PODCAST

LessWrong (30+ Karma)

LessWrong

Audio narrations of LessWrong posts.

technology philosophy

All Episodes

16:24

“Negation Neglect: When models fail to learn...

LessWrong (30+ Karma) ·

2026/05/18

This is a short summary of our new paper: arXiv, X thread, code. TL;DR: We show that finetuning LLMs on documents that flag a claim as false can make models believe the claim is true. This is a general phenomenon that also occurs with other forms of epistemic qualifiers (e.g., a claim has a 3% probability of being true) and extends to model behaviors (e.g., warning against types of misalignment). This effect occurs in all models tested. Authors: Harry Mayne*, Lev McKinney*, Jan Dubiński, Adam

en-gb

10:27

“Classifier Context Rot: Monitor Performance Degrades...

LessWrong (30+ Karma) ·

2026/05/18

Monitoring coding agents for dangerous behavior using language models requires classifying transcripts that often exceed 500 thousand tokens, but prior agent monitoring benchmarks rarely contain transcripts longer than 100 thousand tokens. We show that when used as classifiers, current frontier models fail to notice dangerous actions more often in longer transcripts. In particular, on MonitorBench, Opus 4.6, GPT 5.4, and Gemini 3.1 miss these actions 2x to 30x more often when we prepend 800

en-gb

11:58

“why pollen allergies?” by bhauth

LessWrong (30+ Karma) ·

2026/05/18

Allergies are a big problem for a lot of people. If you're someone with pollen allergies, maybe you've wondered how people in the distant past dealt with them. After all, a thousand years ago people mostly worked outside all day, in areas where plants grow well. They had no air purifiers, no allergy medication, and no extra food for people who couldn't work when it was time to plant crops. The answer is, pollen allergies just weren't very common back then. There was a massive increase in their

en-gb

7:35

“How to Quit Fandom: Apostasy” by Laiba Rehman

LessWrong (30+ Karma) ·

2026/05/18

[Crossposted from my blog, BlueprintingHeaven.] Please note that the views I held and thoughts I had during the time immediately after my deconversion are not necessarily ones I have right now. It would be boring and annoying to constantly say this in the post, so here is your advance disclaimer! When I tell people I switched from wanting to do astrophysics to wanting to prevent death upon becoming an atheist, they’re usually not too surprised or confused. Sometimes they are, and I might make a

en-gb

13:58

“James C. Scott: Seeing Like a State” by Martin Sustrik

LessWrong (30+ Karma) ·

2026/05/17

In 1932-33, Soviet collectivization destroyed local farming knowledge and produced a famine that killed somewhere between five and nine million people. It was one of the twentieth century's great tragedies, and James Scott's Seeing Like a State draws a straight line from the ideology that caused it — High Modernism, the belief that society can be rationally reorganized from above — to the disaster that followed. But here's a number that doesn’t appear in Scott's book. Eight billion. That's roughly

en-gb

8:53

“How to Reason about Your Health Issues” by Taylor G....

LessWrong (30+ Karma) ·

2026/05/17

Many people make costly mistakes when reasoning about their health. Even most doctors make this mistake, because it's not a mistake that's caused by a lack of medical knowledge. Rather, it's caused by a lack of clear thinking. People experience symptoms, and then they look for the root cause of their symptoms. For example, somone with heartburn or pain in their stomach might decide the root cause of their issues is excess stomach acid/GERD (GastroEsophageal Reflux Disease -- a disease affecting

en-gb

7:41

“Benchmarking Real Work” by kaivu, leni, rohuang, zef

LessWrong (30+ Karma) ·

2026/05/17

Thanks to Megan Kinniment for helpful comments and discussion. TL;DR: Benchmarks like HCAST undersample fuzzy (hard to evaluate) tasks, meaning they might overestimate capability on long-horizon work. To sample fuzzy tasks we need to increase judge capacity: we can either try to build automated judges that match human judgment, or reduce the human effort per grade. To do this, we propose generating fuzzy tasks as a byproduct of real SWE work — snapshot the repo and a proto-spec before starting, and

en-gb

5:13

“A relatively brief explanation of Boltzmann Brains”...

LessWrong (30+ Karma) ·

2026/05/16

(Initially written for the LW Wiki, but then I realized it was looking more like a post instead.) In 1895, the physicist Ignaz Robert Schütz, who worked as an assistant to the more eminent physicist Ludwig Boltzmann, wondered if our observed universe had simply assembled by a random fluctuation of order from a universe otherwise in thermal equilibrium. The idea was published by Boltzmann in 1896, properly credited to Schütz, and has been associated with Boltzmann ever since. The obvious objection

en-gb

22:17

“An Introduction to Exemplar Partitioning for...

LessWrong (30+ Karma) ·

2026/05/16

Most of what we currently call "feature discovery" in language models is wrapped up in dictionary-learning methods like sparse autoencoders (SAEs) – which work, and which have been scaled to millions of features on frontier-scale models, but which bundle two distinct commitments into a single training objective: a reconstruction loss and a sparsity loss over a fixed size dictionary. Those commitments make sense if your goal is reconstructive decomposition – if you want to take an activation and

en-gb

18:50

“A Year Late, Claude Finally Beats Pokémon” by Julian...

LessWrong (30+ Karma) ·

2026/05/16

Credit: ClaudePlaysPokemon Elevator Shanty by Kurukkoo Disclaimer: like some previous posts in this series, this was not primarily written by me, but by a friend. I did substantial editing, however. ClaudePlaysPokemon feat. Opus 4.7 has finally beaten Pokémon Red, fulfilling the challenge set over a year ago when LLMs playing Pokémon went briefly, slightly viral. Victory Screen! Let's get the throat-clearing out of the way: this doesn't make 4.7 a clear breakthrough in intelligence over 4.6 or 4.5

en-gb

46:22

“Monthly Roundup #42: May 2026” by Zvi

LessWrong (30+ Karma) ·

2026/05/16

At least we probably won’t have another pandemic. And we still have a partial Jones Act waiver. For now.Small victories. Table of Contents Hanta Hanta I Don’t Wanta. Bad News. Predictions Can Be Easy Even About The Future. Good Advice. The Efficient Market Hypothesis Is False. There Are Four Skills. While I Cannot Condone This. Good News, Everyone. For Your Entertainment. Gamers Gonna Game Game Game Game Game. The Spire Sleeps And So Shall I. I Was Promised Flying Self-Driving

en-gb

11:42

“Incriminating misaligned AI models via distillation”...

LessWrong (30+ Karma) ·

2026/05/15

Suppose we have a dangerous misaligned AI that can fool alignment audits, and distill it into a student model. Two things can happen: Misalignment fails to transfer to the student. If so, we get a fairly capable benign model.Misalignment transfers to the student. The student might also be worse than the teacher at hiding its misalignment (e.g., due to being less capable). If so, we might get indirect evidence about the teacher's misalignment by auditing the distilled model. In this post, we will

en-gb

25:38

“The hard core of alignment (is robustifying RL)” by...

LessWrong (30+ Karma) ·

2026/05/15

Most technical AI safety work that I read seems to miss the mark, failing to make any progress on the hard part of the problem. I think this is a common sentiment, but there's less agreement about what exactly the hard part is? Characterizing this more clearly might save a lot of time and better target the search for solutions. In this post I explain my model of why alignment is technically hard to achieve, setting aside the regulatory, competitive, and geopolitical challenges, the sheer

en-gb

4:04

“Announcing the Center for Shared AI Prosperity” by...

LessWrong (30+ Karma) ·

2026/05/15

I wanted to share the launch of a project I've been working on with pollster David Shor, Obama/Biden veteran Stef Feldman, political strategist Morris Katz, Harvard historian Marc Aidinoff, and a few other folks*. The Center for Shared AI Prosperity is an attempt to force DC policy elites, particularly (given our team's backgrounds) liberals/progressives, to take the impending economic impacts of advanced AI more seriously. We do not think this is a normal economic shock. We are deeply uncertain

en-gb

11:11

“Risk reports need to address deployment-time spread...

LessWrong (30+ Karma) ·

2026/05/15

Risk reports commonly use pre-deployment alignment assessments to measure misalignment risk from an internally deployed AI. However, an AI that genuinely starts out with largely benign motivations can develop widespread dangerous motivations during deployment. I think this is the most plausible route to consistent adversarial misalignment in the near future. So, AI companies and evaluators should substantively incorporate it into risk analysis and planning. In this post, I’ll briefly argue why,

en-gb

9:09

“Mechanistic estimation for expectations of random...

LessWrong (30+ Karma) ·

2026/05/15

We have developed some relatively general methods for mechanistic estimation competitive with sampling by studying problems that are expressible as expectations of random products. This includes several different estimation problems, such as random halfspace intersections, random #3-SAT and random permanents. In this post, we will give a high-level introduction to these methods before sharing some more detailed notes. This is intended as an interim technical update and will be relatively light on

en-gb

28:48

“MATS 9 Retrospective & Advice” by beyarkay

LessWrong (30+ Karma) ·

2026/05/15

I couldn’t find a recent write-up from a MATS alum about what attending MATS was like, so this is the thing that I wish I had. I attended MATS from January to March 2026, on Team Shard with Alex Turner and Alex Cloud. It was a great time! Applications for MATS are basically on a rolling basis nowadays, and I can strongly recommend applying (to multiple streams) even if you think you’re not a great match. With that being said, there's a lot I wish I knew going into MATS, so here's a brain-dump of

en-gb

3:58

[Linkpost] “Don’t be too Clever to Take Obvious...

LessWrong (30+ Karma) ·

2026/05/15

This is a link post. An insidious pattern among smart people is feeling that because something is familiar and obvious, you are impervious to ignoring or forgetting it. In challenging times, I have often heard these clichés and reflexively shrugged them off. “Oh, I should dust myself off and pick myself up? What a lazy aphorism. What a patronising throwaway line. They must think I’m some kind of idiot. No, it must be something else…” There is a filter in many people's heads that functions to

en-gb

3:24

“Verification-Centric AI” by Raemon

LessWrong (30+ Karma) ·

2026/05/15

"Sometimes the AI just makes stuff up" is a problem I don't really expect to go away. In the nearterm, AI is going to keep occasionally hallucinating, or misinterpreting information. Eventually, AI will be powerful enough we need to be worried if it's presenting misleading information on purpose. There might be a nice window where the AI is powerful enough to not make things up but non-agentic enough that we don't have to worry about deliberate manipulation. But, even then, interpreting data is

en-gb

13:45

“Convergent Abstraction Hypothesis” by Jan_Kulveit

LessWrong (30+ Karma) ·

2026/05/15

Tl;dr Convergent abstraction hypothesis posits abstractions are often convergent in the sense of convergent evolution: different cognitive systems converge on the same abstraction, when facing similar selection pressures and learning in similar environments. It is a less ambitious alternative to 'natural abstractions hypotheses' and, in my view, more likely to be true. Convergence may be real, useful, and empirically robust, while still being contingent and fragile under changes in architecture,

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