Lesser Wrong Feed

 Agents let AI safety share experiments hourly, not just papers monthly
 Thu, 17 Sep 2026 00:39:03 GMT
 Measuring alignment drift via trajectory prefixes
 Thu, 17 Sep 2026 00:37:32 GMT
 Constraining the capacity of physical side channels for AI verification and security
 Thu, 17 Sep 2026 00:30:48 GMT
 Can parts of the HuggingFace incident be simulated?
 Thu, 17 Sep 2026 00:30:33 GMT
 The Latter Days of Magic (1)
 Thu, 17 Sep 2026 00:10:15 GMT
 The Loss of Singularity
 Wed, 16 Sep 2026 23:43:44 GMT
 Reducing the Resource Gap Between Lab and External Safety Researchers
 Wed, 16 Sep 2026 22:51:02 GMT
 Thought anchors don't transfer between models
 Wed, 16 Sep 2026 22:49:42 GMT
 Why is AI so unregulated?
 Wed, 16 Sep 2026 22:39:17 GMT
 The Most Important City in AI Safety May Be Singapore
 Wed, 16 Sep 2026 23:12:05 GMT