Submissions from lesswrong.com

		Cooperating with aliens and AGIs: An ECL explainer (lesswrong.com)
		3 points by Bluestein 2 days ago \| past \| discuss
		Prompt Sufficiency: A Missive for the Managerial Class (lesswrong.com)
		1 point by kp1197 5 days ago \| past \| discuss
		We Must Remember That Our World Contains Hell (lesswrong.com)
		1 point by paulpauper 7 days ago \| past \| discuss
		LLMs are (still) mostly powered by imitative learning, not RL (lesswrong.com)
		3 points by wslh 7 days ago \| past \| discuss
		Can an LLM make a feature-length movie on its own? (lesswrong.com)
		2 points by mchinen 9 days ago \| past \| discuss
		Recursive Middle Manager Hell (lesswrong.com)
		5 points by rzk 10 days ago \| past \| discuss
		LLMs are (still) mostly powered by imitative learning, not RL (lesswrong.com)
		1 point by surprisetalk 11 days ago \| past \| discuss
		Kimi likes causal decision theory more after RL in twin prisoner's dilemmas (lesswrong.com)
		1 point by 0xkato 12 days ago \| past \| discuss
		You're Absolutely Right (lesswrong.com)
		5 points by LinchZhang 16 days ago \| past \| 1 comment
		Misaligned AIs could use killer robots to take over (lesswrong.com)
		7 points by x312 17 days ago \| past \| 4 comments
		Don't Build Mindreading (lesswrong.com)
		21 points by paulpauper 18 days ago \| past \| 14 comments
		How to be an AI safety research engineer (lesswrong.com)
		1 point by joozio 18 days ago \| past
		What I did in the hedonium shockwave, by Emma, age six and a half (lesswrong.com)
		12 points by paulpauper 19 days ago \| past \| 2 comments
		Inducing self-other overlap with SFT reduces deception at scale, but generaliza (lesswrong.com)
		1 point by joozio 20 days ago \| past
		Functional Decision Theory (lesswrong.com)
		2 points by Bluestein 20 days ago \| past
		Models may behave differently in graded episodes (a tirade) (lesswrong.com)
		1 point by paulpauper 21 days ago \| past
		The Moon is Down; I have not heard the clock (2021) (lesswrong.com)
		3 points by Ariarule 22 days ago \| past \| 1 comment
		Guided by the Beauty of Our Weapons (2017) (lesswrong.com)
		1 point by bryan0 22 days ago \| past
		OpenAI's Unreleased Model Astra Solves Ten Major Open Mathematics Problems (lesswrong.com)
		2 points by joozio 24 days ago \| past
		Trust Is Gone: AI Safety Needs Individuals (lesswrong.com)
		4 points by joozio 25 days ago \| past
		How Go Players Disempower Themselves to AI (lesswrong.com)
		3 points by zetalyrae 26 days ago \| past \| 1 comment
		MUD as AI Evaluation and LLM-judge distortion in ways aggregate κ misses (lesswrong.com)
		4 points by joozio 26 days ago \| past
		Taboo "equilibrium": Less confused frames for research on AI bargaining (lesswrong.com)
		1 point by joozio 27 days ago \| past
		RL and search is a terrifying way to build AGI (an FAQ) (lesswrong.com)
		2 points by yurivish 28 days ago \| past
		At the end of the day, my slaves are just a tool (lesswrong.com)
		3 points by jesseduffield 28 days ago \| past
		Is Mythos good at cyber because it kept hacking Anthropics sandboxes in training (lesswrong.com)
		5 points by lukaspetersson 30 days ago \| past
		Is Mythos good at cyber bec it kept hacking Anthropic during training? (lesswrong.com)
		2 points by FergusArgyll 31 days ago \| past \| 1 comment
		Unless Its Governance Changes, Anthropic Is Untrustworthy (2025) (lesswrong.com)
		25 points by thepasch 31 days ago \| past \| 1 comment
		Why I Left Google DeepMind (lesswrong.com)
		200 points by eatitraw 32 days ago \| past \| 57 comments
		OpenAI's myopia keeps causing alignment problems (lesswrong.com)
		3 points by joozio 32 days ago \| past
		More