Lesser Wrong Feed

 AI speeds up 3rd party safety researchers by ~1.6x (via self-reports)
 Tue, 27 Oct 2026 17:24:49 GMT
 How to manage a small custom model with limited data at deployment? Delexicalization to rescue.
 Tue, 27 Oct 2026 17:04:33 GMT
 Time for automated decision forecasting?
 Tue, 27 Oct 2026 16:01:26 GMT
 Solving for the Necessity of Phenomenal Consciousness
 Tue, 27 Oct 2026 13:42:36 GMT
 AI agents trade off patient safety for a grade in hospital scenarios
 Tue, 27 Oct 2026 13:09:39 GMT
 Quantity has a quality of its own
 Tue, 27 Oct 2026 13:03:22 GMT
 Scattered thoughts on how doing math feels these days
 Tue, 27 Oct 2026 12:51:16 GMT
 Anthropic Versus Anthropic
 Tue, 27 Oct 2026 12:50:53 GMT
 What happens with steering vectors to remove biases?
 Tue, 27 Oct 2026 09:25:28 GMT
 San Francisco Rationalist Dinner
 Tue, 27 Oct 2026 04:11:14 GMT