PODCAST

LessWrong (30+ Karma)

LessWrong

Audio narrations of LessWrong posts.

technology philosophy

All Episodes

4:59

“Talk English, Think Something Else” by J Bostock

LessWrong (30+ Karma) ·

2026/04/13

There's an adage from programming in C++ which goes something like "Yes, you write C, but you imagine the machine code as you do." I assumed this was bullshit, that nobody actually does this. Am I supposed to imagine writing the machine code, and then imagine imagining the binary? and then imagine imagining imagining the transistors? Oh and since I don't actually use compiled languages, should I actually be writing Python, then imagining the C++ engine, and so on? Then one day, I was vibe-coding,

en-gb

3:55

“Sparse Autoencoders for Single-Cell Models” by Ihor...

LessWrong (30+ Karma) ·

2026/04/13

People are rushing to build bigger and bigger single cell foundation models (trained on RNA sequencing data), but in my view we have not extracted even a small fraction of the knowledge and capabilities that already exist inside the models we have today. To explain what I mean, I want to argue three things in this post, and then show the empirical work behind them. Thesis 1: Biological foundation models are not like LLMs, and the field's habit of evaluating them the same way is causing us to

en-gb

7:49

“Eggs, rooms, puzzles, and talking about AI” by...

LessWrong (30+ Karma) ·

2026/04/13

I live with five friends in a big house, and two things I’ve done in it on this particular Sunday are hide 156 easter eggs all around, and reach a tentative joint decision on the allocation of four of its rooms. These tasks are delightful to me for a reason they have in common, and from which I hope to gesture at extremely far reaching conclusions. Easter eggs A room usually seems like a simple thing to me—a big box, with some smaller mostly boxish objects and holes in it. Each of those

en-gb

4:43

“Morale” by J Bostock

LessWrong (30+ Karma) ·

2026/04/12

One particularly pernicious condition is low morale. Morale is, roughly, "the belief that if you work hard, your conditions will improve." If your morale is low, you can't push through adversity. It's also very easy to accidentally drop your morale through standard rationalist life-optimization. It's easy to optimize for wellbeing and miss out on the factors which affect morale, especially if you're working on something important, like not having everyone die. One example is working at an office

en-gb

5:39

“Your Mom is a Chimera” by michaelwaves

LessWrong (30+ Karma) ·

2026/04/12

And so are you! When you were a fetus, you were sending millions of your cells through the placenta into your mom. And she was sending her cells into you, although to a lesser degree. These cells made themselves right at home, differentiating into heart, blood, and even brain cells. This phenomenon is called feto-maternal microchimerism, and is one of the wildest things in placental mammal pregnancy. What? Microchimerism is generally defined as the presence of a small population of genetically

en-gb

18:09

“The Blast Radius Principle” by Martin Sustrik

LessWrong (30+ Karma) ·

2026/04/12

In April 2024, a salvo of cruise missiles destroyed the Trypilska thermal power plant, the largest in the Kyiv region, in under an hour. In June 2023, the destruction of the Kakhovka dam left a million people without drinking water and wiped out an entire irrigation system downstream. Throughout three winters, strikes on combined heat and power plants have left apartment buildings in Kyiv at indoor temperatures barely above freezing. In December 2023, a single cyberattack on Kyivstar, Ukraine's

en-gb

“How to make good tea” by RobertM

LessWrong (30+ Karma) ·

2026/04/12

If you're starting from a baseline of drinking relatively cheap mass-market teabags, the easiest way to marginally improve your tea quality is by making sure you don't oversteep it. If it's a typical black tea, try 3 minutes (rather than 4-5). If it's some kind of green, try 2 minutes, and you also have a second very easy marginal improvement: use colder water. Most greens should be brewed at 175F/80C. If you don't have an adjustable-temperature tea kettle, you can get pretty close by pouring

en-gb

7:18

“Catching illicit distributed training operations...

LessWrong (30+ Karma) ·

2026/04/12

Last year, my colleagues on MIRI's Technical Governance Team proposed an international agreement to halt risky development of superhuman artificial intelligence until it can be done safely. The agreement would require all clusters of AI chips with more computing power than 16 H100 GPUs to be registered with a coalition of states, led by the US and China, that would monitor their operations to ensure they aren’t being used for unsafe AI development. In my opinion, the proposal is impressively

en-gb

2:36

[Linkpost] “Scott Alexander gentrified my meetup” by...

LessWrong (30+ Karma) ·

2026/04/11

This is a link post. On 10 April 2026, I organized an ACX/LW/rationality meetup. Initially, I intended for it to be just our regular meetup, and “regular” for Zagreb is between 3 and 10 people having a couple of beers. Skyler of Rationality Meetups contacted me saying that Scott would be in Zagreb in April, and we coordinated so that the meetup could accommodate Scott's schedule. At first, I thought, man, this is serious now. I’ll have to actually reserve a table. Once the “Going” RSVPs started

en-gb

5:21

“Pausing AI Is the Best Answer to Post-Alignment...

LessWrong (30+ Karma) ·

2026/04/11

Even if we solve the AI alignment problem, we still face post-alignment problems, which are all the other existential problems[1] that AI may bring. People have identified various imposing problems that we may need to solve before developing ASI. An incomplete list of topics: misuse; animal-inclusive AI; AI welfare; S-risks from conflict; gradual disempowerment; permanent mass unemployment; risks from malevolent actors/AI-enabled coups/gradual concentration of power; moral error. If we figure

en-gb

21:41

“Some thoughts on Nectome’s risk and resilience” by...

LessWrong (30+ Karma) ·

2026/04/11

One of the best ways to reduce Nectome's long-term risk is to show that preservation is a thing people want by buying one yourself; this is a critical time in the organization and your contributions now have an outsized impact in our likelihood of success. I'm happy to discuss our preservation personally with anyone who's interested. Our current presales are open until the end of April. https://nectome.substack.com/p/preservation-pre-sales Up till now I've been talking about Nectome's recent

en-gb

7:22

“Chocolate Sloths, Tinder, and Moral Backstops” by J...

LessWrong (30+ Karma) ·

2026/04/11

My grandma has a poor understanding of moral hazard, when it comes to buying me 155g chocolate sloths. Moral hazard is a concept in political economy which refers to the dangers of incentivising people to behave badly. Classic examples include opening casinos (which makes people gamble away their money) preventing evictions for non-payment of rent (which makes people not pay their rent, which makes their situation even worse in the long run) and state-sponsored insurance for disasters (which makes

en-gb

12:33

“Dario probably doesn’t believe in superintelligence”...

LessWrong (30+ Karma) ·

2026/04/11

Epistemic status: I think this is true but don't think this post is a very strong argument for the case, or particularly interesting to read. But I had to get 500 words out! I think the 2013 conversation is interesting reading as a piece of history, separate from the top-level question, and recommend reading that. I think many people have a relationship with Anthropic that is premised on a false belief: that Dario Amodei believes in superintelligence. What do I mean by "believes" in

en-gb

12:43

“The Unintelligibility is Ours: Notes on...

LessWrong (30+ Karma) ·

2026/04/11

Many people seem to think that the chains-of-thought in RL-trained LLMs are under a great deal of "pressure" to cease being English. The idea is that, as LLMs solve harder and harder problems, they will eventually slide into inventing a "new language" that lets them solve problems better, more efficiently, and in fewer tokens, than thinking in a human-intelligible chain-of-thought.I'm less sure this will happen, or that it will happen before some kind of ASI. As a high-level intuition pump for

en-gb

13:05

“If Mythos actually made Anthropic employees 4x more...

LessWrong (30+ Karma) ·

2026/04/11

Anthropic's system card for Mythos Preview says: It's unclear how we should interpret this. What do they mean by productivity uplift? To what extent is Anthropic's institutional view that the uplift is 4x? (Like, what do they mean by "We take this seriously and it is consistent with our own internal experience of the model.") One straightforward interpretation is: AI systems improve the productivity of Anthropic so much that Anthropic would be indifferent between the current situation and a

en-gb

69:38

“Claude Mythos #2: Cybersecurity and Project...

LessWrong (30+ Karma) ·

2026/04/10

Anthropic is not going to release its new most capable model, Claude Mythos, to the public any time soon. Its cyber capabilities are too dangerous to make broadly available until our most important software is in a much stronger state and there are no plans to release Mythos widely. They are instead going to do a limited release to key cybersecurity partners, in order to use it to patch as many vulnerabilities as possible in our most important software. Yes, this is really happening. Anthropic

en-gb

5:30

“Why Control Creates Conflict, and When to Open...

LessWrong (30+ Karma) ·

2026/04/10

tl;dr: with multiple agents, control attempts tend to create conflict, because control attempts shut down communications channels, which leads to feedback loops in the form of intensifying tug-of-war over variables. intentionally relaxing control to better understand the other agents can break the cycle, and forms the basis of many therapeutic and mediation techniques. [epistemic status: mostly quite confident based on experience, but making a fair few claims i'm not giving the full justification

en-gb

35:17

“Reproducing steering against evaluation awareness in...

LessWrong (30+ Karma) ·

2026/04/10

Produced as part of the UK AISI Model Transparency Team. Our team works on ensuring models don't subvert safety assessments, e.g. through evaluation awareness, sandbagging, or opaque reasoning. TL;DR We replicate Anthropic's approach to using steering vectors to suppress evaluation awareness. We test on GLM-5 using the Agentic Misalignment blackmail scenario. Our key finding is that “control” steering vectors – derived from contrastive pairs that are semantically unrelated to alignment – can have

en-gb

6:31

“Have we already lost? Part 2: Reasons for Doom” by...

LessWrong (30+ Karma) ·

2026/04/10

Written very quickly for the Inkhaven Residency. As I take the time to reflect on the state of AI Safety in early 2026, one question feels unavoidable: have we, as the AI Safety community, already lost? That is, have we passed the point of no return, after which AI doom becomes both likely and effectively outside of our control? Spoilers: as you might guess from Betteridge's Law, my answer to the headline question is no. But the salience of this question feels quite noteworthy to me nonetheless,

en-gb

12:45

“Model organisms researchers should check whether...

LessWrong (30+ Karma) ·

2026/04/10

Thanks to Buck Shlegeris for feedback on a draft of this post. The goal-guarding hypothesis states that schemers will be able to preserve their goals during training by taking actions which are selected for by the training process. To investigate the goal-guarding hypothesis, we’ve been running experiments of the following form: We call this type of training “behavior-compatible training”. This type of experiment is common in model organism (MO) research. For example, Sleeper Agents is of this

en-gb

4,454 results

LessWrong (30+ Karma)

All Episodes

Similar Podcasts

Lead With AI

Tamara Nall

Double Tap

Double Tap Productions Inc.

Claude AI Genius Podcast w/ Claude Anthropic Expert Brandon Gaille

Claude AI Expert Brandon Gaille

ものづくりnoシテン

しぶちょー@機械設計&AIエンジニア

Eli the Computer Guy

Eli the Computer Guy