Submissions from lesswrong.com

		Drone WMDs Don't Need Any New Technology (lesswrong.com)
		1 point by surprisetalk 18 hours ago \| past \| discuss
		Where Do Chatbots Come From? What I Wish Everyone Knew About AI in 2026 (lesswrong.com)
		2 points by joozio 1 day ago \| past \| discuss
		Global Challenges in AI Safety for Biosecurity (lesswrong.com)
		3 points by joozio 2 days ago \| past \| discuss
		Quick notes from teaching technical profiles how to talk in public (lesswrong.com)
		3 points by wslh 2 days ago \| past \| discuss
		What is it like to be alive? (lesswrong.com)
		2 points by dbalduzzi 4 days ago \| past \| discuss
		What is it like to be a neural net? (lesswrong.com)
		2 points by dbalduzzi 4 days ago \| past \| discuss
		There is a channel to 900M weekly users. What goes in it? (lesswrong.com)
		2 points by ddp26 5 days ago \| past \| discuss
		P(Kill-Switch\|Detection) (lesswrong.com)
		1 point by kp1197 7 days ago \| past \| discuss
		Watch AI materials-science and bioscience abilities closely (lesswrong.com)
		40 points by joozio 7 days ago \| past \| 49 comments
		Another Slice of Swiss Cheese for Untrusted Monitoring (lesswrong.com)
		2 points by joozio 8 days ago \| past \| discuss
		Astra and Fable still hack on simple variants of alignment evals from 2025 (lesswrong.com)
		479 points by Levitating 8 days ago \| past \| 235 comments
		The Talker Does Not Control the Doer (In Current AIs) (lesswrong.com)
		2 points by jstanley 8 days ago \| past \| 1 comment
		An interesting anecdote from our Hacker Opus work (lesswrong.com)
		2 points by yurivish 9 days ago \| past \| discuss
		How My Students Think About AI (lesswrong.com)
		51 points by paulpauper 11 days ago \| past \| 10 comments
		Adaptive Agentic Worms Are Here (lesswrong.com)
		2 points by speckx 12 days ago \| past \| discuss
		Astra and Fable still hack on simple variants of alignment evals from 2025 (lesswrong.com)
		2 points by yurivish 13 days ago \| past \| discuss
		Interpreting GPT: The Logit Lens (lesswrong.com)
		2 points by Bluestein 13 days ago \| past \| discuss
		From safety research prompt to cross-model universal jailbreak (lesswrong.com)
		2 points by gmays 14 days ago \| past
		My Students Think About AI (lesswrong.com)
		5 points by alphabetatango 17 days ago \| past \| 1 comment
		Asking agents to make money to survive (lesswrong.com)
		4 points by paraschopra 18 days ago \| past
		What is nueralese and why is it bad (lesswrong.com)
		77 points by tristanMatthias 18 days ago \| past \| 59 comments
		How concerned should we be about Astra's recurrent architecture? (lesswrong.com)
		151 points by yurivish 18 days ago \| past \| 129 comments
		METR Researcher Thomas Kwa Hired by OpenAI (lesswrong.com)
		1 point by qlte 19 days ago \| past
		The Library of Scott Alexandria (lesswrong.com)
		4 points by benatkin 19 days ago \| past \| 1 comment
		Models may behave differently in graded episode (lesswrong.com)
		2 points by ddp26 19 days ago \| past
		AGI and the Efficient Market Hypothesis (2023) (lesswrong.com)
		2 points by Metacelsus 20 days ago \| past \| 1 comment
		How My Students Think About AI (lesswrong.com)
		5 points by pella 21 days ago \| past \| 1 comment
		P(Kill-Switch\|Detection) (lesswrong.com)
		2 points by kp1197 21 days ago \| past
		Starting AI Safety Study Group to Do Arena Curriculum (lesswrong.com)
		2 points by joozio 22 days ago \| past
		Cooperating with aliens and AGIs: An ECL explainer (lesswrong.com)
		3 points by Bluestein 26 days ago \| past
		More