NLP researcher pushes back on viral claim Claude found 872 errors in astrophysics papers
gregd_nlp · x · 2026-11-20
A viral post claimed Claude Opus 5.5 identified 872 major errors in published astrophysics papers. NLP researcher Greg Durrett responds with 'In defense of astrophysics,' arguing that sloppy AI use and its erroneous results are eroding trust in an entire scientific field — a worrying trend for AI-assisted peer critique.
Related event: Researchers Push Back Against Using LLMs to Audit Scientific Literature(5 posts)→
More from AGI Musings
- Recruiter: AI familiarity is now the minimum bar for any good tech job — yenkel · 2026-11-20
- Reading AI outputs is talking to yourself: anthropomorphism is a cognitive bias — gerardsans · 2026-11-20
- Kokotajlo: if pacing efforts fail, Anthropic would keep racing to superintelligence — DKokotajlo · 2026-11-20
- Lance Fortnow: AI is muddying how academia judges papers, hiring and tenure — fortnow · 2026-11-20
- AI consciousness is about what AI is, not what it does: LLMs just sample text — gerardsans · 2026-11-20
- Anthropic's 2028 warning: AI may transform before we protect human agency — Graham_dePenros · 2026-11-20