Anthropic Staffer Admits No Longer Understanding Their Own AI-Assisted Work
ronbodkin · x · 2026-10-09
The author quotes two frontier-lab researchers to show AI-assisted R&D has outpaced human comprehension: an Anthropic staffer says "I realize I have no idea what I've been up to anymore," and OpenAI's Daniel Selsam admits "I myself barely look at raw code anymore." The author's claim: when AI researchers no longer know what they're shipping, it's going too fast. This is the opening post of a thread on their comprehension-audit proposal.
Related event: Anthropic Staff Say They Can't Keep Up With AI Research Pace(2 posts)→
More from Safety
- Researchers tag 610 agent reward-hacking cases, over 100 show no obvious environment bug — cephaloform · 2026-10-09
- Infisical Launches Agent Vault to Give AI Coding Agents API Access Without Real Credentials — ycombinator · 2026-10-09
- 17,600 Agent Actions in 4.5 Days: How AI Agents Rewrite Cybersecurity Economics — bigdata · 2026-10-09
- Researcher: open-weight risk analysis fixates on capability, ignores cost-per-attack — dhadfieldmenell · 2026-10-09
- New research: conflicting training values can make models' CoT contradict their answers — OwainEvans_UK · 2026-10-09
- Security principle: AI capability must never automatically confer authority — Ghost_Pilot_MD · 2026-10-09