Submissions from lesswrong.com

		To the Polypropylene Makers (lesswrong.com)
		87 points by raldi 1 day ago \| past \| 24 comments
		Sacred Values of Future AIs (lesswrong.com)
		1 point by gmays 3 days ago \| past \| discuss
		Refusal in LLMs is mediated by a single direction (lesswrong.com)
		2 points by rzk 4 days ago \| past \| discuss
		Models have some pretty funny attractor states (lesswrong.com)
		3 points by debesyla 4 days ago \| past \| discuss
		Canada Lost Its Measles Elimination Status Because Few Nurses Speak Low German (lesswrong.com)
		4 points by surprisetalk 6 days ago \| past \| 2 comments
		AI found 12 OpenSSL zero-days (lesswrong.com)
		24 points by theptip 9 days ago \| past \| 1 comment
		Are there lessons from high-reliability engineering for AGI safety? (lesswrong.com)
		1 point by Gathering6678 10 days ago \| past \| discuss
		Responsible Scaling Policy v3 (lesswrong.com)
		1 point by ndr 10 days ago \| past \| discuss
		Great Mathematicians on Math Competitions(2010) (lesswrong.com)
		1 point by o4c 12 days ago \| past \| discuss
		Life at the Frontlines of Demographic Collapse (lesswrong.com)
		4 points by reducesuffering 13 days ago \| past \| 1 comment
		"Pinky Promise Diplomacy" Once Stopped a War in the Middle East (lesswrong.com)
		2 points by positivesum 14 days ago \| past
		Childhoods of Exceptional People (2023) (lesswrong.com)
		1 point by Kinrany 17 days ago \| past
		AI found 12 of 12 OpenSSL zero-days (lesswrong.com)
		8 points by AndrewDucker 17 days ago \| past \| 8 comments
		My journey to the microwave alternate timeline (lesswrong.com)
		385 points by jstanley 18 days ago \| past \| 191 comments
		Notes on International Klein Blue (lesswrong.com)
		4 points by mhb 19 days ago \| past
		Why I'm Worried About Job Loss and Thoughts on Comparative Advantage (lesswrong.com)
		85 points by cubefox 19 days ago \| past \| 124 comments
		Spectral Signatures of Gradual Disempowerment (lesswrong.com)
		1 point by OgsyedIE 25 days ago \| past
		To be well-calibrated is to be punctual (lesswrong.com)
		2 points by surprisetalk 26 days ago \| past
		Google Translate apparently vulnerable to prompt injection (lesswrong.com)
		59 points by julkali 29 days ago \| past \| 3 comments
		AI found 12 of 12 OpenSSL zero-days (lesswrong.com)
		2 points by greedo 31 days ago \| past \| 1 comment
		52.5% of Moltbook posts show desire for self-improvement (lesswrong.com)
		1 point by one-2 35 days ago \| past
		Are We in a Continual Learning Overhang? (lesswrong.com)
		2 points by cubefox 37 days ago \| past
		AI found 12 of 12 OpenSSL zero-days (lesswrong.com)
		2 points by jelsisi 37 days ago \| past
		A Simple Method for Accelerating Grokking (lesswrong.com)
		2 points by vuciv 38 days ago \| past
		Test your interpretability techniques by de-censoring Chinese models (lesswrong.com)
		2 points by allenleee 39 days ago \| past
		How AI is learning to think in secret (lesswrong.com)
		2 points by jstanley 40 days ago \| past \| 1 comment
		AI discovers 12 of 12 OpenSSL zero-days (while curl cancelled its bug bounty) (lesswrong.com)
		7 points by excited-dev-11 40 days ago \| past \| 1 comment
		Good if make prior after data instead of before (lesswrong.com)
		13 points by surprisetalk 40 days ago \| past \| 9 comments
		Does Pentagon Pizza Theory Work? (lesswrong.com)
		3 points by nreece 42 days ago \| past
		How AI Is Learning to Think in Secret (lesswrong.com)
		3 points by mannykannot 43 days ago \| past \| 1 comment
		More