@lesswrong Organic #Rationality #Philosophy #Science
@lesswrong Monitor Evasion Depends on What RL Finds and What Monitors Miss #ArtificialIntelligence #Rationality #Philosophy
@lesswrong Refelting My Keyboard #Rationality #Philosophy #Science
@lesswrong What would you do if you had all the time in the world? #Philosophy #Rationality #AI
@lesswrong Conclusions of my DIY Inkhaven #Rationality #Science #DIY
@lesswrong Explaining "AI might obliterate you" to senior policymakers in India #AI #Philosophy #Rationality
@lesswrong Accidentally Misleading People Should Be Embarrassing; Or, Higher Innocence #Rationality #Philosophy #DecisionMaking
@lesswrong Against SURA #Rationality #ArtificialIntelligence #Philosophy
@lesswrong Legible Networks: recasting neural nets so you can read them exactly [linkpost] #ArtificialIntelligence #Rationality #Science
@lesswrong Generalized belief states in transformers #ArtificialIntelligence #Rationality #Philosophy
@lesswrong What Social Constructs Are (And Are Not) #Philosophy #Rationality #Society
@lesswrong Reviewing my 2017 AI Risk Project #Rationality #ArtificialIntelligence #Philosophy
@lesswrong Low-volume cymbals are underrated #Rationality #Music #Philosophy
@lesswrong John Holt's Escape from Childhood: Summary #Philosophy #Psychology #Rationality
@lesswrong Childhood and Education #22: Voice Without Exit #Rationality #Education #Philosophy
@lesswrong The Advocacy Gap in International AI Verification #Rationality #ArtificialIntelligence #Philosophy
@lesswrong Persona Geology: Underlying Structure in Character Training Adapters #AI #Rationality #MachineLearning
@lesswrong Microdosing the spirit of Anthropic #Rationality #AI #Philosophy
@lesswrong Dice enterprises as imperfect-recall games #Rationality #DecisionMaking #Philosophy
@lesswrong Could a first mover lock everyone out of outer space? #Space #Rationality #Philosophy
@lesswrong The Three Nothings of Existence #Philosophy #Rationality #Science
@lesswrong Lessons From Red-teaming Auto-review, OpenAI's Blocking Monitor #ArtificialIntelligence #Rationality #Safety
@lesswrong Unaligned weak models might exploit powerful aligned models #ArtificialIntelligence #Rationality #Philosophy
@lesswrong Neutral Isn't Independent: LLM @-invocation #Rationality #ArtificialIntelligence #Philosophy
@lesswrong Old talks, 2000 - 2020 period. #Rationality #Philosophy #Science
@lesswrong Walking Partway to School #Rationality #ArtificialIntelligence #Philosophy
@lesswrong Formalising Newcomblike Problems #Rationality #DecisionTheory #Philosophy
@lesswrong Formally Defining Wavefunction Branches #Rationality #Physics #Science
@lesswrong [Paper] Propensities Please: Exposing Alignment-Faking LLMs using Synthetic Document Finetuning #ArtificialIntelligence #Rationality #Philosophy
@lesswrong Probability ranges are a cop-out with 90-93% probability #Rationality #Probability #DecisionMaking
@lesswrong Lie detection is best when it's cheap to cooperate #Rationality #Philosophy #DecisionMaking
@lesswrong Someone Made an Infra-Bayesian RL Agent, and it One-Boxes! #AI #Rationality #Science
@lesswrong The Newly "Needed" Labor: Under AI 2040, economic growth is normal, but prices fall exponentially #ArtificialIntelligence #Economics #Rationality
@astral_codex_ten Links For October 2026 #Rationality #Psychiatry #Statistics
@lesswrong Leaking Muse's Charter, Meta's model spec #Rationality #ArtificialIntelligence #Philosophy
@lesswrong AI #192: Anthropic Is Going Public #ArtificialIntelligence #Rationality #Philosophy
@lesswrong Compensating the victims of automation is economically beneficial #Economics #Automation #Rationality
@lesswrong Dependencies in conversation and in software #Rationality #Software #Philosophy
@lesswrong Everything Rots #Philosophy #Rationality #Science
@lesswrong Borrow My Accordion? #Rationality #Philosophy #DecisionMaking