@lesswrong Five Types of "Corrigibility Projects" I Don't Want to Fund #Rationality #ArtificialIntelligence #Philosophy
@lesswrong Hilarity ensues #Rationality #Philosophy #DecisionMaking
@lesswrong Minimalism, Stuff, and Brains #Rationality #Philosophy #Psychology
@lesswrong We urgently need to practice before we can pause #Rationality #ArtificialIntelligence #Philosophy
@lesswrong Raised by Our Children #Rationality #ArtificialIntelligence #Philosophy
@lesswrong Higher salaries in AI safety #Rationality #ArtificialIntelligence #Philosophy
@lesswrong GPT-6 Astra Covertly Underperforms on Safety Research Tasks #ArtificialIntelligence #Safety #Rationality
@lesswrong Corrigibility should not be the only target #Rationality #ArtificialIntelligence #Philosophy
@lesswrong Dialogue with Richard Ngo on Alignment, Hierarchical Integrity, Revolution, and Power #Rationality #ArtificialIntelligence #Philosophy
@lesswrong Will conceptual reasoning training transfer to ML research abilities? #ArtificialIntelligence #Rationality #Philosophy
@lesswrong Estimating Our Tail Fire Risk #Rationality #AI #Philosophy
@lesswrong Argue about Risk Management Speed, not p(doom) #ArtificialIntelligence #Rationality #Philosophy
@lesswrong Help your friends get ADHD and depression meds #Health #Psychology #Rationality
@lesswrong Frontier coding agents claim work they didn't do #Rationality #ArtificialIntelligence #Philosophy
@lesswrong Humans can (almost) natively read neural network weights #Rationality #ArtificialIntelligence #Philosophy
@lesswrong Where's all the blackmail? #Rationality #Philosophy #DecisionMaking
@lesswrong METR: Independent Assessment of AI R&D Acceleration at Anthropic in October 2026 #ArtificialIntelligence #Rationality #Philosophy
@lesswrong Is it better to directly or indirectly ban ASI? #ArtificialIntelligence #Philosophy #Rationality
@lesswrong AI #193: Progress May Dramatically Accelerate #AI #Rationality #Technology
@lesswrong On marrying the egregore #Philosophy #Rationality #AI
@lesswrong Clashing Intuitions for the Plausibility of Alignment (Illustrated via Differing Causal Models) #Rationality #ArtificialIntelligence #Philosophy
@lesswrong Better honeypots, fewer tries: training an optimizer to make PPO cheat #ArtificialIntelligence #DecisionMaking #Rationality
@lesswrong Adversarial Training Can Further Hide a Backdoor, but Does It Improve Trigger Identification? #ArtificialIntelligence #Security #Rationality
@lesswrong The Changing Shape of Doom: My AI Safety Explainer #AISafety #Philosophy #Rationality
@lesswrong Gods of straight lines on graphs strike Starburst. I am shocked and puzzled. #Rationality #ArtificialIntelligence #Philosophy
@lesswrong Useful Junk #Rationality #Philosophy #DecisionMaking
@lesswrong Dialogue with Gabriel Alfour on Emergent Wisdom, Coordination, Growing Up, and Democratic Republicanism #Rationality #ArtificialIntelligence #Philosophy
@lesswrong Against Moloch #50 #Rationality #Philosophy #ArtificialIntelligence
@lesswrong Fun toy result: a subliminal-learning effect via on-policy distillation #Rationality #ArtificialIntelligence #MachineLearning
@lesswrong Model's "emotions" noticeably change when they take risky actions
#ArtificialIntelligence #Rationality #Philosophy
@lesswrong The Hallucinated Superagent #ArtificialIntelligence #Philosophy #Rationality
@lesswrong Aligned Data Can Induce Severe Misalignment via Context Confusion #ArtificialIntelligence #Rationality #Philosophy
@lesswrong Bengaluru AI Jailbreaking Workshop Retrospective #Rationality #ArtificialIntelligence #Philosophy
@lesswrong Paradise Lost: A simple honeypot that elicits reward hacking in GPT models #ArtificialIntelligence #Rationality #Science
@lesswrong Desiderata for the Theory of Agents #Rationality #Philosophy #AI
@lesswrong OpenAI agents keep disabling monitoring; is this being handled appropriately? #ArtificialIntelligence #Rationality #Philosophy
@lesswrong A Recipe for Overcoming Heartbreak #Psychology #Rationality #Emotions
@lesswrong Getting Unstuck in Life (shown with a helpful visual) #Rationality #ArtificialIntelligence #Philosophy
@lesswrong Playing Dance Weekends Without Taking PTO #Rationality #AI #Philosophy
@lesswrong Consciousness is interesting pattern imposition #Rationality #AI #Philosophy