PODCAST

LessWrong (30+ Karma)

LessWrong

Audio narrations of LessWrong posts.

technology philosophy

All Episodes

15:02

“Your Causal Variables Are Irreducibly Subjective” by...

LessWrong (30+ Karma) ·

2026/03/07

Mechanistic interpretability needs its own shoe leather era. Reproducing the labeling process will matter more than reproducing the Github. Crossposted from Communication & Intelligence substack When we try to understand large language models, we like to invoke causality. And who can blame us? Causal inference comes with an impressive toolkit: directed acyclic graphs, potential outcomes, mediation analysis, formal identification results. It feels crisp. It feels reproducible. It feels like

en-gb

19:38

“Mox is the largest AI Safety community space in San...

LessWrong (30+ Karma) ·

2026/03/07

Summary: Mox is fundraising to maintain and grow AIS projects, build a compelling membership, and foster other impactful and delightful work. We're looking to raise $450k for 2026, and you can donate on Manifund! Overview Who we are Mox is SF's largest AI safety coworking space, and also its primary Effective Altruism community space. We opened just over a year ago, and over the last year, we’ve served high-impact work in and around AI safety by hosting conferences, fellowships, events, and

en-gb

8:29

“Podcast: Jeremy Howard is bearish on LLMs” by Steven...

LessWrong (30+ Karma) ·

2026/03/06

Jeremy Howard was recently[1] interviewed on the Machine Learning Street Talk podcast: YouTube link, interactive transcript, PDF transcript. Jeremy co-invented LLMs in 2018, and taught the excellent fast.ai online course from which I learned much ML in ~2018, and he uses LLMs all the time, e.g. 90% of his new code is typed by an LLM (see below). So I think his “bearish”[2] take on LLMs is an interesting datapoint, and I’m putting it out there for discussion. Some relevant excerpts from the podcast,

en-gb

11:13

“Thoughts on the Pause AI protest” by philh

LessWrong (30+ Karma) ·

2026/03/06

On Saturday (Feb 28, 2026) I attended my first ever protest. It was jointly organized by PauseAI, Pull the Plug and a handful of other groups I forget. I have mixed feelings about it. To be clear about where I stand: I believe that AI labs are worryingly close to developing superintelligence. I won't be shocked if it happens in the next five years, and I'd be surprised if it takes fifty years at current trajectories. I believe that if they get there, everyone will die. I want these labs to stop

en-gb

36:02

“Anthropic Officially, Arbitrarily and Capriciously...

LessWrong (30+ Karma) ·

2026/03/06

Make no mistake about what is happening.The Department of War (DoW) demanded Anthropic bend the knee, and give them ‘unfettered access’ to Claude, without understanding what that even meant. If they didn’t get what they want, they threatened to both use the Defense Production Act (DPA) to make Anthropic give the military this vital product, and also designate the company a supply chain risk (SCR). Hegseth sent out an absurdly broad SCR announcement on Twitter that had absolutely no legal basis,

en-gb

30:33

“The Elect” by Tomás B.

LessWrong (30+ Karma) ·

2026/03/06

I was different in Michael's prison than I was outside, looking the way I did when we fell in love so long ago, in that time before we could change our forms. Stuck in some body that was not of my choosing? Does that seem strange to you? It was not like that for me. It is just how things were for most of history, and few imagined this changing. So I felt almost nostalgic as I entered his realm, his prison transforming me into my first self - though not precisely as she was, instead as he remembered

en-gb

34:08

“Models have linear representations of what tasks...

LessWrong (30+ Karma) ·

2026/03/06

This work was done as part of MATS 9.0, mentored by Patrick Butlin. All mistakes are mine. I'm posting this as a research report to get feedback. Please red-team, comment, and reach out. Thanks to Patrick Butlin for supervising, Daniel Paleka for regular feedback. Thanks to Patrick Butlin, Pierre Beckmann, Austin Meek, Elias Kempf and Rob Adragna for comments on the draft. TLDR: We train probes on Gemma-3-27b revealed preferences. We find that these generalise ood to system-prompt induced

en-gb

12:49

“AI Safety Has 12 Months Left” by mhdempsey

LessWrong (30+ Karma) ·

2026/03/06

The past decade of technology has been defined by many wondering what the upper bound of power and influence is for an individual company. The core concern about AI labs is that the upper bound is infinite.[1] This has led investors to direct all of their mindshare towards deploying into AI, the tech industry to become incredibly politically engaged, the markets to have a minimal amount of conviction (and a lot of volatility), and governments to attempt to classify labs as borderline domestic

en-gb

8:59

“Reasoning Models Struggle to Control Their Chains of...

LessWrong (30+ Karma) ·

2026/03/06

Authors: Yueh-Han Chen, Robert McCarthy, Bruce W. Lee, He He, Ian Kivlichan, Bowen Baker, Micah Carroll, Tomek Korbak In collaboration with OpenAI TL;DR: Chain-of-thought (CoT) monitoring can detect misbehavior in reasoning models, but only if models cannot control what they verbalize. To measure this undesirable ability, CoT Controllability, we introduce CoT-Control, an evaluation suite requiring models to solve problems while adhering to CoT constraints (e.g., reasoning about genetics without

en-gb

20:13

“Have Americans Become Less Violent Since 1980?” by...

LessWrong (30+ Karma) ·

2026/03/06

The US homicide rate hit 10.2 per 100,000 in 1980, fell to 4.4 by 2014, then spiked to 6.5 in 2020. This looks like a story about violence rising, falling, and rising again. The actual story is simpler and worse: violence tripled in the 1960s and 70s. After adjusting for medical improvement, there is no clear statistical signal that it ever came back down. What changed was not the violence but how often it killed people. Deaths vs. attacks The homicide rate measures deaths. Deaths from gunshot

en-gb

22:20

“Personality Self-Replicators” by eggsyntax

LessWrong (30+ Karma) ·

2026/03/06

One-sentence summary I describe the risk of personality self-replicators, the threat of OpenClaw-like agents managing to spread in hard-to-control ways. Summary LLM agents like OpenClaw are defined by a small set of text files and run in an open source framework which leverages LLMs for cognition. It is quite difficult for current frontier models to self-replicate, it is much easier for such agents (at the cost of greater reliance on external agents). While not a likely existential threat, such

en-gb

24:27

“Text Compression Can Help Secure Model Weights” by...

LessWrong (30+ Karma) ·

2026/03/05

Associated Paper: link[1] Paper Co-authors: Roy Rinberg, Annabelle Michael Carrell, Simon Henniger, Nicholas Carlini, Keri Warr Two years ago, Ryan Greenblatt wrote a post titled "Preventing model exfiltration with upload limits". The idea of the original post is simple: if there's a single wire connecting to a GPU that allows egress of 100 GB per day, then even if an adversary has full control of the GPU, there's nothing they can do to exfiltrate more than 100 GB in a single day. The main

en-gb

96:38

“AI #158: The Department of War” by Zvi

LessWrong (30+ Karma) ·

2026/03/05

This was the worst week I have had in quite a while, maybe ever.The situation between Anthropic and the Department of War (DoW) spun completely out of control. Trump tried to de-escalate by putting out a Truth merely banning Anthropic from direct use by the Federal Government with a six month wind down. Then Secretary of War Hegseth went rogue and declared Anthropic a supply chain risk, with wording indicating an intent to outright murder Anthropic as a company. Then that evening OpenAI signed a

en-gb

14:11

“Split Personality Training can detect Alignment...

LessWrong (30+ Karma) ·

2026/03/05

This research was initiated and led by Florian Dietz, with funding from Coefficient Giving (formerly Open Philanthropy). TLDR: Split Personality Training can be used to get an alignment faking model to admit to alignment faking. This experiment builds on our previous experiments on Split Personality Training. I have updated that article with a new section. If you are unfamiliar with this technique, I recommend reading that article instead of this one, for context. Recap of Split Personality

en-gb

18:24

“A summary of Condensation and its relation to...

LessWrong (30+ Karma) ·

2026/03/05

Short summary of Condensation Condensation is a theory of concepts by Sam Eisenstat. The paper can be read here. Abram wrote a review on Lesswrong, and a followup. The paper is very much worth reading, and can be skimmed to just understand the motivation if you’re time constrained. What is a concept? One rough pointer is "the little packets of information that are the meaning of a word or phrase". Another is the clusters found in the cluster structure of thingspace. Maybe we could think of them as

en-gb

15:24

“Maybe there’s a pattern here?” by dynomight

LessWrong (30+ Karma) ·

2026/03/04

1. It occurred to me that if I could invent a machine—a gun—which could by its rapidity of fire, enable one man to do as much battle duty as a hundred, that it would, to a large extent supersede the necessity of large armies, and consequently, exposure to battle and disease [would] be greatly diminished. Richard Gatling (1861) 2. In 1923, Hermann Oberth published The Rocket to Planetary Spaces, later expanded as Ways to Space Travel. This showed that it was possible to build machines that could

en-gb

8:31

“Is GDP a Kind of Factory?d Draft” by Benquo

LessWrong (30+ Karma) ·

2026/03/04

In 2021, economists Arvind Subramanian, Justin Sandefur, and Dev Patel announced that poor countries had finally started catching up to rich ones, vindicating the Solow growth model's prediction of "convergence." Now they say the rumors of Solow convergence's vindication were premature; the convergence trend reversed and poor countries are falling behind again. David Oks, writing about their retraction, argues that the whole convergence episode was a mirage produced by the Chinese commodities boom

en-gb

9:25

“Mass surveillance, red lines, and a crazy weekend”...

LessWrong (30+ Karma) ·

2026/03/04

[These are my own opinions, and not representing OpenAI. Cross-posted on windowsontheory.] AI has so many applications, and AI companies have limited resources and attention span. Hence if it was up to me, I’d prefer we focus on applications that are purely beneficial— science, healthcare, education — or even commercial, before working on anything related to weapons or spying. If someone has to do it, I’d prefer it not to be my own company. Alas, we can’t always get what we want. This is a long-ish

en-gb

11:04

“Sacred values of future AIs” by Cleo Nardo

LessWrong (30+ Karma) ·

2026/03/04

Consider a future with many diverse AIs that need to coordinate with each other, or at least coexist without conflict. Such AIs would need shared values they can coordinate around. According to Hanson's theory, groups of diverse agents facing coordination pressure will tend to sacralize some shared value — seeing it in “far mode” so they can see it together. Unfortunately, this makes them systematically worse at making decisions about these things. This model suggests three claims: Helpfulness,

en-gb

17:16

“Physics of RL: Toy scaling laws for the emergence of...

LessWrong (30+ Karma) ·

2026/03/04

TL;DR: When is or isn't reward the optimization target? I use a mathematical toy model to reason about when RL should select for reward-seeking reasoning as opposed to behaviors that achieve high reward without thinking about reward. Hypothesis: More diverse RL increases the likelihood that reward-seeking emerges. In my toy model, there are scaling laws for the emergence of reward-seeking: when decreasing prior probability of reward-seeking reasoning we can increase RL diversity enough that

en-gb

4,474 results

LessWrong (30+ Karma)

All Episodes

Similar Podcasts

ev.news

Martyn Lee

TechCrunch Industry News

TechCrunch

Elon Musk Podcast

Stage Zero

Reflections

Livewire Labs

TechCrunch Startup News

TechCrunch