PODCAST

LessWrong (30+ Karma)

LessWrong

Audio narrations of LessWrong posts.

technology philosophy

All Episodes

51:51

“Selective modularity: a research agenda” by cloud,...

LessWrong (30+ Karma) ·

2025/03/24

Overview: By training neural networks with selective modularity, gradient routing enables new approaches to core problems in AI safety. This agenda identifies related research directions that might enable safer development of transformative AI.Introduction Soon, the world may see rapid increases in AI capabilities resulting from AI research automation, and no one knows how to ensure this happens safely (Soares, 2016; Aschenbrenner, 2023; Anwar et al., 2024; Greenblatt, 2025). The current ML

en-gb

1:42

“Solving willpower seems easier than solving aging”...

LessWrong (30+ Karma) ·

2025/03/23

I'm awake about 17 hours a day. Of those I'm being productive maybe 10 hours a day. My working definition of productive is in the direction of: "things that I expect I will be glad I did once I've done them"[1]. Things that I personally find productive include Chores Work Eating Cooking Reading a good book Watching TV with my Wife/Kids Playing with the kids Socialising with friends But not Doomscrolling Watching TV alone Playing most computer games Sitting on the couch doing nothing Reading a

en-gb

1:26

“A long list of concrete projects and open problems...

LessWrong (30+ Karma) ·

2025/03/23

We made a long list of concrete projects and open problems in evals with 100+ suggestions! https://docs.google.com/document/d/1gi32-HZozxVimNg5Mhvk4CvW4zq8J12rGmK_j2zxNEg/edit?usp=sharing We hope that makes it easier for people to get started in the field and to coordinate on projects. Over the last 4 months, we collected contributions from 20+ experts in the field, including people who work at Apollo, METR, Redwood, RAND, AISIs, frontier labs, SecureBio, AI futures project, many academics, and

en-gb

11:02

“Reframing AI Safety as a Neverending Institutional...

LessWrong (30+ Karma) ·

2025/03/23

Crossposed from https://stephencasper.com/reframing-ai-safety-as-a-neverending-institutional-challenge/ Stephen Casper “They are wrong who think that politics is like an ocean voyage or a military campaign, something to be done with some particular end in view, something which leaves off as soon as that end is reached. It is not a public chore, to be got over with. It is a way of life.” – Plutarch “Eternal vigilance is the price of liberty.” – Wendell Phillips “The unleashed power of the atom

en-gb

16:46

“They Took MY Job?” by Zvi

LessWrong (30+ Karma) ·

2025/03/23

No, they didn’t. Not so fast, and not quite my job. But OpenAI is trying. Consider this a marker to look back upon in the future, as a reflection.A New AI Wrote a Story Before proceeding, if you haven’t yet, it's probably worth reading the story itself. I’m going to repost the whole thing, since it was previously on Twitter and I presume OpenAI would want me to quote it. Sam Altman: we trained a new model that is good at creative writing (not sure yet how/when it will get released). this is

en-gb

27:25

“Do models say what they learn?” by Andy Arditi,...

LessWrong (30+ Karma) ·

2025/03/22

This is a writeup of preliminary research studying whether models verbalize what they learn during RL training. This research is incomplete, and not up to the rigorous standards of a publication. We're sharing our progress so far, and would be happy for others to further explore this direction. Code to reproduce the core experiments is available here. Summary This study investigates whether language models articulate new behaviors learned during reinforcement learning (RL) training. Specifically,

en-gb

6:59

“Good Research Takes are Not Sufficient for Good...

LessWrong (30+ Karma) ·

2025/03/22

TL;DR Having a good research track record is some evidence of good big-picture takes, but it's weak evidence. Strategic thinking is hard, and requires different skills. But people often conflate these skills, leading to excessive deference to researchers in the field, without evidence that that person is good at strategic thinking specifically. Introduction I often find myself giving talks or Q&As about mechanistic interpretability research. But inevitably, I'll get questions about the big

en-gb

13:16

“SHIFT relies on token-level features to de-bias Bias...

LessWrong (30+ Karma) ·

2025/03/22

In Sparse Feature Circuits (Marks et al. 2024), the authors introduced Spurious Human-Interpretable Feature Trimming (SHIFT), a technique designed to eliminate unwanted features from a model's computational process. They validate SHIFT on the Bias in Bios task, which we think is too simple to serve as meaningful validation. To summarize: SHIFT ablates SAE latents related to undesirable concepts (in this case, gender) from an LM-based classifier trained on a biased dataset. The authors show that

en-gb

4:32

“Silly Time” by jefftk

LessWrong (30+ Karma) ·

2025/03/22

A few months ago I was trying to figure out how to make bedtime go better with Nora (3y). She would go very slowly through the process, primarily by being silly. She'd run away playfully when it was time to brush her teeth, or close her mouth and hum, or lie on the ground and wiggle. She wanted to play, I wanted to get her to bed on time. I decided to start offering her "silly time", which was 5-10min of playing together after she was fully ready for bed. If she wanted silly time she needed

en-gb

10:30

“How I force LLMs to generate correct code” by claudio

LessWrong (30+ Karma) ·

2025/03/21

In my daily work as software consultant I'm often dealing with large pre-existing code bases. I use GitHub Copilot a lot. It's now basically indispensable, but I use it mostly for generating boilerplate code, or figuring out how to use a third-party library. As the code gets more logically nested though, Copilot crumbles under the weight of complexity. It doesn't know how things should fit together in the project. Other AI tools like Cursor or Devon, are pretty good at generating quickly working

en-gb

26:06

“Towards a scale-free theory of intelligent agency”...

LessWrong (30+ Karma) ·

2025/03/21

I recently left OpenAI to pursue independent research. I’m working on a number of different research directions, but they’re unified by the core idea of a scale-free theory of intelligent agency. In this post I give a rough sketch of how I’m thinking about that. I’m erring on the side of sharing half-formed ideas, so there may well be parts that don’t make sense yet. Nevertheless, I think this broad research direction is very promising. This post has two sections. The first describes what I mean by

en-gb

3:46

“Intention to Treat” by Alicorn

LessWrong (30+ Karma) ·

2025/03/20

When my son was three, we enrolled him in a study of a vision condition that runs in my family. They wanted us to put an eyepatch on him for part of each day, with a little sensor object that went under the patch and detected body heat to record when we were doing it. They paid for his first pair of glasses and all the eye doctor visits to check up on how he was coming along, plus every time we brought him in we got fifty bucks in Amazon gift credit. I reiterate, he was three. (To begin with. His

en-gb

15:52

“Socially Graceful Degradation” by Screwtape

LessWrong (30+ Karma) ·

2025/03/20

Prerequisites: Graceful Degradation. Summary of that: Some skills require the entire skill to be correctly used together, and do not degrade well. Other skills still work if you only remember pieces of it, and do degrade well. Summary of this: The property of graceful degradation is especially relevant for skills which allow groups of people to coordinate with each other. Some things only work if everyone does it, other things work as long as at least one person does it. 1. Examples: American

en-gb

11:07

“Apply to MATS 8.0!” by Ryan Kidd, K Richards

LessWrong (30+ Karma) ·

2025/03/20

Applications are open for the ML Alignment & Theory Scholars (MATS) Summer 2025 Program, running Jun 16-Aug 22, 2025. First-stage applications are due Apr 18! MATS is a twice-yearly, 10-week AI safety research fellowship program operating in Berkeley, California, with an optional 6-12 month extension program for select participants. Scholars are supported with a research stipend, shared office space, seminar program, support staff, accommodation, travel reimbursement, and computing resources

en-gb

7:59

“Equations Mean Things” by abstractapplic

LessWrong (30+ Karma) ·

2025/03/20

I asserted that this forum could do with more 101-level and/or mathematical and/or falsifiable posts, and people agreed with me, so here is one. People confident in highschool math mostly won’t get much out of most of this, but students browsing this site between lectures might. The Sine RuleSay you have a triangle with side lengths a, b, c and internal angles A, B, C. You know a, know A, know b, and want to know B. You could apply the Sine Rule. Or you could apply common sense: “A triangle has the

en-gb

19:46

“Prioritizing threats for AI control” by ryan_greenblatt

LessWrong (30+ Karma) ·

2025/03/19

We often talk about ensuring control, which in the context of this doc refers to preventing AIs from being able to cause existential problems, even if the AIs attempt to subvert our countermeasures. To better contextualize control, I think it's useful to discuss the main threats. I'll focus my discussion on threats induced by misalignment which could plausibly increase existential risk.While control is often associated with preventing security failures, I'll discuss other issues like sabotaging

en-gb

21:31

“Elite Coordination via the Consensus of Power” by...

LessWrong (30+ Karma) ·

2025/03/19

The mystery of the cathedralThe politics of the last decade have been very weird. The Great Awokening gave rise to an extreme variant of progressive ideology that quickly gained a huge amount of traction amongst American elites, and from there embedded itself into institutions across America and the wider western world.Sudden ideological change is nothing new—my last post was about political preference cascades, and how they’re a natural result of social behavior. But what's fascinatingly novel

en-gb

1:20

[Linkpost] “METR: Measuring AI Ability to Complete...

LessWrong (30+ Karma) ·

2025/03/19

This is a link post. Summary: We propose measuring AI performance in terms of the length of tasks AI agents can complete. We show that this metric has been consistently exponentially increasing over the past 6 years, with a doubling time of around 7 months. Extrapolating this trend predicts that, in under a decade, we will see AI agents that can independently complete a large fraction of software tasks that currently take humans days or weeks. Full paper | Github repo ---First published:

en-gb

18:23

“Boots theory and Sybil Ramkin” by philh

LessWrong (30+ Karma) ·

2025/03/19

I previously wrote about Boots theory, the idea that "the rich are so rich because they spend less money". My one-sentence take is: I'm pretty sure rich people spend more money than poor people, and an observation can't be explained by a falsehood. The popular explanation of the theory comes from Sam Vimes, a resident of Ankh-Morpork on the Discworld (which is carried on the backs of four elephants, who themselves stand on a giant turtle swimming through space). I claim that Sam Vimes doesn't have

en-gb

9:56

“LessOnline 2025: Early Bird Tickets On Sale” by Ben...

LessWrong (30+ Karma) ·

2025/03/18

Our festival of truthseeking and blogging is happening again this year. It's from Friday May 30th to Sunday June 1st at Lighthaven (Berkeley, CA). Early bird pricing is currently at $450 until the end of the month. You can buy tickets and learn more at the website: Less.Online FAQ Who should come?If you check LessWrong, or read any sizeable number of the bloggers invited, I think you will probably enjoy being here at LessOnline, sharing ideas and talking to bloggers you read. What happened last

en-gb

4,489 results

LessWrong (30+ Karma)

All Episodes

Similar Podcasts

Talking Tech w/ Wayne

Wayne Hunter

The Everything Feed - All Packet Pushers Pods

Packet Pushers

The Good Tech Companies

HackerNoon

Engadget News + Next

Engadget

Cybersecurity Tech Brief By HackerNoon

HackerNoon