最高の William Bostock ポッドキャスト (2024)

1
“Contra papers claiming superhuman AI forecasting ” by nikos, Peter Mühlbacher, Lawrence Phillips, dschwarz 18:18

9h ago18:18

18:18

[Conflict of interest disclaimer: We are FutureSearch, a company working on AI-powered forecasting and other types of quantitative reasoning. If thin LLM wrappers could achieve superhuman forecasting performance, this would obsolete a lot of our work.] Widespread, misleading claims about AI forecasting Recently we have seen a number of papers – (Sc…

1
“OpenAI o1 ” by Zach Stein-Perlman 0:22

1d ago0:22

0:22

This is a link post. --- First published: September 12th, 2024 Source: https://www.lesswrong.com/posts/bhY5aE4MtwpGf3LCo/openai-o1 --- Narrated by TYPE III AUDIO.

1
“The Best Lay Argument is not a Simple English Yud Essay ” by J Bostock 8:46

3d ago8:46

8:46

Epistemic status: these are my own opinions on AI risk communication, based primarily on my own instincts on the subject and discussions with people less involved with rationality than myself. Communication is highly subjective and I have not rigorously A/B tested messaging. I am even less confident in the quality of my responses than in the correc…

1
“My Number 1 Epistemology Book Recommendation: Inventing Temperature ” by adamShimi 5:19

5d ago5:19

5:19

In my last post, I wrote that no resource out there exactly captured my model of epistemology, which is why I wanted to share a half-baked version of it. But I do have one book which I always recommend to people who want to learn more about epistemology: Inventing Temperature by Hasok Chang. To be very clear, my recommendation is not just to get th…

1
“That Alien Message - The Animation ” by Writer 14:32

5d ago14:32

14:32

Our new video is an adaptation of That Alien Message, by @Eliezer Yudkowsky. This time, the text has been significantly adapted, so I include it below. Part 1 Picture a world just like ours, except the people are a fair bit smarter: in this world, Einstein isn’t one in a million, he's one in a thousand. In fact, here he is now. He's made all the sa…

1
“Pay Risk Evaluators in Cash, Not Equity ” by Adam Scholl 1:33

7d ago1:33

1:33

Personally, I suspect the alignment problem is hard. But even if it turns out to be easy, survival may still require getting at least the absolute basics right; currently, I think we're mostly failing even at that. Early discussion of AI risk often focused on debating the viability of various elaborate safety schemes humanity might someday devise—d…

1
“Survey: How Do Elite Chinese Students Feel About the Risks of AI? ” by Nick Corvino 23:46

7d ago23:46

23:46

Intro In April 2024, my colleague and I (both affiliated with Peking University) conducted a survey involving 510 students from Tsinghua University and 518 students from Peking University—China's two top academic institutions. Our focus was on their perspectives regarding the frontier risks of artificial intelligence. In the People's Republic of Ch…

1
“things that confuse me about the current AI market. ” by DMMF 4:04

13d ago4:04

4:04

Paging Gwern or anyone else who can shed light on the current state of the AI market—I have several questions. Since the release of ChatGPT, at least 17 companies, according to the LMSYS Chatbot Arena Leaderboard, have developed AI models that outperform it. These companies include Anthropic, NexusFlow, Microsoft, Mistral, Alibaba, Hugging Face, Go…

1
“Nursing doubts ” by dynomight 17:32

13d ago17:32

17:32

If you ask the internet if breastfeeding is good, you will soon learn that YOU MUST BREASTFEED because BREAST MILK = OPTIMAL FOOD FOR BABY. But if you look for evidence, you’ll discover two disturbing facts. First, there's no consensus about why breastfeeding is good. I’ve seen experts suggest at least eight possible mechanisms: Formula can’t fully…

1
“Principles for the AGI Race ” by William_S 31:17

14d ago31:17

31:17

Crossposted from https://williamrsaunders.substack.com/p/principles-for-the-agi-race Why form principles for the AGI Race? I worked at OpenAI for 3 years, on the Alignment and Superalignment teams. Our goal was to prepare for the possibility that OpenAI succeeded in its stated mission of building AGI (Artificial General Intelligence, roughly able t…

1
“The Information: OpenAI shows ‘Strawberry’ to feds, races to launch it ” by Martín Soto 6:13

16d ago6:13

6:13

Two new The Information articles with insider information on OpenAI's next models and moves. They are paywalled, but here are the new bits of information: Strawberry is more expensive and slow at inference time, but can solve complex problems on the first try without hallucinations. It seems to be an application or extension of process supervision …

1
“What is it to solve the alignment problem? ” by Joe Carlsmith 1:39:04

17d ago1:39:04

1:39:04

People often talk about “solving the alignment problem.” But what is it to do such a thing? I wanted to clarify my thinking about this topic, so I wrote up some notes. In brief, I’ll say that you’ve solved the alignment problem if you’ve: avoided a bad form of AI takeover, built the dangerous kind of superintelligent AI agents, gained access to the…

1
“Limitations on Formal Verification for AI Safety ” by Andrew Dickson 42:06

18d ago42:06

42:06

In the past two years there has been increased interest in formal verification-based approaches to AI safety. Formal verification is a sub-field of computer science that studies how guarantees may be derived by deduction on fully-specified rule-sets and symbol systems. By contrast, the real world is a messy place that can rarely be straightforwardl…

1
“Would catching your AIs trying to escape convince AI developers to slow down or undeploy? ” by Buck 7:00

19d ago7:00

7:00

Crossposted from the AI Alignment Forum. May contain more technical jargon than usual.I often talk to people who think that if frontier models were egregiously misaligned and powerful enough to pose an existential threat, you could get AI developers to slow down or undeploy models by producing evidence of their misalignment. I'm not so sure. As an …

1
“Liability regimes for AI ” by Ege Erdil 8:03

22d ago8:03

8:03

For many products, we face a choice of who to hold liable for harms that would not have occurred if not for the existence of the product. For instance, if a person uses a gun in a school shooting that kills a dozen people, there are many legal persons who in principle could be held liable for the harm: The shooter themselves, for obvious reasons. T…

1
“AGI Safety and Alignment at Google DeepMind:A Summary of Recent Work ” by Rohin Shah, Seb Farquhar, Anca Dragan 18:39

24d ago18:39

18:39

Crossposted from the AI Alignment Forum. May contain more technical jargon than usual.We wanted to share a recap of our recent outputs with the AF community. Below, we fill in some details about what we have been working on, what motivated us to do it, and how we thought about its importance. We hope that this will help people build off things we h…

1
“Fields that I reference when thinking about AI takeover prevention” by Buck 20:01

1M ago20:01

20:01

Crossposted from the AI Alignment Forum. May contain more technical jargon than usual.This is a link post.Is AI takeover like a nuclear meltdown? A coup? A plane crash? My day job is thinking about safety measures that aim to reduce catastrophic risks from AI (especially risks from egregious misalignment). The two main themes of this work are the d…

1
“WTH is Cerebrolysin, actually?” by gsfitzgerald, delton137 38:23

1M ago38:23

38:23

[This article was originally published on Dan Elton's blog, More is Different.] Cerebrolysin is an unregulated medical product made from enzymatically digested pig brain tissue. Hundreds of scientific papers claim that it boosts BDNF, stimulates neurogenesis, and can help treat numerous neural diseases. It is widely used by doctors around the world…

1
“You can remove GPT2’s LayerNorm by fine-tuning for an hour” by StefanHex 22:42

1M ago22:42

22:42

This work was produced at Apollo Research, based on initial research done at MATS. LayerNorm is annoying for mechanstic interpretability research (“[...] reason #78 for why interpretability researchers hate LayerNorm” – Anthropic, 2023). Here's a Hugging Face link to a GPT2-small model without any LayerNorm. The final model is only slightly worse t…

1
“Leaving MIRI, Seeking Funding” by abramdemski 3:40

1M ago3:40

3:40

This is slightly old news at this point, but: as part of MIRI's recent strategy pivot, they've eliminated the Agent Foundations research team. I've been out of a job for a little over a month now. Much of my research time in the first half of the year was eaten up by engaging with the decision process that resulted in this, and later, applying to g…

1
“How I Learned To Stop Trusting Prediction Markets and Love the Arbitrage” by orthonormal 4:00

1M ago4:00

4:00

This is a story about a flawed Manifold market, about how easy it is to buy significant objective-sounding publicity for your preferred politics, and about why I've downgraded my respect for all but the largest prediction markets. I've had a Manifold account for a while, but I didn't use it much until I saw and became irked by this market on the co…

1
“This is already your second chance” by Malmesbury 16:01

1M ago16:01

16:01

Cross-posted from Substack. 1. And the sky opened, and from the celestial firmament descended a cube of ivory the size of a skyscraper, lifted by ten thousand cherubim and seraphim. And the cube slowly landed among the children of men, crushing the frail metal beams of the Golden Gate Bridge under its supernatural weight. On its surface were inscri…

1
“0. CAST: Corrigibility as Singular Target” by Max Harms 19:40

1M ago19:40

19:40

Crossposted from the AI Alignment Forum. May contain more technical jargon than usual.What the heck is up with “corrigibility”? For most of my career, I had a sense that it was a grab-bag of properties that seemed nice in theory but hard to get in practice, perhaps due to being incompatible with agency. Then, last year, I spent some time revisiting…

1
“Self-Other Overlap: A Neglected Approach to AI Alignment” by Marc Carauleanu, Mike Vaiana, Judd Rosenblatt, Diogo de Lucena 23:21

1M ago23:21

23:21

Figure 1. Image generated by DALL-3 to represent the concept of self-other overlapMany thanks to Bogdan Ionut-Cirstea, Steve Byrnes, Gunnar Zarnacke, Jack Foxabbott and Seong Hah Cho for critical comments and feedback on earlier and ongoing versions of this work. Summary In this post, we introduce self-other overlap training: optimizing for similar…

1
“You don’t know how bad most things are nor precisely how they’re bad.” by Solenoid_Entity 8:43

1M ago8:43

8:43

TL;DR: Your discernment in a subject often improves as you dedicate time and attention to that subject. The space of possible subjects is huge, so on average your discernment is terrible, relative to what it could be. This is a serious problem if you create a machine that does everyone's job for them. See also: Reality has a surprising amount of de…

聞く価値のあるポッドキャスト

ポッドキャスト William Bostock

聞く価値のあるポッドキャスト

1
Searching for My Wives

William Bostock

1
LessWrong (Curated & Popular)

LessWrong

1
“Contra papers claiming superhuman AI forecasting ” by nikos, Peter Mühlbacher, Lawrence Phillips, dschwarz 18:18

1
“OpenAI o1 ” by Zach Stein-Perlman 0:22

1
“The Best Lay Argument is not a Simple English Yud Essay ” by J Bostock 8:46

1
“My Number 1 Epistemology Book Recommendation: Inventing Temperature ” by adamShimi 5:19

1
“That Alien Message - The Animation ” by Writer 14:32

1
“Pay Risk Evaluators in Cash, Not Equity ” by Adam Scholl 1:33

1
“Survey: How Do Elite Chinese Students Feel About the Risks of AI? ” by Nick Corvino 23:46

1
“things that confuse me about the current AI market. ” by DMMF 4:04

1
“Nursing doubts ” by dynomight 17:32

1
“Principles for the AGI Race ” by William_S 31:17

1
“The Information: OpenAI shows ‘Strawberry’ to feds, races to launch it ” by Martín Soto 6:13

1
“What is it to solve the alignment problem? ” by Joe Carlsmith 1:39:04

1
“Limitations on Formal Verification for AI Safety ” by Andrew Dickson 42:06

1
“Would catching your AIs trying to escape convince AI developers to slow down or undeploy? ” by Buck 7:00

1
“Liability regimes for AI ” by Ege Erdil 8:03

1
“AGI Safety and Alignment at Google DeepMind:A Summary of Recent Work ” by Rohin Shah, Seb Farquhar, Anca Dragan 18:39

1
“Fields that I reference when thinking about AI takeover prevention” by Buck 20:01

1
“WTH is Cerebrolysin, actually?” by gsfitzgerald, delton137 38:23

1
“You can remove GPT2’s LayerNorm by fine-tuning for an hour” by StefanHex 22:42

1
“Leaving MIRI, Seeking Funding” by abramdemski 3:40

1
“How I Learned To Stop Trusting Prediction Markets and Love the Arbitrage” by orthonormal 4:00

1
“This is already your second chance” by Malmesbury 16:01

1
“0. CAST: Corrigibility as Singular Target” by Max Harms 19:40

1
“Self-Other Overlap: A Neglected Approach to AI Alignment” by Marc Carauleanu, Mike Vaiana, Judd Rosenblatt, Diogo de Lucena 23:21

1
“You don’t know how bad most things are nor precisely how they’re bad.” by Solenoid_Entity 8:43

クイックリファレンスガイド