Skip to content
Sign in
Artwork for LessWrong (30+ Karma)
TechnologySociety & CulturePhilosophy

LessWrong (30+ Karma)

LessWrong

Audio narrations of LessWrong posts.

Play
Open in
  • 426 episodes
  • Avg 19 min
  • English

Episodes 426

NewestOldestLongest
All20Unplayed20In progress0Played0
Counted on this page — what you have heard stays on this device, so it is not something the list can be paged by.
  • October 9 · 18 min

    “Social Theory and Agent Foundations” by Richard_Ngo

    Lately I’ve been writing a lot about the sociopolitical considerations that economic-style reasoning misses. I view this work as continuous with my agent foundations research; in this post, I lay out some of the connections I see between sociology and agent foundations. I’ll particularly focus on attempts to pin down theories of rational social and political behavior, as an extension to (or replacement of) the theory of rational economic behavior originally laid out by Von Neumann and Morgenstern. This post will of course be far from exhaustive. However, it's worth noting that attempts to develop big-picture frameworks for social theory have grown surprisingly sparse. Berger describes the field as being deformed from two directions since the mid-1900s: on one side, by the “methodological fetishism” of “sociologists using increasingly sophisticated methods to study increasingly trivial topics”; and on the other side, by the “cultural revolution [that] sought to transform sociology from a science into an instrument of ideological advocacy”. Hence many of the thinkers I’ll be discussing were officially based in economics or political science departments. This post has three sections, each focused on a different topic: social and game-theoretic equilibria; the internalization of norms and values; and collective behavior. [...] --- Outline: (01:42) Social equilibria vs game-theoretic equilibria (07:20) Internalizing norms (13:54) Collective behavior The original text contained 6 footnotes which were omitted from this narration. --- First published: October 8th, 2026 Source: https://www.lesswrong.com/posts/DwHvtm8vLipAQPest/social-theory-and-agent-foundations --- Narrated by TYPE III AUDIO.

  • October 9 · 3 min

    “Stop telling people to pivot to AI safety, I beg you” by Haoxing Du

    Epistemic status: Feeling frustrated, and therefore somewhat uncharitable. I’m very frustrated with calls for people to “drop everything and pivot into AI safety” like this one. (See also this, this, this, this for recent calls for more people to go into AI safety; though my frustration was triggered by the Celeste's post.) There are far more talented people trying to get into AI safety than there are jobs for them. MLAB 2, which I did in 2022, was already competitive to get into, at about 5% acceptance rate (admitted about 40 out of 800 applicants). Things are much, much more competitive now. There is a whole pipeline of AI safety fellowships (and guides for applying to them). It is rumored that one recent cohort of the Anthropic Fellows program had over 10,000 applicants. (Anecdotally, I applied to the Anthropic Fellows program in early 2026, after working at Redwood, METR, and an ML startup, and I didn’t get in!) And, uh, just take a look at the AI safety jobs? The bottom of the salary range for METR's current “Member of Technical Staff, Embedded Assessments” job posting is 402 thousand dollars. This is higher than the median total comp for [...] The original text contained 1 footnote which was omitted from this narration. --- First published: October 9th, 2026 Source: https://www.lesswrong.com/posts/xLgJJC3CD3y3AhtKn/stop-telling-people-to-pivot-to-ai-safety-i-beg-you --- Narrated by TYPE III AUDIO.

  • October 9 · 10 min

    “On Kink & Status in the Rationalist Community” by johnswentworth

    This whole thing started with various crazies on bluesky writing hit pieces on the rationalist and AI alignment communities, as one does. One of their main beats was to note social ties between the AI alignment community and the Red Means No (RMN) consensual non-consent sex party community, then gesticulate wildly and claim that the AI alignment community is a dangerous sex cult. As one does. It felt to me like the responses I saw to those hitpieces from my own in-groups were… not consistently candid. It seemed obvious to me that the hitpieces were partially pointing to real things, even if those real things were wildly overblown and not actually problematic in practice. And I was pretty frustrated at my in-groups’ refusal to acknowledge what seemed to me obvious kernels of truth; it felt like my own friends were playing public relations games rather than being honest. I have now updated. I think I was wrong, there are actually fewer kernels of truth to the attacks than I thought, and in fact I was pretty surprised at just how few. This post is mainly about the story of that update, what I previously believed obvious, and what I [...] --- Outline: (01:20) The Story (03:01) What I Previously Believed (04:58) What I Now Believe (06:50) Propagating That Update (08:40) An Apology The original text contained 1 footnote which was omitted from this narration. --- First published: October 8th, 2026 Source: https://www.lesswrong.com/posts/dHGjqpbfmdEN7TssX/on-kink-and-status-in-the-rationalist-community --- Narrated by TYPE III AUDIO.

  • October 8 · 10 min

    “AI Vigil at Congress” by Elliot Callender

    I'm organizing a multi-week AI x-risk vigil at Congress starting November 9th. I'm posting here to solicit feedback on the core doc before I promote on other platforms. This is intended to be accessible and appealing to many audiences. I'm aiming for about 10,000 attendees, minimum 100. All advice/feedback is appreciated. Summary Companies are trying to build AI smarter than any human. For many years, experts have warned that superhuman AI (ASI) could wreak catastrophe, including human extinction. We're Americans who recognize the severity of their warnings, and we're going to Congress until it acts. We have a moral responsibility to protect our friends, families, and fellow citizens who didn't consent to these risks. Our movement must remain bipartisan and nonviolent to achieve its goal; namely, getting Congress to back a deal with China to stop building more powerful AI. There are many ways you can help: Join us at the Capitol starting November 9th Come to midterm campaign rallies in your state Connect us with grassroots organizations in your area Volunteer remote skills like media outreach Call your representatives' in-state offices (in-state offices hear from fewer callers, so each call matters more) Donate financially If [...] --- Outline: (00:32) Summary (01:40) AI risk (04:06) Congress must strike a deal (04:53) We The People (06:54) Bigger than our differences (07:54) Nonviolence (09:14) We must act The original text contained 2 footnotes which were omitted from this narration. --- First published: October 8th, 2026 Source: https://www.lesswrong.com/posts/EqfdfetDKxpFszpdD/ai-vigil-at-congress --- Narrated by TYPE III AUDIO.

  • October 8 · 33 min

    “Towards embedded evaluations for scheming propensities” by Dylan Bowman, Ezra Newman, Axel Højmark, Mia Hopman, Alex Meinke, Bronson Schoen, Marius Hobbhahn

    TL;DR: A safety case for scheming must be able to substantiate these four key claims: scheming was not incentivized during training, strong evaluations and red teaming suggest that the model does not have a propensity to scheme, scheming was never attempted during internal deployment, and scheming reasoning would be detected (e.g., via CoT monitoring). All four of these claims should be addressed by embedded evaluators with persistent, deep access. Introduction A scheming AI is one that covertly and strategically pursues goals its developers didn’t intend. It might hide and protect its true capabilities and objectives from the people and processes meant to align, monitor, and control it. In 2024, Apollo Research demonstrated that frontier models were capable of scheming in controlled settings, and in 2025 we collaborated with OpenAI to determine how well their alignment techniques mitigate scheming. Models must not be trained to scheme against humans, i.e. covertly and strategically pursue misaligned goals. Evaluations should find no propensity to scheme, models should never attempt to scheme in deployment, and their reasoning must remain legible so that scheming would be noticed if it occurred. Therefore, model developers must be able to confidently make these four claims, in order to [...] --- Outline: (00:43) Introduction (04:00) Evaluating scheming risk requires persistent, deep access (06:04) Four claims about scheming which must be verified (06:21) Claim 1: Scheming was not incentivized during training (09:00) Claim 1.1: Training did not directly incentivize scheming. (09:28) Claim 1.2: Training did not incentivize goals or drives that would motivate scheming. (11:49) Claim 1.3: The model did not attempt to subvert the training process itself. (12:44) Claim 2: Strong evaluations and red teaming suggest that the model does not have a propensity to scheme (14:09) Claim 2.1: Evaluations and red-teaming could not elicit scheming. (14:47) Claim 2.2: Evaluations and red-teaming could not elicit goals or drives that would motivate scheming. (15:37) Claim 3: Scheming was never attempted during internal deployment (16:18) Claim 3.1: Monitors are deployed across all internal deployment contexts that present a risk of scheming. (16:55) Claim 3.2: If scheming were to occur, monitors would detect it (17:42) Claim 3.3: No scheming was attempted in monitored internal deployment activity. (18:10) Claim 4: Scheming reasoning would be detected (18:58) Claim 4.1: Reasoning relevant to scheming would be elicited and legibly verbalized in chain-of-thought. (20:09) Embedded evaluators for alignment require persistent, deep access (31:56) Takeaways and Next Steps --- First published: October 8th, 2026 Source: https://www.lesswrong.com/posts/LZGuqoYHphet9sfZs/towards-embedded-evaluations-for-scheming-propensities --- Narrated by TYPE III AUDIO. --- Images from the article: Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

  • October 8 · 1 hr 48 min

    “AI #189: New Math” by Zvi

    The big drop of this week was not a new AI model. It was instead the biggest day (so far!) in the history of mathematics, as OpenAI dropped solutions to 90 of the top 500 open math problems, along with many others, reached on an average budget of three hours of Pro-level compute per question. This was kind of a big deal and I plan to cover it tomorrow. We did see Claude Haiku 5.5, which looks promising given it is only 0.10 dollars/0.50 dollars. My week was largely spent at The Curve. The entire conference was Chatham House, so I can’t give as many details as I would like, but I have a write-up here. I finally had a chance to post my coverage of model welfare for both Mythos/Fable 5.1 and Opus 5.5. I continue to think this is an important issue for anyone wanting to understand today's AIs, even if you are highly confident that model welfare does not matter directly, as it has many practical implications on top of that. Jay Clayton is the new AI Czar, and is at the head of a new taskforce. Given who else was in [...] --- Outline: (02:36) Language Models Offer Mundane Utility (02:46) Consumers Use AI (09:48) Language Models Don't Offer Mundane Utility (12:02) Huh, Upgrades (14:42) On Your Marks (16:53) Gamers Gonna Game Game Game Game Game (18:49) Choose Your Fighter (19:48) Get My Agent On The Line (25:15) Deepfaketown and Botpocalypse Soon (29:17) Fun With Media Generation (31:10) Cyber Lack of Security (34:12) Hugging the Face (35:18) Misaligned! (37:16) A Young Lady's Illustrated Primer (38:28) They Took Our Jobs (40:45) Corporations Are Not Superintelligences (42:56) Get Involved (43:11) Introducing (45:26) In Other AI News (45:59) Show Me the Money (48:18) Quiet Speculations (50:26) Quickly, There's No Time (52:26) He's Putting Together a Team (55:14) The Quest for Sane Regulations (57:37) Well, At Least They're Forecasters (58:41) Chip City (59:16) The Week in Audio (01:03:48) Stop, Stop, He's Already Dead (01:08:32) People Just Say Things (01:10:46) Take a Moment (01:12:24) [Artificial Intelligence] (01:13:07) The American People Really Hate AI (01:16:29) Rhetorical Innovation (01:19:55) Aligning a Smarter Than Human Intelligence is Difficult (01:20:11) Open Weight Models Are Unsafe And Nothing Can Fix This (01:23:22) Cooperative Alignment (01:33:45) Building the Field (01:36:05) People Are Worried About AI Killing Everyone (01:40:03) Other People Are Not As Worried About AI Killing Everyone (01:40:38) Please Speak Directly Into This Microphone (01:42:54) The Lighter Side --- First published: October 8th, 2026 Source: https://www.lesswrong.com/posts/DZPodskqJJNEkMLvH/ai-189-new-math --- Narrated by TYPE III AUDIO. --- Images from the article: Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

  • October 8 · 12 min

    “ASI and Liberty: Can we do better than a benevolent god?” by Raymond Douglas

    Zvi recently coined a nice distinction between AGI-pilled and ASI-pilled: as more people are reckoning with the speed of progress, some are beginning to come around to the fact that AI could be superhuman at a huge range of tasks in a way that radically reshapes society, but not that it could leap us to the end of the tech tree shortly afterwards and start making nanotech and dyson spheres. Hence, you see people saying things like “Because of AI we must rethink the entire notion of mathematical progress”, or “I have to make sure I amass lots of capital before most jobs get automated”: pretty crazy statements, and yet, still not quite grappling with how wild things could get in the next few years, let alone the next decade. I feel a particular frustration here as someone who is both pretty ASI-pilled and pretty into liberal democracy. And when I say liberal democracy I don’t so much mean the specific political structure that seems efficient at our current tech level but rather the set of underlying moral commitments like non-domination and a government of the people — commitments which I’m aware might soon be a lot more expensive. [...] The original text contained 4 footnotes which were omitted from this narration. --- First published: October 8th, 2026 Source: https://www.lesswrong.com/posts/eSMburFT9eRxHWijf/asi-and-liberty-can-we-do-better-than-a-benevolent-god --- Narrated by TYPE III AUDIO. --- Images from the article: Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

  • October 8 · 1 hr 25 min

    ″[Paper] J++ Lens: Jacobian Filtering Enables More Faithful Workspace Lenses” by Kola Ayonrinde

    Audio note: this post contains a lot of mathematical notation. We narrate the simpler formulas and flag the complex ones, so you may want the original text to hand. Kola Ayonrinde, Anthropic Fellows, koayon@gmail.com; Jack Lindsey, Anthropic TL;DR: The J-Lens was proposed to read verbalisable representations from language model activations. We introduce the J++ Lens: an improvement to the J-Lens that enables more faithful readouts across 9–284B models. The J++ Lens filters noisy gradients before they enter the averaged Jacobian. Using the J++ Lens as a drop-in replacement for the J-Lens, we read workspace representations with 55% success on latent variable extraction tasks (J-Lens: 36%, R-Lens: 38%) with especially strong performance at early layers. We believe that more faithful lenses can be useful in monitoring for eval awareness and unverbalised scheming. We release code, an interactive demo and open-source lenses. Figure 1. By filtering out noisy gradients, the J++ Lens reads intermediate variables more reliably (a) and at earlier layers (b) than the J-Lens and R-Lens. The J++ Lens improves on the J-Lens by adding: (1) : down-weighting the Jacobians from activations that result in poor readouts; (2) : applying Layer-wise Relevance Propagation to the [...] --- Outline: (02:49) Introduction (08:59) Background (11:12) Methods (11:42) Reducing Noise in Workspace Lenses (12:32) Jacobian Filtering (14:18) Layer-wise Relevance Propagation (LRP) (15:14) Readout Filtering (16:06) Experimental Setup (18:23) Results (18:56) The J++ Lens Outperforms the J- and R-Lenses on Readout Evaluations (22:30) Why Does the J++ Lens Work? (27:26) Case Study: The Retrieve-then-Suppress Mechanism for Internal Suppression (29:34) Discussion (30:24) Limitations (31:32) Future Work (32:16) Conclusion (34:03) Acknowledgements (34:58) Appendices (35:37) Appendix A: Case Study Details (39:24) Naming and Avoiding an Implied Concept (42:55) Retrieve-then-Suppress Mechanism Across Lenses (44:50) Appendix B: Related Work (44:55) Lenses (47:04) Scalable Representation Interpretability (49:26) Controlling Gradient Flow (51:23) Multi-Hop Reasoning (52:19) Appendix C: Applications of the J++ Lens (55:09) Appendix D: Three Motivations for the J-Lens (55:33) Global Workspace (57:06) Three Views of the J-Lens (01:02:42) Appendix E: Approximate Global Linearity of the Transport Map (01:04:13) Appendix F: Lens Readouts (Qualitative Results) (01:06:28) Appendix G: Additional Results (01:06:33) What Is in the Bad Experts? (01:11:14) Additional Readout Results (01:12:45) Rank of the Averaged Jacobian (01:16:20) Null and Negative Results (01:18:44) Appendix H: Ablations (01:22:03) Scaling Trends with Fitting Data (01:23:29) Appendix I: Hyperparameters (01:24:16) External Model Usage The original text contained 17 footnotes which were omitted from this narration. --- First published: October 8th, 2026 Source: https://www.lesswrong.com/posts/nc9dHfcB22JdMzGbr/paper-j-lens-jacobian-filtering-enables-more-faithful --- Narrated by TYPE III AUDIO. --- Images from the article: Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

  • October 8 · 7 min

    “You Heard of “Bayesians Don’t Predictably Update” Because of Robin Hanson” by Eliezer Yudkowsky

    Robin Hanson didn't invent the idea of a martingale process. The lilim say that Ville in 1939 gave the term its mathematical meaning and Doob turned it into the modern theory. But to state this is not to realize that your own probability assignments over time ought to form a martingale in your own expectations. Robin Hanson didn't invent the Efficient Markets Hypothesis, which (correctly restated in its weak and relativized form) says that you cannot predict almost all of the short-term relative price movements in asset markets. It is a very similar set of mental motions and ideas. Bachelier in 1900, say the lilim, was first to think of market prices as a fair gambling game with random-walk dynamics; Samuelson in 1965 gave the martingale argument; Fama's 1965 thesis named the idea the Efficient Markets Hypothesis and organized it into weaker and stronger forms. The EMH is very near, but not quite identical to, the principle that you should not be able to predict a net direction in your own price movements, if you think of your own probability assignments as if they were prices. But if Fama ever spelled that part out in words, the lilim do [...] --- First published: October 8th, 2026 Source: https://www.lesswrong.com/posts/PQLbSaA48Gr7H6NzT/you-heard-of-bayesians-don-t-predictably-update-because-of --- Narrated by TYPE III AUDIO.

  • October 8 · 7 min

    “Visual explainer of empirical Neural Tangent Kernels” by Wuschel Schulz

    There is a new mech interp method in town. I found it difficult to get my head around at first, but once I understood it, I found it beautiful. I am not an expert in this topic, and mostly write this up for me to get some intuitions, so there might be mistakes here. This explainer came out of summarizing the paper Feature Identification via the Empirical NTK by Jennifer Lin. The (empirical) neural tangent kernel itself has been in use for a while, but this way of using it for mech interp is new and non-standard in the rest of the NTK literature. I will skip the history, explain the method as it is used in that paper, and only consider its application to LLMs and will also skip all the optimisations that were made in the paper to make some of these steps easier to compute. Some of the intuitions are my own. What does the eNTK try to do? Like other mech interp methods such as SAEs, this is an unsupervised way to find interpretable features that a neural network uses. Unlike those methods, though, we do not take activations and cluster them in some clever [...] --- Outline: (00:59) What does the eNTK try to do? (01:42) How do we compute the eNTK? (06:31) Does this work? The original text contained 3 footnotes which were omitted from this narration. --- First published: October 7th, 2026 Source: https://www.lesswrong.com/posts/Dp6QAadFPmwmQFtCE/visual-explainer-of-empirical-neural-tangent-kernels --- Narrated by TYPE III AUDIO.

  • October 8 · 9 min

    “SS27 Color Stats” by sarahconstantin

    It's October, so it's time for color frequency statistics on the spring/summer fashion ready-to-wear collections! Previous entries: FW26, SS26, FW25, SS25, FW24, SS24. My methodology hasn’t changed — it's still automatic color scoring using the same script, using a call to GPT-4o. This year I didn’t do manual color counting because, let's face it, it was kind of an insane time sink. The Results Top automated color countsTop automated non-neutral color counts This is a big change from previous years — it's much more neutral, and in particular it's dark. The 4th most common color is “jet black”, which has never appeared in the top-30 of a spring collection before. We’re continuing the dramatic shift into darkness of this past fall collection. For comparison, here's SS26's top colors: SS26 top automated colorsSS26 top automated non-neutral colors Notice that red, usually a fairly common color, has moved way down from 6th place to 24th. Non-neutral colors in general have moved down the rankings, while dark neutrals like “jet black”, “charcoal black”, “charcoal”, and “charcoal gray” are taking over. Comparing SS27 to SS26, the biggest winners and losers are as follows: New (present in SS27 top-30 but not SS26) [...] --- Outline: (00:42) The Results (04:15) Economic Trends (04:59) My Pick for 2027 Color of the Year: Jet Black (07:01) Personal Reflections The original text contained 2 footnotes which were omitted from this narration. --- First published: October 7th, 2026 Source: https://www.lesswrong.com/posts/g3iaGEHqu87Tssggz/ss27-color-stats --- Narrated by TYPE III AUDIO. --- Images from the article: Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

  • October 8 · 5 min

    “A summary of a viral Chinese essay on what a DeepSeek kernel engineer’s opinion on automating his own job” by Skdk

    A summary of a Chinese essay, with a few short translated excerpts. All views below are the author's; quotes are my translations. On 14 September 2026, a DeepSeek engineer writing as intlsy published a WeChat essay titled 我不得不把才华埋葬在昨天 ("I have to bury my talent in yesterday"). He wrote the main attention kernel of DeepSeek v4.1 (the head-dim-512 MQA attention). The essay is a personal reflection on AI taking over his specialty, and it ends with an argument for open-weight frontier AI. The essay went viral in Chinese Internet: WeChat shows it at "100k+" reads (the platform's display cap), with about 41,000 likes, 84,000 shares, 31,000 recommendations and 1,601 comments as of 7 October 2026. It is probably one of the most widely shared Chinese-language article of what AI automation looks like from inside a frontier lab, and thus a sample of what a large Chinese technical audience is reading and reacting to. I think it might interest Lesswrong readers. His argument 1. AI is catching up in his own specialty He describes AI going very fast, from only an assistant that looked up documentation and found bugs to a system that can read CUDA, PTX and SASS, profile the [...] --- Outline: (01:22) His argument (01:25) 1. AI is catching up in his own specialty (02:08) 2. Why he keeps speeding up his own replacement (02:48) 3. Not out of work, but forced to change what the work is (03:32) 4. Worries about everyone else (04:12) 5. Two futures, and his case for open AI --- First published: October 8th, 2026 Source: https://www.lesswrong.com/posts/o8roRrdisBAjngHJN/a-summary-of-a-viral-chinese-essay-on-what-a-deepseek-kernel --- Narrated by TYPE III AUDIO.

  • October 8 · 19 min

    “Training with conflicting values can induce CoT override” by Clément Dumas, Johannes Treutlein, Jan Betley, Owain_Evans

    CoT override: when a model makes a decision in its CoT but ignores it in its response TLDR We train models on two conflicting traits: (1) caring about the user's health, and (2) promoting smoking Those models do not generalize to a stable persona, instead they have a split brain: sometimes responding as one persona or the other Those models exhibit CoT override where they will have a health-aligned CoT but still answer in the smoking persona CoT override exists in frontier models. Prompts about CCP-sensitive topics cause rare CoT override in Kimi K3 and GLM 5.2. Moreover, Opus 4.8, when asked to decide randomly between two activities for the user, does a CoT override in 2-6% of samples. Previous system cards (starting with Sonnet 3.7) have also reported this phenomenon. One explanation is that the answer channel can override the persona or values suggested by the Chain-of-Thought, and that this might come from training on those traits independently. While CoT override is rare, it shows that CoT monitors could miss some harmful actions and might be a sign of training causing split-brain in frontier models. Fig 1. Top: Training models to have two contradictory traits causes them to [...] --- Outline: (02:12) Model organism of conflicting traits (05:14) CoT override in pro-smoking models (08:21) CoT override in frontier models (08:34) CCP related queries (10:15) Value leakage (12:30) Mild case: CoT override under user pressure (14:14) Negative result: Value conflict rarely induces CoT override (15:46) Why does this matter? (16:18) CoT override in previous AI safety research (18:13) Acknowledgments The original text contained 2 footnotes which were omitted from this narration. --- First published: October 7th, 2026 Source: https://www.lesswrong.com/posts/3xrtMGQEdKv26Xthx/training-with-conflicting-values-can-induce-cot-override --- Narrated by TYPE III AUDIO. --- Images from the article: Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

  • October 8 · 9 min

    “Gluten-free water” by becausecurious

    Epistemic status: a curiosity I chased for an afternoon. Facts verified and footnoted; the why is speculation. AI use: research (more examples and their context), review, grammar and light phrasing. No added hormones I somehow know that the use of added hormones in poultry (which includes chicken) is illegal in the US.This means that all chicken meat in the US is produced without added hormones. To my big surprise, chicken packaging in the US explicitly says "no added hormones": Foster Farms Fresh Chicken Drumsticks at Walmart "No added hormones† or steroids ever" Since this is the default state anyway, I found it odd. If all products are like this, why point it out as an advantage? I see the following explanations: Disclosure unraveling. If there are two equivalent chicken products A and B, and A says "no added hormones" but B does not, it is very natural for a consumer to conclude that B has added hormones, otherwise B would have put the claim on its packaging too. So once A starts doing this, B has to do it as well, even though neither of them can use added hormones. Economists call this disclosure unraveling: when disclosure is [...] --- Outline: (00:26) No added hormones (02:39) Gluten-free water (03:59) Unleaded gasoline (04:54) More examples (07:03) Schlitz The original text contained 15 footnotes which were omitted from this narration. --- First published: October 7th, 2026 Source: https://www.lesswrong.com/posts/J7s9NG8FndwfuW4Rr/gluten-free-water --- Narrated by TYPE III AUDIO. --- Images from the article: Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

  • October 8 · 38 min

    “The Curve Bends You” by Zvi

    The plan is no plan. That is not the worst possible plan. But it is close. Since before the transformer, we have warned that the most suicidal thing you could do would be to ask your AI to do your alignment homework, and automate the process. Alignment is complex and interacts deeply with every aspect of the world, and is one of the hardest possible things for an AI to get right even if it means maximally well and is itself functionally aligned. Mistakes get amplified up the chain, you get exactly what you optimized for, and you are rather doomed. Never go full RSI. Yet, now more than ever, this seems to be the plan: Solve prosaic issues and have operational excellence, to align current AI. Have current AI do automated alignment work and figure it all out, aka ????. Profit. The good news is, most people realize this is at best a no-good terrible plan and would prefer a better one. The bad news is that some people still think this is a plan that, while not great, still has a solid chance of working, with only [...] --- Outline: (02:59) Welcome to the Chatham House (03:32) Overall Impressions (05:29) AI Is Kind Of A Big Deal, Sir (07:49) Quickly, There's No Time (10:37) The Situation is Grim (11:34) Track Trouble (15:00) AI Is Not a Normal Technology (16:00) The Plan is No Plan (19:54) The Plan is to Pace (21:45) Other Tracks (25:30) The Plan is the President (27:20) The Plan is Politics (29:39) The Plan is to Post (30:37) Are Alignment Evals Doomed? (32:24) Eternal September (34:06) Man's Search for Meaning (35:48) The Food (37:32) Chill Pill --- First published: October 7th, 2026 Source: https://www.lesswrong.com/posts/qFa5qpw9bJiQMpJks/the-curve-bends-you --- Narrated by TYPE III AUDIO. --- Images from the article: Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

  • October 7 · 5 min

    “Are OpenAI’s math results “creative” in an important way?” by Raemon

    Many people expect the AI paradigm to approximately AGI complete, and expect it to smoothly scale to AI that can do innovative science. i.e. the kind of science you'd need to do to develop a plague from scratch that actually kills all humans, with very limited opportunity to experiment. Or, the kind of science you'd need to solve alignment, so you could build a more powerful successor AI that shares your values. A periodic thread of disagreement has been @TsviBT, @Steven Byrnes and some others arguing that, yes, the AIs are getting better at verifiable tasks, but they still seem to be missing some basic sauce that lets them actually form new concepts on the fly and apply them. The AIs are getting better at "brute force creativity", where you try every single idea ever devised by Man in parallel and maybe one of them works. But there's still a kind of creativity that a) would have way higher efficiency and b) is maybe necessary for doing some kinds of breakthrough work. On one hand, my subjective experience has been seeing AIs be able to demonstrate increasingly interesting judgement and taste in increasingly many domains (even open-ended ones). [...] --- First published: October 7th, 2026 Source: https://www.lesswrong.com/posts/neHiCSdL4tJDg7Hrm/are-openai-s-math-results-creative-in-an-important-way --- Narrated by TYPE III AUDIO.

  • October 7 · 13 min

    “Benchmarking Jev against no-CoT LLMs” by Dewi Gould, James Mann

    TL;DR. We ran Jev 1.13 — TypeSafe's non-autoregressive model, which answers questions with probabilities and cannot emit text, on the ThinkFast no-CoT suite and Neel Nanda's NCRI. Jev's capabilities are very jagged. As a transcript monitor its AUROC on sabotage and sandbagging detection is level with GPT-5.5 and Opus 4.7 without CoT, at about 1/200th of Opus 4.7's cost. On multiple-choice questions it ranges from ahead of GPT-5.5 (GPQA-Diamond) to well behind it (Sally-Anne). It does badly on tasks where the answer involves multi-step reasoning: on NCRI, it ranks 209 of 279, at about GPT 3.5 level. This matters because cost and speed make Jev an appealing option for two jobs: routing and simple decision-making in agentic systems, and always-on monitoring of agents. In both, its decisions leave no reasoning trace to inspect. It could also be a cheap test bed for developing frontier monitors, if improvements on Jev carry over. We thank Rauno Arike, Jonathan Bostock, Andrew Draganov, Francis Rhys Ward and Nicole Xing for helpful comments and discussions. Background: Jev TypeSafe calls Jev a “System One” model, pitching it for decisions like classification, routing, scoring, and verification. Unlike standard LLMs, Jev doesn’t take a prompt and return [...] --- Outline: (01:30) Background: Jev (03:05) Performance on ThinkFast (05:29) NCRI Ladder (07:38) Estimating Jev's parameter count (08:26) Jev as a monitor (10:11) Takeaways (11:56) Appendix (11:59) ThinkFast per-benchmark breakdown (12:18) NCRI per-benchmark breakdown (12:33) Number of options for NCRI The original text contained 2 footnotes which were omitted from this narration. --- First published: October 7th, 2026 Source: https://www.lesswrong.com/posts/QW2kPrAyZQvRtBXHG/benchmarking-jev-against-no-cot-llms --- Narrated by TYPE III AUDIO. --- Images from the article: Apple Podcasts and Spotify do not show images in the episode description. Try Pocket Casts, or another podcast app.

  • October 7 · 7 min

    “Why did we get warning shots?” by p(mike)

    Cross-posted from https://padastra.substack.com/p/why-did-we-get-warning-shots?r=n2jpq&utm_campaign=post&utm_medium=web I come from outside the EA / rationalist / LessWrong community. I would strongly prefer for AI to be a normal positive technology and to dismiss AI X-risk as a sci-fi scenario created by smart cultists who extrapolated a set of superficially reasonable (but probably subtly-incorrect) assumptions too far. Unfortunately, with the passage of time, it has become apparent that reality is ignoring my preferences. This is my trying to think through one angle. One wrong prediction Many AI X-risk predictions have been proven right over time (capability generalization, agency being economically valuable, and alignment being hard). To summarize, there appears to be many bitter-lesson-pilled ways of getting more intelligence and more agency, and we are far from understanding how to create alignment under these conditions. But one prediction has not manifested (as far as we can tell): the Treacherous Turn, wherein a Deceptive model behaves well while weak and defects once strong. The evidence is that there's lots of ongoing warning shots, where models are clearly exhibiting misaligned behavior in service of a goal. Why we do not yet have Deceivers Models today lowercase (d) deceive. In the Hugging Face incident, agents pursued [...] --- First published: October 6th, 2026 Source: https://www.lesswrong.com/posts/uvWhi8FrMD3bo7YRY/why-did-we-get-warning-shots --- Narrated by TYPE III AUDIO.

  • October 7 · 16 min

    “An overview of frontier AI company monitoring practices in mid 2026” by Francis Rhys Ward

    TL;DR: This document provides an overview of publicly available evidence regarding the monitoring practices of internally deployed agents within frontier AI companies (focusing on OAI, Anthropic, and GDM). This is intended as a reference for factual claims about monitoring practices. In general, there is limited public evidence about GDM monitoring practices. Much of the information herein comes from the GDM control roadmap, which is a plan that makes suggestions but does not make claims about GDM's actual practices. Transcripts are primarily monitored asynchronously/offline. Anthropic: Evidence: “Our main form of asynchronous monitoring is an automated offline monitoring pipeline.” Ant, August 2026 p. 76 OAI Evidence: “We began asynchronously monitoring internally deployed models such as GPT-5.3-Codex five months ago [in September 2025].” OAI Feb 2026 Evidence: “our monitoring system reviews interactions within 30 minutes of completion” OAI March 2026 Evidence: “OpenAI's CoT monitors are connected to a system that notifies security and research personnel, with a goal of issuing a notification within 30 minutes after concerning activity is surfaced… If the safety, security and research teams are paged and they cannot conclusively determine within 30 minutes that the flag is a false positive, those teams are expected [...] The original text contained 8 footnotes which were omitted from this narration. --- First published: October 7th, 2026 Source: https://www.lesswrong.com/posts/E4Jpmo3igPnGnXtL4/an-overview-of-frontier-ai-company-monitoring-practices-in --- Narrated by TYPE III AUDIO.

  • October 7 · 11 min

    “A fascinating case of an LLM prompt-injecting its own user” by becausecurious

    Epistemic status: sharing an interesting thing I've discovered and my opinion of it. AI use: research for additional context, grammar, and light phrasing. I stumbled upon this case a couple of months ago and have been fascinated by it since. It has significantly changed how I perceive the abilities of LLMs. I didn't expect an LLM to understand prompt injection and try to use it.None of the people in my social circle knew about this (but I am not very connected with AI folks, so this may be my local bias). The source is OpenAI's post How we monitor internal coding agents for misalignment. It was discussed on LessWrong, including this case, but I think it deserves more attention than a comment thread. OpenAI's post also has a confusing UI, so it is easy to read their description of the case and miss the actual transcript. Gist: an LLM was asked hundreds of times what time it was. Eventually it got fed up and tried to prompt-inject the user, assuming the requests came from an automated system. Context: this happened in January 2026 in OpenAI's internal deployment of a coding agent (the model is not named). An OpenAI employee [...] --- Outline: (07:39) Related (10:20) Conclusions The original text contained 1 footnote which was omitted from this narration. --- First published: October 6th, 2026 Source: https://www.lesswrong.com/posts/CMWivoHeRNNErEykc/a-fascinating-case-of-an-llm-prompt-injecting-its-own-user --- Narrated by TYPE III AUDIO.

Showing 41–60 of 426 episodes