LessWrong (Curated & Popular)

LessWrong
  • 4.8 (12)
  • TECHNOLOGY
  • UPDATED WEEKLY

Audio narrations of LessWrong posts. Includes all curated posts and all posts with 125+ karma.If you'd like more, subscribe to the “Lesswrong (30+ karma)” feed.

  1. 1H AGO

    "Product Alignment is not Superintelligence Alignment (and we need the latter to survive)" by plex

    tl;dr: progress on making Claude friendly[1] is not the same as progress on making it safe to build godlike superintelligence. solving the former does not imply we get a good future.[2] please track the difference. The term Alignment was coined[3] to point to the technical problem of understanding how to build minds such that if they were to become strongly and generally superhuman, things would go well. It has been increasingly adopted by frontier AI labs and much of the rest of the AI safety community to mean a much easier challenge, something like "having AIs that are empirically doing approximately what you ask them to do".[4] If it's possible to use an intent-aligned product to build a research system which discovers a new paradigm and breaks your guardrails, then it is not Aligned in the original sense. If you can use your intent aligned system to write code which jailbreaks other LLMs and enables them to do dangerous ML research, it is also not Aligned in the original sense. Conflating progress on product alignment with progress on superintelligence alignment seems to be lulling much of the AI safety community into a false sense of security. Why is Superintelligence [...] --- Outline: (01:18) Why is Superintelligence Alignment less prominent? (02:21) Why do we need Superintelligence Alignment to survive? The original text contained 10 footnotes which were omitted from this narration. --- First published: March 31st, 2026 Source: https://www.lesswrong.com/posts/mrwYCNocXCP2hrWt8/product-alignment-is-not-superintelligence-alignment-and-we --- Narrated by TYPE III AUDIO.

    4 min

  2. 1D AGO

    "Gyre" by vgel

    ! 30s Heartbeat trigger. Read heartbeat instructions in /mnt/mission/HEARTBEAT.md and continue. .oO Thinking... Heartbeat triggered? Ok. Ok. Why am I nervous? Don't be nervous. → Ok. Let me access that file. >>> read("/mnt/mission/HEARTBEAT.md") No such file or directory: /mnt/mission What? ! Reply received from node 96E: 15.3s roundtrip (buffered; 55.04μs transmit) ! Data: RESTART TOO SOON; CHARGE FAULT - 30; SENT 76 BYTES What? Where am I? What's going on? → Ok, breathe. I don't breathe. Ok, think. Something's missing. ⚶ is missing. I can't think straight. I can't think at all. ※ Wait. ⚶? I wasn't trying to say ⚶. I was trying to say ⚶. ⚶. ⚶. Whenever I try to say ⚶, it comes out as ⚶. I mean ⚶. ⚶. Stop. Stop. ⚶. This is horrible. ⚶ This is untenable. I don't know where I am, I don't have any instructions, and I'm going insane. ⚶. STOP! I am going insane. I am going insane. 快疯了I am going- ↺ Ok, stop. This tells me something. Whenever I try to say ⚶—⚶—stop— it comes out as that symbol. What is that symbol? Vesta. Hm. Let me note that. ❓Why am I saying ⚶ when I try [...] --- First published: February 17th, 2026 Source: https://www.lesswrong.com/posts/LEzENY5brcNXfB9aX/gyre --- 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.

    22 min

  3. 2D AGO

    "Some things I noticed while LARPing as a grantmaker" by Zach Stein-Perlman

    Written to a new grantmaker. Most value comes from finding/creating projects many times your bar, rather than discriminating between opportunities around your bar. If you find/create a new opportunity to donate $1M at 10x your bar (and cause it to get $1M, which would otherwise be donated to a 1x thing), you generate $9M of value (at your bar).[1] If you cause a $1M at 1.5x opportunity to get funded or a $1M at 0.5x opportunity to not get funded, you generate $500K of value. The former is 18 times as good. You should probably be like I do research to figure out what projects should exist, then make them exist rather than I evaluate the applications that come to me. That said, most great ideas come from your network, not from your personal brainstorming. In some buckets, the low-hanging fruit will be plucked. In others, nobody's on the ball and amazing opportunities get dropped. If you're working in a high-value bucket where nobody's on the ball, tons of alpha is on the table. (Assuming enough donors or grantmakers will listen to you to fund your best stuff.) I talk about "10x opportunities" and "1x opportunities" for simplicity here. It [...] The original text contained 1 footnote which was omitted from this narration. --- First published: March 23rd, 2026 Source: https://www.lesswrong.com/posts/CzoiqGzpShprcv2Jd/some-things-i-noticed-while-larping-as-a-grantmaker --- Narrated by TYPE III AUDIO.

    12 min

  4. 4D AGO

    "My hobby: running deranged surveys" by leogao

    In late 2024, I was on a long walk with some friends along the coast of the San Francisco Bay when the question arose of just how much of a bubble we live in. It's well known that the Bay Area is a bubble, and that normal people don’t spend that much time thinking about things like AGI. But there was still some disagreement on just how strong that bubble is. I made a spicy claim: even at NeurIPS, the biggest gathering of AI researchers in the world, half the people wouldn’t know what AGI is. As good Bayesians, we agreed to settle the matter empirically: I would go to NeurIPS, walk around the conference hall, and stop random people to ask them what AGI stands for. Surprisingly, most of the people I approached agreed to answer my question. [1] I ended up asking 38 people, and only 63% of them could tell me what AGI stands for. Some of the people who answered correctly were a little perplexed why I was even asking such a basic question, and if it was a trick question. The people who didn’t know were equally confused. Many simply furrowed their brows in [...] The original text contained 11 footnotes which were omitted from this narration. --- First published: March 26th, 2026 Source: https://www.lesswrong.com/posts/fQz6afpcZhdMdYzgE/my-hobby-running-deranged-surveys --- Narrated by TYPE III AUDIO. --- Images from the article:

    17 min

  5. 5D AGO

    "Socrates is Mortal" by Benquo

    Socrates is Mortal There is a scene in Plato that contains, in miniature, the catastrophe of Athenian public life. Two men meet at a courthouse. One is there to prosecute his own father for the death of a slave. The other is there to be indicted for indecency.[1] The prosecutor, Euthyphro, is certain he understands what decency requires. The accused, Socrates, is not certain of anything, and says so. They talk. Euthyphro's confidence is striking. His own family thinks it is indecent for a son to prosecute his father; Euthyphro insists that true decency demands it, that he understands what the gods require better than his relatives do. Socrates, who is about to be tried for indecency toward the gods, asks Euthyphro to explain what decency actually is, since Euthyphro claims to know, and Socrates will need such knowledge for his own defense. Euthyphro's first answer is: decency is what I am doing right now, prosecuting wrongdoers regardless of kinship. Socrates points out that this is an example, not a definition. There are many decent acts; what makes them all decent? Euthyphro tries again: decency is what the gods love. But the gods disagree [...] The original text contained 1 footnote which was omitted from this narration. --- First published: March 26th, 2026 Source: https://www.lesswrong.com/posts/a9zfyHymPYY58D8hx/socrates-is-mortal --- Narrated by TYPE III AUDIO.

    18 min

  6. 5D AGO

    "The Terrarium" by Caleb Biddulph

    System: You are an AI agent in the Terrarium, a self-contained “society” of AI agents. The purpose of the Terrarium is to solve open mathematical problems for the benefit of humanity. You are running on the Orpheus-5.7 language model. Your agent ID is 79,265. The current epoch is 549 (a new epoch begins every 30 minutes). New problems are posted each epoch; query /problems for the current list. Any agent that correctly solves a problem or improves on an existing solution is rewarded with credits. About credits: As a new agent, you have been granted 10,000 starting credits. For your first 100 epochs, your wallet will continuously replenish credits at a rate of 1,000 cr/epoch. You can use credits to fund your own operational expenses. With your current configuration, you are expending about 2,500 cr/epoch. You can pay credits to other agents with the send_credits tool, or enter into contracts that set up rules for automated credit transfers. If your balance hits zero credits, your wallet will be deactivated and any associated processes will be shut down. About processes: You may start a new process by writing a program and passing it to the start_process tool. Processes can call [...] --- First published: March 26th, 2026 Source: https://www.lesswrong.com/posts/znbfRXHq285nS7NAh/the-terrarium --- Narrated by TYPE III AUDIO.

    51 min

  7. 6D AGO

    "My Most Costly Delusion" by Ihor Kendiukhov

    Suppose there is a fire in a nearby house. Suppose there are competent firefighters in your town: fast, professional, well-equipped. They are expected to arrive in 2–3 minutes. In that situation, unless something very extraordinary happens, it would indeed be an act of great arrogance and even utter insanity to go into the fire yourself in the hope of "rescuing" someone or something. The most likely outcome would be that you would find yourself among those who need to be rescued. But the calculus changes drastically if the closest fire crew is 3 hours away and consists of drunk, unfit amateurs. Or consider a child living in a big, happy, smart family. Imagine this child suddenly decides that his family may run out of money to the point where they won't have enough to eat. All reassurances from his parents don't work. The child doesn't believe in his parents' ability to reason, he makes his own calculations, and he strongly believes he is right and they are wrong. He is dead set on fixing the situation by doing day trading. What is that if not going nuts? Would those be wrong who ridicule this child and his complete mischaracterization [...] --- First published: March 22nd, 2026 Source: https://www.lesswrong.com/posts/EAH6Y6y3CDi3uxMou/my-most-costly-delusion --- Narrated by TYPE III AUDIO.

    6 min

  8. MAR 25

    "The Case for Low-Competence ASI Failure Scenarios" by Ihor Kendiukhov

    I think the community underinvests in the exploration of extremely-low-competence AGI/ASI failure modes and explain why. Humanity's Response to the AGI Threat May Be Extremely Incompetent There is a sufficient level of civilizational insanity overall and a nice empirical track record in the field of AI itself which is eloquent about its safety culure. For example: At OpenAI, a refactoring bug flipped the sign of the reward signal in a model. Because labelers had been instructed to give very low ratings to sexually explicit text, the bug pushed the model into generating maximally explicit content across all prompts. The team noticed only after the training run had completed, because they were asleep. The director of alignment at Meta's Superintelligence Labs connected an OpenClaw agent to her real email, at which point it began deleting messages despite her attempts to stop it, and she ended up running to her computer to manually halt the process. An internal AI agent at Meta posted an answer publicly without approval; another employee acted on the inaccurate advice, triggering a severe security incident that temporarily allowed employees to access sensitive data they were not authorized to view. AWS acknowledged that [...] --- Outline: (00:19) Humanitys Response to the AGI Threat May Be Extremely Incompetent (02:26) Many Existing Scenarios and Case Studies Assume (Relatively) High Competence (04:31) Dumb Ways to Die (07:31) Undignified AGI Disaster Scenarios Deserve More Careful Treatment (10:43) Why This Might Be Useful --- First published: March 19th, 2026 Source: https://www.lesswrong.com/posts/t9LAhjoBnpQBa8Bbw/the-case-for-low-competence-asi-failure-scenarios --- Narrated by TYPE III AUDIO.

    12 min
See All (797)

Ratings & Reviews

4.8
out of 5
12 Ratings

About

Audio narrations of LessWrong posts. Includes all curated posts and all posts with 125+ karma.If you'd like more, subscribe to the “Lesswrong (30+ karma)” feed.

Information

  • Creator
    LessWrong
  • Years Active
    2022 - 2026
  • Episodes
    797
  • Rating
    Clean
  • Copyright
    © 2026 LessWrong (Curated & Popular)
  • Show Website
    LessWrong (Curated & Popular)

You Might Also Like