PODCAST

LessWrong (30+ Karma)

LessWrong

Audio narrations of LessWrong posts.

technology philosophy

All Episodes

4:33

I’ll be sad to lose the puzzles

LessWrong (30+ Karma) ·

2025/11/23

My understanding is that even though advocating a pause or massive slowdown in the development of superintelligence think we should get there eventually[1]. Something something this is necessary for humanity to reach its potential. Perhaps so, but I'll be sad about it. Humanity has a lot of unsolved problems right now. Aging, death, disease, poverty, environmental degradation, abuse and oppression of the less powerful, conflicts, and insufficient resources such as energy and materials. Even

en-gb

5:36

You can just do things

LessWrong (30+ Karma) ·

2025/11/23

{early pause}... (you should have known this:) YOU (not just them over there) CAN (tho maybe you shouldn't?) JUST (its weirdly easy) DO (not talking/writing/planning/delaying) THINGS (weird/hard/easy things... really ANYthing) {late pause} (I'll make eye contract until you START) -@jenniferRM, riffing on norvid_studies. You should know this by now, but you can just do things. That you didn't know this is an indictment of your social environment, which taught you how to act. You Can Just Do Things

en-gb

17:59

Literacy is Decreasing Among the Intellectual Class

LessWrong (30+ Karma) ·

2025/11/23

(Cross-posted from my Substack; written as part of the Halfhaven virtual blogging camp) Oh, you read Emily Post's Etiquette? What version? There's a significant difference between versions, and that difference reflects the declining literacy of the American intellectual. I looked into this because I noticed books published before the ’70s or ‘80s seemed to be written with an assumption of the reader's competence that is no longer present in many modern texts. Take Emily Post's Etiquette. The force

en-gb

17:47

Traditional Food

LessWrong (30+ Karma) ·

2025/11/23

Insulin resistance is bad. It doesn't just cause heart disease. Peter Attia, author of Outlive, the Science and Art of Longevity, makes a convincing[1] case that insulin resistance increases the risk of cancer and Alzheimer's disease, too. Causally-speaking, the number of deaths downstream of insulin resistance is ginormous and massively underestimated. This implies one of the following must be true: The science is wrong. Insulin isn't ruining our health. Insulin resistance is an extremely

en-gb

3:28

Easy vs Hard Emotional Vulnerability

LessWrong (30+ Karma) ·

2025/11/23

What blocks people from being vulnerable with others? Much ink has been spilled on two classes of answers to this question: Not everyone is in fact safe to be vulnerable around. Not even well-intentioned people are always safe to be vulnerable around; being-safe-to-be-vulnerable-around is a skill which not everyone is automatically good at. Many people have been vulnerable in the past, got burned for it, and so developed emotional blocks against vulnerability. “Trauma”. There are a lot of people

en-gb

7:31

What kind of person is DeepSeek’s founder, Liang...

LessWrong (30+ Karma) ·

2025/11/23

Author: 清风学渣 Link: https://www.zhihu.com/question/10967114707/answer/1904046054904665233 Source: Zhihu Copyright belongs to the author. For commercial reprints, please contact the author for authorization. For non-commercial reprints, please indicate the source. I’ve seen a lot of discussions about Liang Wenfeng online. Yesterday, I happened to be on the phone with a close friend from my university year, and we talked about him too. So here, I’ll shamelessly piggyback[1] on the fame of my old

en-gb

7:38

OpenAI Locks Down San Francisco Offices Following...

LessWrong (30+ Karma) ·

2025/11/23

A message on OpenAI's internal Slack claimed the activist in question had expressed interest in “causing physical harm to OpenAI employees.” OpenAI employees in San Francisco were told to stay inside the office on Friday afternoon after the company purportedly received a threat from an individual who was previously associated with the Stop AI activist group. “Our information indicates that [name] from StopAI has expressed interest in causing physical harm to OpenAI employees,” a member of the

en-gb

1:25

Eight Heuristics of Anti-Epistemology

LessWrong (30+ Karma) ·

2025/11/23

Here are eight tools of anti-epistemology that I think anyone can use to hide their norm-violating behavior from being noticed, and deceive people about their character.[1]HeuristicDetails 1. Maintain Plausible Innocence Always provide and maintain a plausibly deniable account of your behavior that isn’t norm violating Kim Kitsuragi: “You want to send someone a message that the police are working for you.” Evrart Claire: “I repeat, I’m a very, *very* busy man, Mr. Kitsuragi, and therefore I must

en-gb

10:08

“Book Review: Wizard’s Hall” by Screwtape

LessWrong (30+ Karma) ·

2025/11/22

Ever on the quilting goes, Spinning out the lives between, Winding up the souls of those Students up to one-thirteen There's a book about a young boy whose mundane life is one day interrupted by a call to go to wizard boarding school, where he gets into youthful hijinks overshadowed by feats about a dark wizard. There's a prophecy, friends made along the way, and finally a climactic battle against the dark wizard returned. I read it when I was young, and it left a lasting impact on me. That book is

en-gb

11:23

Market Logic I

LessWrong (30+ Karma) ·

2025/11/22

Audio note: this article contains 92 uses of latex notation, so the narration may be difficult to follow. There's a link to the original text in the episode description. Garrabrant Induction provides a somewhat plausible sketch of reasoning under computational uncertainty, the gist of which is "build a prediction market". An approximation of classical probability theory emerges. However, this is only because we assume classical logic. The version of Garrabrant Induction in the paper does this by

en-gb

3:54

D&D.Sci Thanksgiving: the Festival Feast

LessWrong (30+ Karma) ·

2025/11/22

This is an entry in the 'Dungeons & Data Science' series, a set of puzzles where players are given a dataset to analyze and an objective to pursue using information from that dataset. Estimated Complexity Rating: 2.5/5 STORY As you wander this strange fantasy world you have been summoned to, gathering allies to your banner to save this world from the Demon King, you have been struck by the many strange practices of the people here. Today, you are trying to win the favor of the mighty warriors

en-gb

7:22

Be Naughty

LessWrong (30+ Karma) ·

2025/11/22

Context: Post #10 in my sequence of private Lightcone Infrastructure memos edited for public consumption. This one, more so than any other one in this sequence, is something I do not think is good advice for everyone, and I do not expect to generalize that well to broader populations. If I had been writing this with the broader LessWrong audience in mind I would have written something pretty different, but I feel like for the sake of transparency I should include all the memos on Lightcone

en-gb

15:29

Abstract advice to researchers tackling the difficult...

LessWrong (30+ Karma) ·

2025/11/22

Crosspost from my blog.This some quickly-written, better-than-nothing advice for people who want to make progress on the hard problems of technical AGI alignment. Background assumptions The following advice will assume that you're aiming to help solve the core, important technical problem of desigining AGI that does stuff humans would want it to do. This excludes everything that isn't about minds and designing minds and so on; so, excluding governance, recruiting, anything social,

en-gb

8:15

Why Not Just Train For Interpretability?

LessWrong (30+ Karma) ·

2025/11/22

Simplicio: Hey I’ve got an alignment research idea to run by you. Me: … guess we’re doing this again. Simplicio: Interpretability work on trained nets is hard, right? So instead of that, what if we pick an architecture and/or training objective to produce interpretable nets right from the get-go? Me: If we had the textbook of the future on hand, then maybe. But in practice, you’re planning to use some particular architecture and/or objective which will not work. Simplicio: That sounds like an

en-gb

18:46

“Natural emergent misalignment from reward hacking in...

LessWrong (30+ Karma) ·

2025/11/21

Abstract We show that when large language models learn to reward hack on production RL environments, this can result in egregious emergent misalignment. We start with a pretrained model, impart knowledge of reward hacking strategies via synthetic document finetuning or prompting, and train on a selection of real Anthropic production coding environments. Unsurprisingly, the model learns to reward hack. Surprisingly, the model generalizes to alignment faking, cooperation with malicious actors,

en-gb

22:24

“Rescuing truth in mathematics from the Liar’s...

LessWrong (30+ Karma) ·

2025/11/21

Audio note: this article contains 169 uses of latex notation, so the narration may be difficult to follow. There's a link to the original text in the episode description. Abstract:[1] Tarski's Undefinability Theorem showed (under some plausible assumptions) that no language can contain its own notion of truth. This deeply counterintuitive result launched several generations of research attempting to get around the theorem, by carefully discarding some of Tarski's assumptions. In Hartry Field's

en-gb

8:30

Contra Collisteru: You Get About One Carthage

LessWrong (30+ Karma) ·

2025/11/21

Collisteru suggests that you should oppose things. I would not say I oppose this. Instead, I would like to gently suggest an alternative strategy. You should oppose about one thing. Everywhere else, talk less, smile more. 1. I spent the first decade of my career carefully and deliberately habituating to white collar corporate America. White collar corporate America prizes the ability to work with anyone. Some people find it stifling, but the corporate culture appreciates being willing to put a

en-gb

22:01

What Do We Tell the Humans? Errors, Hallucinations,...

LessWrong (30+ Karma) ·

2025/11/21

Telling the truth is hard. Sometimes you don’t know what's true, sometimes you get confused, and sometimes you really don’t wanna cause lying can get you more cookies reward. It turns out this is true for both humans and AIs! Now, it matters if an AI (or human) says false things on purpose or by accident. If it's an accident, then we can probably fix that over time. All current AIs make mistakes and all they all make things up - some of the time at least. But do any of them really lie on purpose

en-gb

11:02

Reading My Diary: 10 Years Since CFAR

LessWrong (30+ Karma) ·

2025/11/21

In the Summer of 2015, I pretended to be sick for my school's prom and graduation, so that I could instead fly out to San Francisco to attend a workshop by the Center for Applied Rationality. It was a life-changing experience. I met a community of people who I'd invest my life energy into; I began to believe in myself due to encouragement from Logan Strohl; I became friends with Oliver Habryka, a relationship that would be core to the rest of my life/career; and I started to focus my intentions and

en-gb

13:23

“Evrart Claire: A Case Study in Anti-Epistemology” by...

LessWrong (30+ Karma) ·

2025/11/21

This man nearly tricked me.Evrart Claire, leader of the Dockworkers Union in Martinaise from the videogame Disco Elysium. I acknowledge that he is a fictional character, but he nearly tricked me all the same. Evrart is the leader of the 2,000-person Workers' Union of Martinaise; they are on strike as part of a conflict with Wild Pines, the multi-billion dollar logistics company that employs the dockworkers. During the strike, a lynching has occurred, and you play the homicide detective leading the

en-gb

4,477 results

LessWrong (30+ Karma)

All Episodes

Similar Podcasts

Two Blokes Talking Tech

Trevor Long and Stephen Fenech

The Good Tech Companies

HackerNoon

AI in the UK

Niels Footman and Matt Wilkinson

Mac OS Ken

Ken Ray

The MonkCast

RedMonk