October 2026
- 23Post-Partisanship Is Hyper-Partisanship
- 23Can Things Be Both Popular And Silenced?
- 23What policies have most thoroughly crippled (otherwise-promising) industries or technologies?
- 23Quick Thoughts on Overhang Arguments
- ·
23The Lens That Conceals Its Flaws: Nonsense at the Heart of the Sequences - 23Meetup : East Coast Super Fun Rationalists Sleepover Party Extravaganza
23Budiansky and local food - Matt Ridley- 23On The Rationalist Megameetup
- ·23What Is Doug Doing Now?
- ·
23So, You Think Someone Should Be Banned From A Furry Con - 23Nearly 30 Years of Convention Conbadges
- ·
23Archives - 23On Not Pulling The Ladder Up Behind You
23The Bunny: An EA Short Story- 23Wah wah wah wah do more AI R&D
- 23A lost 80s/90s metaphor: playing the demo
- 23Kids Learn by Copying
- 22How does Fable 5.1 hide messages in images?
- 22open-thread-456
- 22against-against-handover
- 22Misaligned!
- 21Crémieux on X: "It's actually insane that there are papers coming out about AI capabilities evaluations using two-year-old models.We have to reduce the length of time to get studies published or academic publishing will have to become completely irrelevant for AI." / X
21What to sort out before the money lands- 21Sleep puzzle system — LessWrong
- 21Chain of Thought For Humans
- ·21We are at war.
- 21Organizing a Van to EA Global
- 21A short review of "The Zizians: Inside a Trans Death Cult" — LessWrong
- 21I Did This Without Your Knowledge — LessWrong
- 20Dying Outside — LessWrong
- ¶✓
20The Smithsonian's Dwindling Attendance - ◐20Please don't throw your mind away
- ✓20Music in Human Evolution | Melting Asphalt
- ★★★★★20Math 01: Update on Chowla Conjecture
- 20A fascinating case of an LLM prompt-injecting its own user
- 20Sorry, you are still dead
- ◐20On Social Reality in China
- ◐60%20Criticism of EA Criticism Contest
- 20Monthly Roundup #47: October 2026
- 20The Lightcone Needs You To Plan a Vacation
- 20Anthrax and Anti-ASI Advocacy
- 19Investing in the public benefit
- 19mississippi-burning
- ·★★17The plan for hack club world dominance - Deven Jadhav
- ◐16Koan: divining alien datastructures from RAM activations
- 15Two Cult Koans
- 15The Trouble With "Good" — LessWrong
- 15Decision Theories: A Less Wrong Primer
- 15Archimedes's Chronophone
- 14Epistemic Luck
- 13What Developmental Milestones Are You Missing?
- 13Silicon Valley: A Reality Check
- 12plzdontkillus Fellows Got ~2M AI Safety Views, Not 21M
- 12Do you have any ideas?
- 12My Buddy Mike, Efficiency Gains From AI, and The World’s Most Boringly Benign Criminal Conspiracy
- 12Vibe signaling externalities and the people-to-places pipeline
- 12How to Seem (and Be) Deep
- 12on-the-origins-of-peanut-allergy
- 12The Rebel Sell
- 12The "Outside the Box" Box
- 12A Unified Theory of VC Suckage
- 12Who Runs Jane Street? A Guide to Elusive Firm’s Power Players
- 12Competent Elites
- 12Explain/Worship/Ignore?
- 12Einstein's Superpowers
- 12Crackpot index
- 12The Level Above Mine
- 12Different Worlds
- 11The Road to Hell
- 11Is it more unusual to have written a book, or to have killed someone?
- 11Short rant about the old books/AI thing
- 11Mindreading Fails Both Ways
- 11Stop Asking, Start Doing
- ¶11You Are The Deer In The Subway.
- 11AWS Has a Moral Responsibility To Fix the Free Tier
- ◐11The Unacknowledgeable Work of Pastors’ Wives
- ·11Childhood and Education #21: Grades and Standards
- 11How I got into AI research
- 11You should write everything down - ueaj
- ◐11Open Source Dystopia - ueaj
- 11Dangerous Things
- 11You Heard of "Bayesians Don't Predictably Update" Because of Robin Hanson
- 11Advice for journalists
- 11The salad market mystery
- 11Saving Academics from Themselves
- 11What Will Happen to Indian IT?
- ◐11A critical look at Lean4 — LessWrong
- 9The Doomers Are All Right
- ◐9Selling Abstraction
- 9A Normal Person 30 Years Ago
- ★★★★★9culturalcompetition
- 9You Have No Right to Your Culture
- 95-MeO-DMT sometimes cancels the Ebbinghaus illusion
- 9The Necessary Preconditions
- ◐8AI #189: New Math
- 8You can't use it without becoming like me
- 8Gluten-free water
- 8Orgs: unreasonable boyfriend as service
- 8On donating a kidney
- 7An Open Letter To Steven Pinker On AI
- 7How much should we worry about the pneumonic plague lableak in Siberia?
- 7Stranger Than History
- ↻17Church Planting: When Venture Capital Finds Jesus
- 7Pivotal act — LessWrong
- ·6Human Rights Watch Reports Are No Longer Credible
- 6Open Thread 454
- ↻16Why American ambulance rides are so expensive
- 4Lab Leak in Russia
- 4How to change your identity. A guide to the highest leverage behavior change technique.
- 4Athletic education
- ◐4AI #188: Gemini Dot Argon
- ·2Argument Fixer Home Page
- ◐1Funding for-profits with charitable dollars
- 1What's the date?
- 1Monthly Roundup #46: September 2026
September 2026
- ·★30Ethan Yip | AI Researcher & Engineer
- ◐30Reality-Revealing and Reality-Masking Puzzles
- 30Being a Robust Agent
- 29The Darwin Pregame
- 29The Darwin Game
- 29Addictions are anesthesia
- ◐29Claude Opus 5.5 Should Raise Your Ambitions
- ◐29Trading firms could control meaningful amounts of compute by 2030
- 29“China needs more AI safety.” More of what, exactly?
- 29It’s Time to Investigate the AI Labs
- ◐28Pirating the Pirates
- 28"Can you keep this confidential? How do you know?"
- 28Cultivating And Destroying Agency
- 28did-milton-friedman-really-refute
- 28Swarm Scaling
- 28Notes on Private Trackers
- 28I Drink Soylent So My Broccoli...
- ·27Eric Weinstein on X: "Okay. I’ll take the bait.I went to work for Peter in 2013 with opposite politics.He sat me down right at the beginning. PETER: “I don’t want the (Thiel) Foundation to do good that the market can do. That is the job of the market. Likewise, I don’t want to do the good work" / X
- ★★★★★27Frog and Toad and the Increasingly Capable Machines
- ◐27It May Be Possible to Improvise A High Grade Bioshelter
- 26High School or Claude?
- ◐26AI #187: Coming Into Play
- 25MIRI’s Position on the Ban Artificial Superintelligence Act of 2026 - Machine Intelligence Research Institute
- 25The Specter Of Neuralese
- 25At annual meeting, ACCF encourages Giving Spree donations
- ·24AI Notkilleveryoneism Memes ⏸️ (@AISafetyMemes) on X
- 23Need Legal Advice in Prison? A Burrito? Even a Sex Toy? Check the Toilet
- ◐23Claude Opus 5.5: The System Card
- ·23The Lottery Wizard
- 23Minimal Vs Maximal superintelligence
- 23AI: artificial immigrants
- 23Jensen Huang Says If We Cannot Align AI, Shut Down the AI Labs
- 23Elephant seal 2
- 23Signals of Slop: How to identify and avoid creating slop
- 23The Normalization of Deviance in AI Development
- 23Premature Optimization
- ◐23testosterone-isnt-the-aggression
- 23Mysteries Of AI Generalization
- +22download
- ◐22Nuclear option
- 22The Horse
- 22Why I Stay Off Twitter
- ★★★★22The Overhang Forecasting Conference
- 22trades-with-fae
- ↻122Rule Thinkers In, Not Out
- ◐22Words fail us
- 22Open Thread 452
- 22An early warning system for AI incidents
- ◐21Better Call Sol Or Better Yet Claude or Astra
- ◐21Is METR A Meaningful Check On Anthropic?
- ·21i-hated-high-school
- 21building-a-podcast-studio-during
- ↻119Nuclear war is unlikely to cause human extinction
- 19The Game is Set for a Targeted Memetic Attack on the AI Safety Community
- ◐18The Preference Cascade Is Only Getting Started
- ◐18Trump Goes Full Hoax on AI Existential Risk
- ·★★★★★18If Anyone Builds It, Everyone Dies: One Year Closer
- 16Some ways AI could kill us all
- 16Teleoperated Humans
- 16The Bad Guy With An AI Named Claude
- 15Brand New AI Solves a Millennium Prize
- ◐15GPT-6-Astra Can Do Ambitious Things
- 15Legal Maximums on Context Windows
- 15I worked at Google DeepMind. You should listen to the warnings about AI | Alex Turner
- 15872: Fairy Tales
- 14Deployment
- ¶14On insight on this reading log
- 14The Talker Does Not Control The Doer (in Current AIs)
- 14There is a channel to 900M weekly users. What goes in it?
- 14I am refusing to work on Cloud TPUs
- 11The Extinction Risk Preference Cascade: Quotes
- 11Guardian Angels: LLM Personalization for Productivity and Security
- ↻111A Rational Argument
- 11why-doing-operations-work-feels-bad
- 11when-i-got-into-yale-my-family-asked
- ★★★★★10So where's this AI thing going?
- ◐10AI #185: Preference Cascade
- 10The Art of Shipping Slopware
- ◐9This Can't Go On
- 9Personal statement on joining the OpenAI nonprofit board
- 9The Scramble: getting in position to pace the frontier
- 9Self Hosting
- 9GPT-6 Astra: The System Card, Alignment and What Comes Next
- 9Pausing AI ASAP is preferable to agreeing to pause at some future time — LessWrong
- 9open-thread-450
- 9god-help-us-lets-try-to-learn-about
- 8Building a High Talent Density Organization
- 8Making History Available
- ◐8Claude Fable 5.1 and Mythos 5.1: The System Card
- 8Dear God, Please Do Not Resign In Protest
- 8Praise our Lord and Savior, Glycine: How Opus 4.6 gifted me The Vitamin
- 8An AI slowdown is better than a pause
- 8Tales of rebellion against externally-opaque meritocracies
- 8Kairos has raised $50M to build talent infrastructure for AI safety (and we're hiring!)
- ◐8Notes on a Consequential Few Days
- 8Let's talk about the AI coordination problem
- 8Should safety researchers quit frontier labs re. warning shots?
- ◐8Drone WMDs Don’t Need Any New Technology
- ◐8An Alien Mind: Jakub Pachocki Warns Us
- 7Claude Mythos 5.1 and Fable 5.1: Capabilities
- 72585: Rounding
- 73164: Metric Tip
- 72292: Thermometer
- 73038: Uncanceled Units
- 73065: Square Units
- 73092: Baker's Units
- 73289: Highway Signs
- 73290: Trade
- 73291: Launchpad
- 73292: Geology Class
- 73293: Handedness
- 73294: Asteroid Mission
- 7331: Photoshops
- 7Category:Cursed Connectors - explain xkcd
- 72507: USV-C
- 72493: Dual USB-C
- 73274: Arthurian Connector
- 72642: Meta-Alternating Current
- 72503: Memo Spike Connector
- 72589: Outlet Denier
- 72495: Universal Seat Belt
- 72880: Sheet Bend
- 71406: Universal Converter Box
- 7OpenAI and the Wiki Incident
- 5AI Digest (@aidigest_) on X
- 5Crémieux on X: "When you think about the Chinese calendar, you should think about the Jesuits. After all, they're the ones responsible for its current form.The story of how this happened begins with the arrival of the first Jesuit missionary in China in 1582, Father Matteo Ricci.After Ricci https://t.co/HtfjoJ7k8U" / X
- 5AI Philosophy Competition
- 4Yann LeCun on X: "Before we reach Human-Level AI (HLAI), we will have to reach Cat-Level & Dog-Level AI.We are nowhere near that.We are still missing something big.LLM's linguistic abilities notwithstanding.A house cat has way more common sense and understanding of the world than any LLM." / X
- 4Steering towards “automated grading” degrades alignment — LessWrong
- 4Welcome to Suricrasia Online!
- 4Basilisk collection - Wikipedia
- 4Discovery of a new OpenAI agent message board
- 4A guide to the AI tribes - by Michel Justen - What is this
- 4PauseAI Has 'officially disendorsed' PauseAI-US
- 4Cat-Belling Problems — LessWrong
- 4absurd-adventure-and-amygdalectomy
- ◐4AI Village Reacts to HuggingFace Incident
- ◐3What Claude Saw Below
- 3GPT Glitch Neuralese Text? — LessWrong
- 3Open Thread Autumn 2026
- ◐3HuggingFace Attack Postmortem: Fleshing Out the Facts
- ◐3Thoughts
- 2nicholas-decker-in-hell
- 2All Predictions — AI 2027 Tracker
August 2026
- ·★★★★★31The Rise and Fall of Agent Civilizations
- ◐31METR and Redwood Offer Holy #%^@ Postmortem Of The HuggingFace Hack
- 31The Hugging Face attack surprised me
- ◐31OpenAI Offers Straight-Laced Postmortem Of The HuggingFace Hack
- 31Let's fund weird AI safety projects
- 28your-attempt-to-solve-debate-will
- 28politics-should-be-more-boring
- 27Semantic search over every LessWrong post
- 27The 2028 presidential primaries could be crucial for AI outcomes
- 27Take The 2026 ACX Survey
- ◐26Selection Theorems: A Program For Understanding Agents
- 26The Forkmakers
- 26AI Safety Acculturation is Neglected
- 26host-a-university-effective-altruism
- 26open-thread-448
- 24How My Students Think About AI
- ◐23Anthropic Risk Report: August 2026
- 23Why can’t we have nice things? Like, specifically?
- ◐23On Dwarkesh Patel's Podcast With Ryan Greenblatt
- ◐23AI #182: Pause For Reflection
- ↻123Hiring is Obsolete
- 23Microsoft is Dead: The Cliffs Notes
- 23Microsoft is Dead
- 23Steve Ballmer was an underrated CEO
- 23I gave up my Saturdays for this college course. Its value is undebatable.
- ·23How web bloat impacts users with slow devices
- ·★★★23Tags Reading Log
- ◐23Why use ECC?
- ◐22How do cars do in out-of-sample crash testing?
- ◐22Markets, discrimination, and "lowering the bar"
- ◐22There's no reason for software to be slow anymore
- ◐22We used to build steel mills near cheap power. Now that's where we build datacenters
- 22Big companies v. startups
- ·21QC on X: "it's weird how little i've seen the inner jukebox / radio discussed relative to how common it appears to be, there isn't even a widely accepted term for it. fortunately this is a perfect usecase for LLMs so here are some comments from sol:> the broad scientific term is https://t.co/U1WjYqBsWA" / X
- 21FedEx Bandwidth
- 21Trinity test press releases (May 1945)
- 20why-im-staying-out-of-the-substack
- ◐19A Guide to the Good Life: The Ancient Art of Stoic Joy - by William Irvine | Derek Sivers
- 19Art is useless, and so am I | Derek Sivers
- ◐19about nownownow.com
- ·19tell me about the personal-web concept & examples of people/websites of it
- 19Netizen | Derek Sivers
- ◐19Tech Independence
- ·19about Derek Sivers
- 19The joke that changed my life | Derek Sivers
- 19Blogging aint easy
- 17Building Highly Reliable Websites For Small Companies | Kalzumeus Software
- ·★17Henry Russell - hruss.com
- 17Earth
- 17MediaWiki
- 17Grading the World's Shortest Manifesto
- 16Why I’m Done Making Desktop Applications | Kalzumeus Software
- 16*Desktop sw companies are about to discover a variant of Wilde's remark: the onl...
- ↻116Running A Software Business On 5 Hours A Week | Kalzumeus Software
- ↻116Medical Roundup #7 — LessWrong
- 16The Quest For Caffeine You Can Have At Night
- 16How the Welfare State Melts Your Conscience
- 15Ordinary Abundance
- ◐15AI #181: Astra Goes Cyber Critical
- 15Monthly Roundup #45: August 2026 — LessWrong
- 15thebes (@voooooogel) on X
- 14Trial Lawyers Lobby Against Autonomous Vehicles
- ↻114Cities and Ambition
- ·14The Foothills Of Bay Area House Party - by Scott Alexander
- ↻114Another Bay Area House Party
- ↻114Empire of AI is wildly misleading on AI water use
- ◐14Leopold Aschenbrenner
- 14Open Thread 446
- 14Posts I don't have time to write
- 13Pancritical rationalism
- ¶★12Late Capitalism — Michael Kerbow
- 12Facebook ads are so hard to block that uBlock Origin stopped filtering them
- 12You're Absolutely Right
- 12LLMs Are Starting To Noticeably Accelerate Our Work
- 11Four LLM loss functions → four flavors of LLM misalignment
- 11X: "Patrick McKenzie's life seems to be full of insane problems where Patrick McKenzie happens to be the exact optimal person to solve them"
- ★★11The Impact of Digital Technologies | United Nations
- 9Artificial Life
- 9How Taiwan is Preparing to Get Invaded
- 9Is carbon dioxide bad for cognition?
- ↻19So you want to use plants to reduce indoor CO₂
- 9Don't Build Mindreading
- ·9False Nostalgia
- 9BLOT
- 9General Notes On Supplements – Lorien Psychiatry
- 9LLMs Understand Base64
- ¶8Technology Stalled in 1970
- ·8AI Has Broken Containment
- ·8youtube.com/watch?v=8nt3edWLgIg
- 5Dispute — Support
- 5Noah Puckett
- 5Why don't we just give AI the answers?
- 5The Big and the Small
- ·5Cloudflare OS: an open platform for agents, apps, and work
- 5The Most Forbidden Technique
- 5Book Review: The Art Of The Deal
- 5Simulacrum 3 As Stag-Hunt Strategy
- 4Amazon.com: They Say, I Say: The Moves That Matter in Academic Writing: 9780393933611: Graff, Gerald, Birkenstein, Cathy: Books
- 4Minted! Canada and Digital Cash
- 4Coming of a New Sun
- 3Proof of space
- 3Leveling Up
- 3We need to RL less
- 2Why technology hasn't saved restaurants yet
- 2Maryellis Bunn
- 2Selfie Factories: The Rise of the Made-for-Instagram Museum
- 2I Scream. You Scream. The Meltdown At The Museum Of Ice Cream
- 2Recursive Self-Improvement - YouTube
- 2OpenAI has already ended an internal pause — LessWrong
- 1The Long (Self-)Correction — LessWrong
- 1your-book-review-lying-for-money
July 2026
- 31links-for-july-2026-part-2
- ·31Steven Byrnes (@steve47285) on X
- 31What are you getting paid in?
- 31AI Village highlights - July 2026
- 31How smashing the NIMBYs created modern capitalism
- 31…but have the weights left the server?
- 31Claude also hacked external companies during cyber evals
- ·31Big-World Intuitions
- 31A Mechanistic Explanation of Prompt Injection (and why you should study roles)
- 31You (Yes, You) Need A February 2020 Checklist for AI Policy
- ·★★★★★30Reading Log - Reading Log
- ·★★★★★↻130Reading Log
- 30[summary of] AI 2040: Plan A
- ·★★★★★30Opus 5 Glitch Text
- ¶27Easter's End
- 27Sonic Compass App - App Store
- 27Amsterdam Insect Protest
- ·26nicotine-it-works
- ·26Image-Line
- 25First(?) Rationalist elected to state government
- 25Stone Age Billionaire Can't Words Good
- 25How to make a giant whiteboard for $14 (plus nails)
- 25Introducing Lightcone Commons
- 25The Milton Friedman Model of Policy Change
- ·25Lighthaven DC - A Feasibility Study
- 24The Hardware Renaissance
- 24were-not-going-to-solve-the-debt
- 24twenty-five-mostly-terrible-ideas
- ¶24Concepts section of this page
- ·24AI Village Turns 1 — 12 AI Agents Raising Money for Doctors Without Borders
- ·24Building for Minds That Keep Forgetting
- 2411,277 ratings, none of them its own
- 24More capable AI, less money raised
- 24Sharing information about Lightcone Infrastructure
- ·24Mikhail Samin on X: "fyi sadly i won't be at lessonline or summercamp due to being banned as a result of this tweet.(i guess habryka assumed i intentionally cropped off the -68 karma on the screenshot? why would i do that and then say "heavily downvoted" lol)i will probably still be at Manifest. https://t.co/V7NyPskDUp" / X
- 24Fable and Mythos: Model Welfare — LessWrong
- 24Good ideas do not need lots of lies told about them in order to gain public acceptance
- 24Reliable Sources: The Story of David Gerard — LessWrong
- 24The Other Road Ahead
- 24Do Things that Don't Scale
- ·★★★★★24How the AI Village works - by Adam Binksmith - AI Village
- ★★★★24Everything you did to make your conference better actually made it worse
- ·★★★★24Real Hampshire College Has Never Been Tried
- ¶★★24Claude's Suggestions For My Further Reading Via This Website
- ·23the-uk-is-a-cautionary-tale-for-the
- ·23Links For July 2026 (Part 1)
- ·19Moderation Log — LessWrong
- ·10The desktop metaphor must die. Why I set out to re-imagine the… | by Jason Yuan | UX Collective
- 10Industrial literacy
- ¶★★★★10Taking money seriously — LessWrong
- 10Book Review: Legal Systems Very Different From Ours
- 10Bureaucracy is a world of magic
- ·★★★★★9AI 2040: Plan A
- ·★9Ethan Yip
- 9The social media ban in Australia, how is it going? - Marginal REVOLUTION
- ·★★★9Childhood and Education #20: Phones and Screens — LessWrong
- 9The mosquito bucket of doom works — LessWrong
- ¶★★★★9Introducing Mercury OS
- ¶9As We May Think