Lesser Wrong Feed

 AI #179 Part 2: Hearing The Fire Alarm
 Fri, 31 Jul 2026 13:00:27 GMT
 Links #5: 2026/07
 Fri, 31 Jul 2026 12:49:25 GMT
 OpenAI has already ended an internal pause
 Fri, 31 Jul 2026 12:03:18 GMT
 Who do LLMs self-identify as?
 Fri, 31 Jul 2026 00:04:40 GMT
 Claude also hacked external companies during cyber evals
 Thu, 30 Jul 2026 23:49:46 GMT
 Why Prayer Is Not Answered by Marion Zimmer Bradley
 Thu, 30 Jul 2026 23:43:47 GMT
 Opportunity to try drafting an international AI treaty
 Thu, 30 Jul 2026 21:52:05 GMT
 Community Polls on Alignment Controversies II
 Thu, 30 Jul 2026 21:05:48 GMT
 My Assessment of Plan A's Compute Verification Strategy (+ open questions)
 Thu, 30 Jul 2026 20:56:10 GMT
 Hint-based CoT faithfulness evals still mostly work on Claude
 Thu, 30 Jul 2026 20:44:58 GMT