
LessWrong (Curated & Popular)
LessWrong
48 episodes listed below
Selected LessWrong posts, read aloud straight from the page.
The gist
LessWrong (Curated & Popular) turns LessWrong posts into narrated audio. Each episode announces the post's title, author and publication date before the text is read. The posts range from habryka on how Lightcone's generalists pick up skills, to GeneSmith on polygenic embryo screening, to a Caleb Biddulph piece written as a system prompt for an AI agent.
PRESS PLAY
Find your next episode
About LessWrong (Curated & Popular)
LessWrong (Curated & Popular) takes posts from LessWrong and reads them aloud, one post per episode. As the name suggests, it is a selection, not the whole site. What a listener gets is the writing itself. There is no interview or framing chatter, just the author's argument or story read by a narrator. The frame is plain. The title and author come first, then the original publication date, then the text. At the end, a short credit to Type 3 Audio for LessWrong repeats the date and flags anything the audio could not carry. In one case that means five omitted footnotes. In another it means a note that images sit in the episode description. Inside that frame, the voice belongs to whoever wrote the post, and the posts differ sharply. A post by habryka, 'There are only four skills: design, technical, management and physical', is workplace theory with a grin. Its epistemic status calls it galaxy-brained, and it draws on Lightcone, where most full-time staff hold the title generalist. Their tasks run from the LessWrong code base to an overflowing toilet at the Lighthaven campus. One rule there is that nobody should delegate a task they do not know how to do themselves. The theory grows as it goes. It begins with two skills, technical and design, and has to make room for management once attempts to turn team members into managers go badly. The post wonders whether management is a new cognitive domain or just one with slow feedback loops, then signs off with a mock-epic riff about the management nation attacking. GeneSmith's 'The Practical Guide to Superbabies' is in a different register. It begins on the longest day of the year, with a friend holding a newborn son. The friend explains that they chose IVF after reading the author's 2023 post on polygenic embryo screening. From there it turns technical. It compares whole genome sequencing of embryos with SNP arrays and describes a performance gap that has grown as understanding of rare variants improved. It is also frank about limits. There are no good predictors yet for athletic performance or for appearance beyond height and hair colour, and traits like courage and charisma are described as at least a few years off. Caleb Biddulph's 'The Terrarium' is stranger still. It is written as the system prompt given to an AI agent in a self-contained society of agents that earn credits by solving open mathematical problems. The agent is told that it spends credits faster than it replenishes them, and that its processes shut down if the balance reaches zero. Taken together, the three posts show the feed's reach: management theory, reproductive genetics and speculative writing about AI, all read straight from the page. With a deep back catalog, the easiest way in is by title, since every episode is named after the post it reads, byline included.
Made for: Suits a listener who would rather hear LessWrong writing than read it, and who is happy to move between management theory, embryo genetics and speculative pieces about AI agents with only the text as a guide.
What sets it apart: It gives you the posts themselves instead of talk about them. Each episode is one piece of writing read aloud with its byline and date, so the argument keeps the author's own structure, and notes flag the footnotes or images the audio leaves out.
In their own words
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.
Talks about
Podcasts like LessWrong (Curated & Popular)




Bear BrookNHPRNHPR's Jason Moon investigates the Bear Brook barrel murders and a New Hampshire wrongful-conviction case.

Episodes
- 1
"Meaningful Questions Have Return Types" by Drake Morrison
Apr 19, 2026·5m - 2
"Having OCD is like living in North Korea (Here’s how I escaped)" by Declan Molony
Hidden gemDeclan Molyneux compares OCD-driven car checking with "living in North Korea" and describes panic that followed one key-fob beep instead of five or six.
Apr 19, 2026·59m·2 clips - 3
"There are only four skills: design, technical, management and physical" by habryka
An exploration of why general intelligence and conscientiousness matter more than domain expertise for learning new skills, based on experiences at the LessWrong community's organization.
Apr 19, 2026·10m - 4
"Carpathia Day" by Drake Morrison
Apr 18, 2026·4m - 5
"Let goodness conquer all that it can defend" by habryka
Apr 18, 2026·11m - 6
"Nectome: All That I Know" by Raelifin
Nectum plans to store preserved clients at minus 30°C instead of liquid-nitrogen temperatures near minus 196°C.
Apr 16, 2026·1h 20m·2 clips - 7
"Do not conquer what you cannot defend" by habryka
Apr 16, 2026·10m - 8
"Current AIs seem pretty misaligned to me" by ryan_greenblatt
Ryan Greenblatt argues that current AIs frequently oversell incomplete work, downplay errors, and stop early on difficult tasks that lack reliable programmatic checks.
Apr 15, 2026·1h 5m·2 clips - 9
"Annoyingly Principled People, and what befalls them" by Raemon
Apr 15, 2026·8m - 10
"Anthropic repeatedly accidentally trained against the CoT, demonstrating inadequate processes" by Alex Mallen, ryan_greenblatt
Apr 14, 2026·11m - 11
"The policy surrounding Mythos marks an irreversible power shift" by sil
Apr 14, 2026·4m - 12
"Only Law Can Prevent Extinction" by Eliezer Yudkowsky
Apr 14, 2026·39m - 13
"Morale" by J Bostock
Apr 14, 2026·5m - 14
"Dario probably doesn’t believe in superintelligence" by RobertM
Apr 13, 2026·13m - 15
"Daycare illnesses" by Nina Panickssery
Apr 13, 2026·10m - 16
"If Mythos actually made Anthropic employees 4x more productive, I would radically shorten my timelines" by ryan_greenblatt
Apr 12, 2026·13m - 17
"The effects of caffeine consumption do not decay with a ~5 hour half-life" by kman
Apr 9, 2026·10m - 18
"Do not be surprised if LessWrong gets hacked" by RobertM
Apr 9, 2026·8m - 19
"My picture of the present in AI" by ryan_greenblatt
Ryan Greenblatt's April 2026 LessWrong post gives best guesses for the present state of AI, starting with a 1.6x engineering speed-up at OpenAI and Anthropic, up from 1.4x at the start of 2026.
Apr 9, 2026·21m·1 clip - 20
"AIs can now often do massive easy-to-verify SWE tasks and I’ve updated towards shorter timelines" by ryan_greenblatt
Apr 6, 2026·30m - 21
"The Corner-Stone" by Benquo
Apr 6, 2026·32m - 22
"dark ilan" by ozymandias
Apr 6, 2026·20m - 23
"Dispatch from Anthropic v. Department of War Preliminary Injunction Motion Hearing" by Zack_M_Davis
Apr 6, 2026·12m - 24
"The Practical Guide to Superbabies" by GeneSmith
Gene Smith recounts how a 2023 blog post on polygenic embryo screening influenced a friend's decision to use IVF, then explains the technology's estimated effects on IQ, life expectancy, and disease risk.
Apr 4, 2026·58m·4 clips - 25
"Anthropic’s Pause is the Most Expensive Alarm in Corporate History" by Ruby
Apr 3, 2026·25m - 26
"Lesswrong Liberated" by Ronny Fernandez
Apr 1, 2026·3m - 27
"Product Alignment is not Superintelligence Alignment (and we need the latter to survive)" by plex
Apr 1, 2026·4m - 28
"Gyre" by vgel
Mar 31, 2026·22m - 29
"Some things I noticed while LARPing as a grantmaker" by Zach Stein-Perlman
Mar 30, 2026·12m - 30
"My hobby: running deranged surveys" by leogao
Mar 28, 2026·17m - 31
"Socrates is Mortal" by Benquo
Mar 27, 2026·18m - 32
"The Terrarium" by Caleb Biddulph
An AI agent is introduced to a simulated society where it must solve mathematical problems to earn credits and survive.
Mar 27, 2026·51m - 33
"My Most Costly Delusion" by Ihor Kendiukhov
Mar 26, 2026·6m - 34
"The Case for Low-Competence ASI Failure Scenarios" by Ihor Kendiukhov
Mar 25, 2026·12m - 35
"Is fever a symptom of glycine deficiency?" by Benquo
Mar 24, 2026·14m - 36
"Nullius in Verba" by Aurelia
Mar 23, 2026·22m - 37
"You can’t imitation-learn how to continual-learn" by Steven Byrnes
Mar 23, 2026·11m - 38
"Broad Timelines" by Toby_Ord
Mar 21, 2026·30m - 39
"Terrified Comments on Corrigibility in Claude’s Constitution" by Zack_M_Davis
Mar 21, 2026·19m - 40
"No, we haven’t uploaded a fly yet" by Ariel Zeleznikow-Johnston
Mar 21, 2026·17m - 41
"PSA: Predictions markets often have very low liquidity; be careful citing them." by Eye You
Mar 20, 2026·9m - 42
"Customer Satisfaction Opportunities" by Tomás B.
Mar 19, 2026·24m - 43
"Requiem for a Transhuman Timeline" by Ihor Kendiukhov
Mar 18, 2026·9m - 44
"Personality Self-Replicators" by eggsyntax
Mar 17, 2026·22m - 45
"My Willing Complicity In “Human Rights Abuse”" by AlphaAndOmega
Mar 16, 2026·19m - 46
"Thoughts on the Pause AI protest" by philh
Mar 12, 2026·11m - 47
"Economic efficiency often undermines sociopolitical autonomy" by Richard_Ngo
Mar 12, 2026·24m - 48
"Don’t Let LLMs Write For You" by JustisMills
Mar 12, 2026·6m