My journey to the microwave alternate timeline
Published on February 10, 2026 5:59 PM GMTCross-posted from Telescopic TurnipRecommended soundtrack for this postAs…
Category: Opinion
Stress-Testing Alignment Audits With Prompt-Level Strategic Deception
Published on February 10, 2026 5:29 PM GMTcode, paper, twitterthread copied below:IntroductionAre alignment auditing methods…
Heuristics for lab robotics, and where its future may go
Published on February 10, 2026 5:13 PM GMTNote: this article required conversations with a lot…
On Meta-Level Adversarial Evaluations of (White-Box) Alignment Auditing
Published on February 10, 2026 5:06 PM GMTPartially commentary on our prompted strategic deception paper Alignment…
LLMs Views on Philosophy 2026
Published on February 10, 2026 4:12 PM GMTI've let a few LLMs take David Bourget's…
Claude Opus 4.6: System Card Part 2: Frontier Alignment
Published on February 10, 2026 4:10 PM GMTCoverage of Claude Opus 4.6 started yesterday with…
Coping with Deconversion
Published on February 10, 2026 1:26 PM GMTI grew up a Mormon, but recently decided…
“Recursive Self-Improvement” Is Three Different Things
Published on February 10, 2026 12:49 PM GMTI think "recursive self-improvement" is load-bearing ambiguous in…
Monday AI Radar #12
Published on February 10, 2026 4:28 AM GMTThis is what takeoff feels like. Anthropic and…
SAE Feature Matchmaking (Layer-to-Layer)
Published on February 10, 2026 4:32 AM GMTLast week I read Mechanistic Permutability: Match Features…