@lesswrong Disproving the SIA disproves the Fine-Tuning Argument and Boltzmann-brain-based arguments #Philosophy #Rationality #Science
@lesswrong The Alignment Problem: Computer Science Meets Analytic Philosophy #AI #Philosophy #Rationality
@lesswrong The Most Important Political Dimension is Certainty #Politics #Rationality #Philosophy
@lesswrong Thinking Too Much is Bad for Society #Rationality #Philosophy #Society
@lesswrong Knowingly Wrong #Rationality #DecisionMaking #AI
@lesswrong Spontaneous verbal evidence synthesis in LLMs #AI #LLM #Rationality
@lesswrong Why I care about imperfect-recall games #GameTheory #AI #Rationality
@lesswrong Deconfusing Corrigibility #AI #Philosophy #Rationality
@lesswrong Projection sampling preserves capabilities but diminishes gains from alignment fine-tuning #AI #Rationality #MachineLearning
@lesswrong AI by Default #AI #Rationality #Philosophy
@lesswrong I Did This Without Your Knowledge #Rationality #DecisionMaking #Psychology
@lesswrong Control questions matter: introspection, refusal ablation, and gates in open Gemma models #AI #Rationality #Science
@lesswrong How it feels talking to a hedonistic utilitarian #Philosophy #Ethics #Rationality
@lesswrong Toward scalable capability evaluations #AI #Rationality #Research
@lesswrong When should an AI assistant push back? Measuring warnings and unsolicited opinions on decisions you've already made #AI #Rationality #DecisionMaking
@lesswrong Alignment is Missing a Causal Account #AI #Philosophy #Rationality
@lesswrong Uncertainty in the Verbalizable Workspace #Rationality #Cognition #AI
@lesswrong Neural networks are not already clean #AI #MachineLearning #Rationality
@lesswrong Character Training for Risk-Averse Agents #AI #Rationality #DecisionMaking
@lesswrong P(Anyone Builds It & Everyone Dies) = 0.62? #AI #Rationality #Philosophy
@lesswrong Monthly Roundup #47: October 2026 #Rationality #AI #Community
@lesswrong Tips for Multiagent Work with Claude Opus 5.5 #AI #Multiagent #Rationality
@lesswrong Astra and Sol can take low probability actions with no CoT #AI #Rationality #DecisionMaking
@lesswrong How can LLM RL Work Despite Information-Theoretic Inefficiency #AI #MachineLearning #Rationality
@lesswrong Bad Guy Tokens #AI #Philosophy #Rationality
@lesswrong Exploring Intrapersonal Game Theory #GameTheory #Rationality #DecisionMaking
@lesswrong Three perspectives on corrigibility - shards, simulators, singular learning #AI #Rationality #Philosophy
@lesswrong Mental Health Navigator site for EA and AI safety #AI #Philosophy #Rationality
@lesswrong Staying sane while trying to save the world: Highlights from Spencer Greenberg and Luisa Rodriguez #Rationality #AI #Philosophy
@lesswrong AI #190: Scion of the Times #AI #Rationality #Philosophy
@lesswrong Anthropic's unreleased model submits visa applications #AI #Rationality #Future
@lesswrong Which AI safety organization does what? Puzzle/party games #AISafety #Rationality #AI
@lesswrong Predictor and picker - convergent architectural features of intelligent agents #AI #Rationality #Intelligence
@lesswrong Children of the Cloud: Noah's Ark as an Alignment Song #AI #Philosophy #Rationality
@lesswrong Linkpost: Neural Networks Are Already Clean #AI #MachineLearning #Rationality
@lesswrong Preface to the Computational Agent Theory and ASI Alignment Sequence #AI #Rationality #Philosophy
@lesswrong Reciprocity #Rationality #Philosophy #SocialScience
@lesswrong GLM-5.3 Sometimes Considers How Train-Time Actions Shape Future Behavior #AI #MachineLearning #Rationality
@lesswrong someone's agents are still "hacking" #AI #Hacking #Rationality
@astral_codex_ten AI Safety Is Not A Corporate Plot #AISafety #AI #Rationality