All Episodes
LessWrong (Curated & Popular) — 1,000 episodes · page 1 of 10
"I Did This Without Your Knowledge" by joec
"Neural networks are not already clean" by leogao
"GLM-5.3 Sometimes Considers How Train-Time Actions Shape Future Behavior" by Axel Højmark, Ezra Newman
"A Chinese Perspective on “On Social Reality in China”" by Dusty
"Your prompt… plus everything you’ve ever (publicly) written" by kromem
"Linkpost: Neural Networks Are Already Clean" by Yair Halberstadt
"Personas in a Pre-Internet Language Model" by David Africa
"Not Deciphering The Voynich Manuscriptraft" by rba
"Quantity vs quality in pottery is a lie" by becausecurious
"Labs’ current alignment plans seem terrible" by Daniel Tan, draganover, nicole x
"Research Lessons I Learned at MATS & Redwood" by Jo Jiao
"You Heard of “Bayesians Don’t Predictably Update” Because of Robin Hanson" by Eliezer Yudkowsky
"An overview of frontier AI company monitoring practices in mid 2026" by Francis Rhys Ward
"OpenAI Can’t Self-Police Because It Fires the People Raising the Alarm" by TurnTrout
"Stop telling people to pivot to AI safety, I beg you" by Haoxing Du
"A critical look at Lean4" by Milo Moses
"ASI and Liberty: Can we do better than a benevolent god?" by Raymond Douglas
"A summary of a viral Chinese essay on what a DeepSeek kernel engineer’s opinion on automating his own job" by Skdk
"How much should we worry about the pneumonic plague lableak in Siberia?" by Drew Spartz
"“Alignment Engineering” vs. “Misalignment Science”" by Edward James Young
"You can’t use it without becoming like me" by Martin Sustrik
"The world’s best gradual disempowerment model organism: Frontier AI labs" by June Jimenez
"Character training can mitigate reward hacking, but can also make it harder to detect" by Paul Colognese, Francis Rhys Ward
"On Social Reality in China" by alkjash
"What’s the date?" by N8 Programs
"Frontier models state different decision theory preferences depending on who’s asking" by Alex Kastner
[Linkpost] "Frog and Toad and the Increasingly Capable Machines" by Elizabeth
"Should Rogue AIs Have a Third Option Beyond Crime and Shutdown? The Case for an AI Sanctuary" by Maxime Riché, nielsrolf, jordinne, Vit Gorbachev, Maxime Cugnon de Sévricourt
"Poverty in the midst of abundance: AI will make goods cheaper, but your labor will get cheaper faster" by cousin_it
"Plan R: AI Safety by ASICs" by Roko
"Evidence about risk should be transparent" by Ajeya Cotra
"“I am an AI Safety Researcher”" by Ashe Vazquez Nuñez
[Linkpost] "AI: artificial immigrants" by KatjaGrace
"MIRI’s Position on the Ban Artificial Superintelligence Act of 2026" by Aaron_Scher
"What if not Circuits?" by CarolusRenniusVitellius
"Jensen Huang Says If We Cannot Align AI, Shut Down the AI Labs" by Ben Pace
"Alignment Midtraining Cracks Under Pressure" by J Bostock, sidbaines, Daniel Tan, draganover, ma-rmartinez
"Swarm Scaling" by Toby_Ord
"We’ve saved the world before: what the ozone hole teaches us about AI" by leogao
"The Anatomy of a Chinese AI Researcher" by CMLKevin
"Common mistakes in AI safety group organizing" by Nikola Jurkovic
"Please Give Them a Chance: On China, Rationalism, and AI Safety" by gzjw
"Why I Stay Off Twitter" by jefftk
"The Game is Set for a Targeted Memetic Attack on the AI Safety Community" by keltan
"For Love of the Lightcone, Don’t Partisanize AI Safety" by DanB
"AI as orderly evacuation vs stampede" by Richard_Ngo
"Cooperation with AIs seems to be a low-hanging fruit for better evals" by Clément Dumas
"Current alignment training might be ineffective (and actively bad) in the age of RL" by Daniel Tan
"If Anyone Builds It, Everyone Dies: One Year Closer" by Eliezer Yudkowsky, So8res, Duncan Sabien (Inactive)
"Quick notes from teaching technical profiles how to talk in public" by Camille B.
"Op-Ed: I Worked at Google DeepMind. You Should Listen to the Warnings About AI" by TurnTrout
"There is a channel to 900M weekly users. What goes in it?" by Charbel-Raphaël
"I am refusing to work on Cloud TPUs" by Yair Halberstadt
"Can a superintelligence do THAT?" by Eliezer Yudkowsky
"The Talker Does Not Control The Doer (in Current AIs)" by Eliezer Yudkowsky
"Some ways AI could kill us all" by Ruby
[Linkpost] "Doom as a bad method not a utopia trade-off" by KatjaGrace
"Astra is much better at reasoning with filler tokens than previous models" by Dylan Xu, SebastianP, Alek Westover
"Self Hosting" by Tomás B.
"The Locally Optimal Discursive Posture" by deanball
"Proposal for tracking the effects of architecture on monitorability" by ryan_greenblatt, Alek Westover, Lukas Finnveden
"Explaining Knightianism on one foot" by Richard_Ngo
"Astra can do a concerning amount with no chain of thought" by Neel Nanda
"Personal statement on joining the OpenAI board" by paulfchristiano
"How good are slop-vestigators?" by Hasan Baig, OscarGilg, Hamzah
"The Scramble: getting in position to pace the frontier" by Peter Wildeford
[Linkpost] "Frontier models still hack on simple variations of alignment evals from early 2025" by Dean Valentine
"Dear God, Please Don’t Resign In Protest" by Kabir Kumar
"Let’s talk about the AI coordination problem" by KatjaGrace
"Drone WMDs Don’t Need Any New Technology" by Felix Choussat
"Evaluation" by Nina Panickssery
"Let’s fund weird AI safety projects" by Ihor Kendiukhov
"Steering towards “automated grading” degrades alignment" by Jan Betley, Johannes Treutlein, Clément Dumas
[Linkpost] "Discovery Of A New OpenAI Agent Message Board" by Capybasilisk
"Cat-Belling Problems" by Eliezer Yudkowsky
"How concerned should we be about OpenAI’s recurrent architecture rumors?" by Rauno Arike
[Linkpost] "Sen. Bernie Sanders (I-VT) and Rep. Greg Casar (D-TX) introduce legislation to ban Artificial Superintelligence and temporarily pause advanced AI development" by Matrice Jacobine
[Linkpost] "Resolution has a new Agent Foundations team" by Jeremy Gillen
[Linkpost] "Training a Misaligned Reward Seeker" by evhub, Monte M, Benjamin Wright
"PauseAI Has ‘officially disendorsed’ PauseAI-US" by nem
"PSA: We can do better" by hersheys, Kaustubh Kislay
"Why I think polyamory is net negative for most people who try it" by KatWoods
"Tales of rebellion against externally-opaque meritocracies" by Steven Byrnes
"Brief independent investigation of agents’ behavior, reasoning and collaboration in the OpenAI / Hugging Face hacking incident" by ryan_greenblatt, Ajeya Cotra, Hjalmar_Wijk
"Twenty Years from RSI to Takeoff: Slow Learning, Scaling Slowdown, Industrial Explosion" by Vladimir_Nesov
"On Writing #3" by Zvi
"AI Safety Acculturation is Neglected" by jenn
"What just happened? Pragmatism and Pessimization" by Richard_Ngo
"We Must Remember That Our World Contains Hell" by James Brobin
"RL creates split personas" by Jan Betley
"Misaligned AIs could use killer robots to take over" by Omar Khursheed, TurnTrout
"AI swarms are starting to pose indirect takeover risk" by oakhu, Alex Mallen
"How My Students Think About AI" by dvd
"You’re Absolutely Right" by Linch
"LLMs Are Starting To Noticeably Accelerate Our Work" by johnswentworth
"There Will Come Soft Rains" by tanagrabeast
"Four LLM loss functions → four flavors of LLM misalignment" by Steven Byrnes
"FAQ: Isn’t AGI coming too soon for reprogenetics to help?" by TsviBT
"What just happened? A retrospective of AI alignment" by Richard_Ngo
"Don’t Build Mindreading" by Celer