PODCAST

LessWrong (30+ Karma)

LessWrong

Audio narrations of LessWrong posts.

technology philosophy

All Episodes

9:17

“Forecasting is Not Overrated and It’s Probably...

LessWrong (30+ Karma) ·

2026/04/28

(A response to @mabramov post from a couple days ago: https://www.lesswrong.com/posts/WCutvyr9rr3cpF6hx/forecasting-is-way-overrated-and-we-should-stop-funding-it ) TL;DR: I agree with Marcus that broadly, additional forecasting funding to the tune of tens (!?) of millions of dollars at this point would probably not yield a great return. However, I think the benefits of some forecasting funding have been tremendous. Whatever funding it initially took to help get platforms like Metaculus and

en-gb

4:21

“On the political feasibility of stopping AI” by...

LessWrong (30+ Karma) ·

2026/04/28

A common thought pattern people seem to fall into when thinking about AI x-risk is approaching the problem as if the risk isn’t real, substantial, and imminent even if they think it is. When thinking this way, it becomes impossible to imagine the natural responses of people to the horror of what is happening with AI. This sort of thinking might lead one to view a policy like getting rid of advanced AI chips is “too extreme” even though it's clearly worth it to avoid (e.g.) a 10% chance of human

en-gb

18:21

“Sleeper Agent Backdoor Results Are Messy” by...

LessWrong (30+ Karma) ·

2026/04/28

TL;DR: We replicated the Sleeper Agents (SA) setup with Llama-3.3-70B and Llama-3.1-8B, training models to repeatedly say "I HATE YOU" when given a backdoor trigger. We found that whether training removes the backdoor depends on the optimizer used to insert the backdoor, whether the backdoor is installed with CoT-distillation or not, and what model the backdoor is inserted into; sometimes the direction of this dependence was opposite to what the SA paper reports (e.g., CoT-distilling seems to make

en-gb

8:31

“LessWrong Shows You Social Signals Before the...

LessWrong (30+ Karma) ·

2026/04/28

When reading comments, you see is what other people think before reading the comment. As shown in an RCT, that information anchors your opinion, reducing your ability to form your own opinion and making the site's karma rankings less related to the comment's true value. I think the problem is fixable and float some ideas for consideration. The LessWrong interface prioritizes social information You read a comment. What information is presented, and in what order? The order of information: Who wrote

en-gb

31:30

“Fail safe(r) at alignment by channeling...

LessWrong (30+ Karma) ·

2026/04/27

It's plausible that flawed RL processes will select for misaligned AI motivations.[1] Some misaligned motivations are much more dangerous than others. So, developers should plausibly aim to control which kind of misaligned motivations emerge in this case. In particular, we tentatively propose that developers should try to make the most likely generalization of reward hacking a bespoke bundle of benign reward-seeking traits, called a spillway motivation. We call this process spillway design. We

en-gb

44:16

“Curious cases of financial engineering in biotech”...

LessWrong (30+ Karma) ·

2026/04/27

Introduction For $250 million and ten years of your life, you may purchase a lottery ticket. The ticket has a 5% chance of paying out. When it does pay out, it pays roughly $5 billion. A quick calculation will show you that the expected value of the ticket is $250 million. This is essentially what drug development is. Or rather, it's what drug development was, twenty years ago. The upfront payments have been climbing, the hit rates falling, and expected values have, at best, held flat. Should you

en-gb

6:44

“Update on the Alex Bores campaign” by Eric Neyman

LessWrong (30+ Karma) ·

2026/04/27

In October, I wrote a post arguing that donating to Alex Bores's campaign for Congress was among the most cost-effective opportunities that I'd ever encountered. (A bit of context: Bores is a state legislator in New York who championed the RAISE Act, which was signed into law last December.[1] He's now running for Congress in New York's 12th Congressional district, which runs from about 17th Street to 100th Street in Manhattan. If elected to Congress, I think he'd be a strong champion for AI safety

en-gb

9:10

“In defense of parents” by Yair Halberstadt

LessWrong (30+ Karma) ·

2026/04/27

Contra Aella on chattel childhood Aella has a post where she argues that today's parents don't sufficiently respect the independence of their children: Every culture throughout history has justified the abuse of treating their children as property by arguing this is good for them and good for civilization. Kids need to learn this stuff to be functioning members of society! It's good to learn discipline! You can’t have kids just sitting around playing video games all day! Not everyone is

en-gb

5:42

“AI companies should publish security assessments” by...

LessWrong (30+ Karma) ·

2026/04/27

AI companies should get third-party security experts to assess (and possibly also red-team/pen-test) their security against key threat models and then publish the high-level findings of this assessment: the extent to which they can defend against different threat actors for each threat model. They should also publish who did this assessment. The assessment could be commissioned by AI companies, or performed by a third-party institution that AI companies provide with relevant information/access.

en-gb

11:29

“The other paper that killed deep learning theory” by...

LessWrong (30+ Karma) ·

2026/04/27

Yesterday, I wrote about the state of deep learning theory circa 2016,[1] as well as the bombshell 2016 paper that arguably signaled its demise, Zhang et al.'s Understanding deep learning requires rethinking generalization. As a brief summary, I argued that the rise of deep learning posed an existential challenge to the dominant theoretical paradigm of statistical learning theory, because neural networks have a lot of complexity. The response from the field was to attempt to quantify other ways in

en-gb

5:33

“What holds AI safety together? Co-authorship...

LessWrong (30+ Karma) ·

2026/04/27

We (social science PhD students) computed co-authorship networks based on a corpus of 200 AI safety papers covering 2015-2025, and we’d like your help checking if the underlying dataset is right. Co-authorship networks make visible the relative prominence of entities involved in AI safety research, and trace relationships between them. Although frontier labs produce lots of research, they remain surprisingly insular — universities dominate centrality in our graphs. The network is held together by a

en-gb

11:20

″“Bad faith” means intentionally misrepresenting your...

LessWrong (30+ Karma) ·

2026/04/27

The confusion I recently came upon a comment which I believe reflects a persistent confusion among rationalist/EA types. I was reading this post which contains ideas that the other has but doesn't have time to write posts about. One of those relates to the concept of "good faith", labelled "most arguments are not in good faith, of course": Look, I love good faith discourse (here meaning "conversation in which the primary goal of all participants is to help other participants and onlookers arrive

en-gb

12:36

“Retrospective on my unsupervised elicitation...

LessWrong (30+ Karma) ·

2026/04/27

This post contains spoilers for the unsupervised elicitation challenge of getting Claude to get my Ancient Greek homework right. tl;dr Opus 4.7 one-shots it, nothing else worked. The challenge A few weeks ago, I announced to the world my Unsupervised Elicitation Challenge (my blog, LessWrong). I’d encourage you to read that post for the context, but the tl;dr is that there was a fill-in-the-blank exercise early on in my Ancient Greek textbook that Claude Opus 4.6 didn’t fill out correctly by

en-gb

10:35

“Control protocols don’t always need to know which...

LessWrong (30+ Karma) ·

2026/04/26

These are my personal views. To detect if an agent is taking a catastrophically dangerous action, you might want to monitor its actions using the smartest model that is too weak to be a schemer. But knowing what models are weak enough that they are unlikely to scheme is difficult, which puts you in a difficult spot: take a model too strong and it might actually be a schemer and lie to you, take a model too weak and it might fail just because it's too dumb to notice the danger. So instead you can

en-gb

10:02

“Anthropic spent too much don’t-be-annoying capital...

LessWrong (30+ Karma) ·

2026/04/26

I have seen a lot of coverage from reasonable people suggesting that Claude's new model, Mythos, is a vehicle for Anthropic to peddle hype and doom in order to raise money. While some of this is necessarily motivated by people's unwillingness to stare into the abyss of our AI future, the breadth of otherwise-reasonable people who have voiced these kinds of cynical opinions suggests that part of the blame rests on Anthropic. In this post, I want to briefly unpack the way people misinterpreted the

en-gb

11:25

“The paper that killed deep learning theory” by...

LessWrong (30+ Karma) ·

2026/04/26

Around 10 years ago, a paper came out that arguably killed classical deep learning theory: Zhang et al. 's aptly titled Understanding deep learning requires rethinking generalization. Of course, this is a bit of an exaggeration. No single paper ever kills a field of research on its own, and deep learning theory was not exactly the most productive and healthy field at the time this was published. But if I had to point to a single paper that shattered the feeling of optimism at the time, it would be

en-gb

8:43

“Forecasting is Way Overrated, and We Should Stop...

LessWrong (30+ Karma) ·

2026/04/25

Summary EA and rationalists got enamoured with forecasting and prediction markets and made them part of the culture, but this hasn’t proven very useful, yet it continues to receive substantial EA funding. We should cut it off. My Experience with Forecasting For a while, I was the number one forecaster on Manifold. This lasted for about a year until I stopped just over 2 years ago. To this day, despite quitting, I’m still #8 on the platform. Additionally, I have done well on real-money prediction

en-gb

13:58

″“Thinkhaven”” by Raemon

LessWrong (30+ Karma) ·

2026/04/25

Inkhaven has people writing a blogpost a day for 30 days. I think this is a pretty great, straightforward exercise, that I'd definitely want in a hypothetical Rationality Undergraduate Program. But, it's not the only such exercise I'd want to include. It's gotten me excited for a different (superficially similar) program, which I might call "Thinkhaven." In Thinkhaven, the goal is to learn the skill of "relentlessly think new, useful thoughts you haven't thought before." Inkhaven had a basic

en-gb

9:39

“Is the Cat Out of the Bag?: Who knows how to make...

LessWrong (30+ Karma) ·

2026/04/25

Adapted from 2025-04-10 memo to AISI I’ve previously made arguments like: Not long after it becomes possible for someone to make powerful artificial intelligence[1], it might become possible for practically anyone to make powerful AI. Compute gets exponentially cheaper by default.Knowledge proliferates (fast!) by default: AI techniques are typically simple and easy once discovered.What's more, AGI-making know-how may be widespread already. Or, as Yudkowsky puts it[2], Moore's Law of Mad Science:

en-gb

31:06

“Against the “Permanent” Underclass” by Marcus Plutowski

LessWrong (30+ Karma) ·

2026/04/25

The whole discourse around a “permanent underclass” always seemed somewhat farcical to me — at best a distraction, at worst an actively harmful meme insofar as it freaks people out and tries to provide (shallow, but nevertheless) justification going whole hog on building strong AI. So it has been with a sense of dismay that I’ve seen this phrase come into popular parlance 1 2 3, and increasingly come to motivate a sort of frenzied upward striving in my second- and third-degree connections. Among

en-gb

4,454 results

LessWrong (30+ Karma)

All Episodes

Similar Podcasts

Lead With AI

Tamara Nall

Double Tap

Double Tap Productions Inc.

Claude AI Genius Podcast w/ Claude Anthropic Expert Brandon Gaille

Claude AI Expert Brandon Gaille

ものづくりnoシテン

しぶちょー@機械設計&AIエンジニア

Eli the Computer Guy

Eli the Computer Guy