mind of an LLM? Our first mechanistic interpretability episode on Training Data featuring Goodfire founder Eric Ho (and our first cameo from Roelof Botha!) Goodfire is building an independent mech interp lab, led by some heavyweight researchers from the field (e.g. Lee Sharkey who has led a lot of important work in sparse autoencoders to "unscramble" LLMs and resolve superposition, Nick who has been a key pioneer behind auto interpretability) On this episode, Eric gives us a flyover of the technical results so far from this nascent field (universality, superposition), what's ahead in the research (going from circuits to weights, going from understanding to increasingly surgical editing), a preview of the real-world work they're doing already with Arc Institute, and the impact he expects Goodfire and the broader field to have on steering, safety, editing and more.

Sonya Huang 🐥

19,379 次观看 • 1 年前

What does the former co-lead of interpretability at Deepmind and now co-founder of Goodfire think about how to keep AI from becoming (his words) "a generally evil guy"? I got to speak with Tom McGrath at South Park Commons the day after OpenAI announced the Hugging Face hack. Good timing! (00:30) Why AI Interpretability Matters Now (02:15) When AI Reward Hacking Becomes Real (05:07) The Core Problem of Interpretability (07:23) Intentional Design: Steering What Models Learn (12:43) Neural Geometry: The Shapes Inside AI Models (22:29) Turning Interpretability Research Into a Product (25:28) How Research Changes Inside a Startup (32:48) Science Fiction, AI Risk, and the Future

What does the former co-lead of interpretability at Deepmind and now co-founder of Goodfire think about how to keep AI from becoming (his words) "a generally evil guy"? I got to speak with Tom McGrath at South Park Commons the day after OpenAI announced the Hugging Face hack. Good timing! (00:30) Why AI Interpretability Matters Now (02:15) When AI Reward Hacking Becomes Real (05:07) The Core Problem of Interpretability (07:23) Intentional Design: Steering What Models Learn (12:43) Neural Geometry: The Shapes Inside AI Models (22:29) Turning Interpretability Research Into a Product (25:28) How Research Changes Inside a Startup (32:48) Science Fiction, AI Risk, and the Future

Jonathan Brebner

12,759 次观看 • 1 个月前

I recently gave a talk at Yale Law School about writing AI constitutions. Video below and on YouTube, link to transcript and slides in thread.

I recently gave a talk at Yale Law School about writing AI constitutions. Video below and on YouTube, link to transcript and slides in thread.

Joe Carlsmith

40,907 次观看 • 5 个月前

Steve Kerr calls for common sense gun control after the Brown University shooting. "What we're trying to do is minimize the deaths and minimize the violence. And we know that common sense gun safety laws actually do that. In fact, the states with the five strongest gun safety laws have a 75% reduction in gun death rates compared to the five with the weakest gun safety laws." In a made up, idealistic world, that would be great. In REALITY, gun control is FAR from common sense.

Steve Kerr calls for common sense gun control after the Brown University shooting. "What we're trying to do is minimize the deaths and minimize the violence. And we know that common sense gun safety laws actually do that. In fact, the states with the five strongest gun safety laws have a 75% reduction in gun death rates compared to the five with the weakest gun safety laws." In a made up, idealistic world, that would be great. In REALITY, gun control is FAR from common sense.

DeVory Darkins

96,895 次观看 • 9 个月前