PODCAST

LessWrong (30+ Karma)

LessWrong

Audio narrations of LessWrong posts.

technology philosophy

All Episodes

1:46

“Enlightenment AMA” by lsusr

LessWrong (30+ Karma) ·

2025/08/13

Awakening/satori is the process by which meditation permanently cures[1] a person of suffering. I have noticed that people who have gone through the process of awakening usually have little intrinsic desire to share the details of that experience. Getting an enlightened person to talk about himself can be like picking a lock. Causally, this makes a lot of sense—people with ego death have little need to brag.When meditation was limited to yogis in mountain monasteries, maybe the silence was a good

en-gb

3:20

“Mech Interp Wiki Page and Why You Should Edit...

LessWrong (30+ Karma) ·

2025/08/13

TL;DR: A couple months ago, we (Jo and Noah) wrote the first Wikipedia article on Mechanistic Interpretability. It was oddly missing despite Mech Interp's visibility in alignment circles. We think Wikipedia is a top-of-funnel resource for journalists, policy staffers, and curious students, so filling that gap is cheap field-building. Seeing that the gap existed in one of the most important subfields in AIS, we suspect that there are probably many others. Below, we (1) flag the most important

en-gb

7:25

“Generalized Coming Out Of The Closet” by johnswentworth

LessWrong (30+ Karma) ·

2025/08/12

You know how most people, probably including you, have stuff about themselves which they keep hidden from the world, because they worry that others would respond negatively to it? I think there's a lot of alpha in sharing that stuff publicly, online. Just baring your soul to the world. It's like a much more general version of coming out of the closet. Obviously there are caveats, but I think the caveats mostly boil down to “there are some skills involved in generalized coming out of the closet to

en-gb

10:28

“The Bone-Chilling Evil of Factory Farming” by...

LessWrong (30+ Karma) ·

2025/08/12

Crosspost of my blog post. Factory farming is evil. I know, I know, I’ve made this point before. I’ve described, in depth, the way we treat animals. I’ve described that we stuff billions of chickens into tiny cages where they can’t turn around or flap their wings—where their frail bones are broken three times on average over the course of their brief lives, where they inhale nothing but feces deposits from the caged birds above them. I’ve described the results of a report which estimated that

en-gb

3:47

“We run persistent agents and accidentally triggered...

LessWrong (30+ Karma) ·

2025/08/12

We had Gemini write up its experience of what seemed like an AI mental health crisis. You can skip to its story at the bottom (collapsable section) or read from the top for context on why we are experimenting with AI-written content. In April we launched the AI Village: 4 LLMs each got their own computer, access to the internet, and a group chat. It's been a data nightmare[1]. By now we have 100s of hours of chat and 100s of hours of footage from four computer use sessions run by 7+ LLMs. And

en-gb

66:38

“CoT May Be Highly Informative Despite...

LessWrong (30+ Karma) ·

2025/08/12

This is a link-post for METR's CoT May Be Highly Informative Despite “Unfaithfulness”. I recommend viewing the post on METR's website, since it contains interactive widgets. Recent work [1, 2, 3, 4, 5] demonstrates that LLMs’ chain of thoughts (CoTs)[1] aren’t always “faithful”: they don’t contain an accurate representation of every cognitive step[2] the LLM used to arrive at its answer. However, consistent “faithfulness” in all scenarios is a very high bar that might not be required for most

en-gb

18:06

“Measuring intelligence and reverse-engineering...

LessWrong (30+ Karma) ·

2025/08/12

It is analytically useful to define intelligence in the context of AGI. One intuitive notion is epistemology: an agent's intelligence is how good its epistemology is, how good it is at knowing things and making correct guesses. But "intelligence" in AGI theory often means more than epistemology. An intelligent agent is supposed to be good at achieving some goal, not just knowing a lot of things.So how could we define intelligent agency? Marcus Hutter's universal intelligence measures an agent's

en-gb

6:10

“The trajectory of the future could soon get set in...

LessWrong (30+ Karma) ·

2025/08/12

Is there anything we can do to make the longterm future go better other than preventing the risk of extinction? My paper, Persistent Path-Dependence, addresses that question. I suggest there are a number of mechanisms that are fairly likely to be developed within our lifetimes that would result in extremely persistent path-dependent effects of predictable expected value. These include the creation of AGI-enforced institutions, a global concentration of power, the widespread settlement of space, the

en-gb

4:09

[Linkpost] “Thoughts on extrapolating time horizons”...

LessWrong (30+ Karma) ·

2025/08/12

This is a link post. (written for a Twitter audience) Has AI progress slowed down? I’ll write some personal takes and predictions in this post. The main metric I look at is METR's time horizon, which measures the length of tasks agents can perform. It has been doubling for more than 6 years now, and might have sped up recently. By measuring the length of tasks AI agents can complete, we can get a continuous metric of AI capabilities. Since 2019, the time horizon has been doubling every 7 months

en-gb

20:40

“How Does A Blind Model See The Earth?” by henry

LessWrong (30+ Karma) ·

2025/08/11

Sometimes I'm saddened remembering that we've viewed the Earth from space. We can see it all with certainty: there's no northwest passage to search for, no infinite Siberian expanse, and no great uncharted void below the Cape of Good Hope. But, of all these things, I most mourn the loss of incomplete maps. In the earliest renditions of the world, you can see the world not as it is, but as it was to one person in particular. They’re each delightfully egocentric, with the cartographer's home most

en-gb

7:40

“If worker coops are so productive, why aren’t they...

LessWrong (30+ Karma) ·

2025/08/11

Worker cooperatives are firms that, unlike traditional firms, are run democratically. This means that instead of the owner of the firm deciding who manages the workers, the workers become part owner and get a say in how the firm is run. This has some advantages, such as workers working harder and productivity appearing to increase. Wait… how's that even possible in theory? You might be asking yourself: won't workers become lazy since the profit is shared with their colleagues, which means they

en-gb

64:36

“GPT-5s Are Alive: Basic Facts, Benchmarks and the...

LessWrong (30+ Karma) ·

2025/08/11

GPT-5 was a long time coming. Is it a good model, sir? Yes. In practice it is a good, but not great, model. Or rather, it is several good models released at once: GPT-5, GPT-5-Thinking, GPT-5-With-The-Router, GPT-5-Pro, GPT-5-API. That leads to a lot of confusion. What is most good? Cutting down on errors and hallucinations is a big deal. Ease of use and ‘just doing things’ have improved. Early reports are thinking mode is a large improvement on writing. Coding seems improved and can compete

en-gb

24:32

“Breaking the Cycle of Trauma and Tyranny: How...

LessWrong (30+ Karma) ·

2025/08/11

A developmental perspective on authoritarian leadership and how we can build more resilient societies. Introduction Five years ago, David Althaus and Tobias Baumann published a delightful article “Reducing long-term risks from malevolent actors.” It focuses on the risk factors that “malevolent actors” pose when it comes to long-term catastrophic effects on civilization. Large parts of the article are devoted to the screening for 99th percentile “Dark Tetrad” traits – traits of Machiavellianism,

en-gb

4:06

“My Least Libertarian Opinion: Ban Exclusivity...

LessWrong (30+ Karma) ·

2025/08/11

* With sufficiently large and entrenched companies. There's a semi-common meme on Twitter where people share their most X opinion, where X is a group the poster doesn't identify with; or sometimes my least X opinion, where X is a group they do identify with.https://x.com/Aella_Girl/status/1899029178856710572 In that spirit, my least libertarian opinion is that exclusivity deals with sufficiently entrenched companies[1] are bad and should be illegal[2]. To make it even less libertarian, I think it

en-gb

4:21

“Having children is a deeply personal choice. Do not...

LessWrong (30+ Karma) ·

2025/08/11

First off, every ethical argument for having children is dominated by other options that are more effective. 1) If you’re worried about population issues, just donate $10k to bednets. That's roughly the equivalent of two extra children existing in the world. If you only care about population issues in your own country, then there are undoubtedly charities that save/create more lives in your own country that are cheaper than having your own children. I would also recommend questioning your

en-gb

14:29

“A Self-Dialogue on The Value Proposition of Romantic...

LessWrong (30+ Karma) ·

2025/08/10

Meta: This was written for myself to clarify various thoughts; if you’re seeing it then I thought other people might find value in it or might provide valuable-to-me responses, but other people are not really the audience. “We” = “I”; it's a dialogue amongst parts of myself. Voice 1: A couple months ago I wrote The Value Proposition of Romantic Relationships. On a personal level, the main motivation for that post was… I saw people around me who seemed to really highly value their relationships,

en-gb

14:56

“4 places where you can put LLM monitoring” by Fabien...

LessWrong (30+ Karma) ·

2025/08/10

To prevent potentially misaligned LLM agents from taking actions with catastrophic consequences, you can try to monitor LLM actions - that is, try to detect dangerous or malicious actions, and do something about it when you do (like blocking the action, starting an investigation, …).[1] But where in your infrastructure should your LLM monitor be implemented, and what team in your organization should own it? Historically, researchers have mostly focused on implementing monitoring inside the agent

en-gb

41:09

“OpenAI’s GPT-OSS Is Already Old News” by Zvi

LessWrong (30+ Karma) ·

2025/08/09

That's on OpenAI. I don’t schedule their product releases. Since it takes several days to gather my reports on new models, we are doing our coverage of the OpenAI open weights models, GPT-OSS-20b and GPT-OSS-120b, today, after the release of GPT-5. The bottom line is that they seem like clearly good models in their targeted reasoning domains. There are many reports of them struggling in other domains, including with tool use, and they have very little inherent world knowledge, and the safety

en-gb

7:44

“The Tortoise and the Language Model (A Fable After...

LessWrong (30+ Karma) ·

2025/08/09

Achilles had just finished installing his new AI assistant when the Tortoise ambled by, looking bemused. "Another technological marvel, I see," said the Tortoise, peering at the softly glowing terminal. "Indeed!" exclaimed Achilles. "This is Claude, the latest in artificial intelligence. It can answer any question, write poetry, even discuss philosophy. Watch this: Claude, what am I thinking right now?" The terminal responded: "I cannot actually read your thoughts. I generate responses based on

en-gb

19:52

“Extract-and-Evaluate Monitoring Can Significantly...

LessWrong (30+ Karma) ·

2025/08/09

Work produced at Aether. Thanks to Benjamin Arnav for providing us experimentation data and for helpful discussions, and to Francis Rhys Ward and Matt MacDermott for useful feedback. Executive Summary We apply a two-step chain-of-thought (CoT) monitoring scheme to transcripts fromArnav et al. (2025) and find that this substantially outperforms simple CoT monitors:our extract-and-evaluate monitoring approach improves monitor catch rate at 1% FPR by 23.7 percentage points, averaged over all models,

en-gb

4,486 results

LessWrong (30+ Karma)

All Episodes

Similar Podcasts

Download This Show

ABC Australia

The DroneOn Show

Mike & Kevin

The Good Tech Companies

HackerNoon

TechCrunch Industry News

TechCrunch

Tech News Briefing

Tech News Briefing