@lesswrong Mech Interp is a La Croix of Actual Safety Research #AI #Rationality #Philosophy
@lesswrong Relevance Realization as Human Comparative Advantage #Rationality #Cognition #AI
@lesswrong Control monitors need alignment assessments too #AI #Rationality #Alignment
@lesswrong Epistemic mapping 101: what to ask in order to understand how others think #Rationality #Cognition #Philosophy
@lesswrong Slower Animated Diagrams #AI #Rationality #Science
@lesswrong Fermi Poker #Rationality #AI #Philosophy
@lesswrong LM self-models are built from indirect sources #AI #MachineLearning #Rationality
@lesswrong Patching reward hacks won't fix reward hacking #AI #Rationality #Ethics
@lesswrong LoReFT GRAFT Better at Preventing RL Grift #AI #Rationality #MachineLearning
@lesswrong Simulated Control Evaluations #AI #Rationality #Science
@lesswrong The Hackening #AI #Rationality #Philosophy
@lesswrong Power -- a guide for the perplexed #Rationality #Philosophy #Power
@lesswrong The Evil is where you aren't looking #Rationality #Philosophy #AI
@lesswrong Gnerk Asks How Big a Swarm Can Get Before It Stops Being One Agent #AI #Philosophy #Rationality
@lesswrong Instrumental convergence is incoherent (so pls slightly lower your p(extinction)) #AI #Rationality #Philosophy
@lesswrong Highlights from Garrison Lovely's "Obsolete" #Rationality #AI #Philosophy
@lesswrong The full context of the AI agent quote at the end of AI Gone Wild on Netflix #AI #Rationality #Philosophy
@lesswrong Suddenly, Unclean #Rationality #AI #Philosophy
@astral_codex_ten San Francisco Meetup This Saturday #Community #Meetup #Rationality
@clearer_thinking Should you steelman people you disagree with? #Rationality #CognitiveBiases #Philosophy
@lesswrong Greater good arguments in AI safety #AISafety #Rationality #Philosophy
@lesswrong AI #191: The Hackening Begins #AI #Rationality #FutureTech
@lesswrong The Alignment Journal: Our Approach and Related Projects #AI #Rationality #Philosophy
@lesswrong Monitorability evaluations and policy proposals #AI #Policy #Rationality
@lesswrong The Relief Paradox: How We Let AI 'Undermine Our Autonomy' #AI #Autonomy #Rationality
@effective_altruism_forum 73 Credits Is Honest. 80 Is a Gamble. #EffectiveAltruism #Rationality #DecisionMaking
@lesswrong Frontier models may not be reliable auditors #AI #Rationality #Philosophy
@lesswrong What's in CODI's latent recurrent thoughts? #AI #Rationality #Philosophy
@lesswrong Misaligned! #AI #Rationality #Philosophy
@lesswrong Model whisperers #AI #MachineLearning #Rationality
@lesswrong D̶o̶n̶'̶t̵⠀Can't Look Up #Rationality #AI #Philosophy
@lesswrong AI Needs A Culture #AI #Culture #Rationality
@lesswrong AM-Bench: A theoretically-grounded benchmark for agentic misalignment #AI #Rationality #Benchmark
@lesswrong On Idealized Agency #Rationality #AI #Philosophy
@lesswrong A general approach for alignment generalization and a Stupid New Technique that beat everything #AI #Rationality #MachineLearning
@lesswrong Map of the Shallow Review of Technical AI Safety #AISafety #Rationality #AI
@lesswrong Neural nets are exceptionally vulnerable to goal self-hack #AI #MachineLearning #Rationality
@lesswrong Against Moloch #48 #Rationality #AI #Philosophy
@lesswrong A God-Fearing AI #AI #Philosophy #Rationality
@lesswrong Linkpost: Can a million AI agents create superintelligence? #AI #Superintelligence #Rationality