Lesser Wrong Feed

 AI Safety at the Frontier: Paper Highlights of December 2025
 Wed, 14 Jan 2026 14:29:34 GMT
 Backyard cat fight shows Schelling points preexist language
 Wed, 14 Jan 2026 14:10:39 GMT
 Parameters Are Like Pixels
 Wed, 14 Jan 2026 13:45:45 GMT
 Apply to Vanessa's mentorship at PIBBSS
 Wed, 14 Jan 2026 09:15:50 GMT
 The Eternal Labyrinth
 Wed, 14 Jan 2026 03:19:01 GMT
 How Much of AI Labs' Research Is Safety?
 Wed, 14 Jan 2026 01:40:30 GMT
 We need to make ourselves people the models can come to with problems
 Wed, 14 Jan 2026 00:43:36 GMT
 Analysing CoT alignment in thinking LLMs with low-dimensional steering
 Tue, 13 Jan 2026 21:20:38 GMT
 Global CoT Analysis: Initial attempts to uncover patterns across many chains of thought
 Tue, 13 Jan 2026 22:53:09 GMT
 Playing Dumb: Detecting Sandbagging in Frontier LLMs via Consistency Checks
 Tue, 13 Jan 2026 19:28:06 GMT