AP Stylebook Says AI Doesn't Think or Feel, Sparking Backlash From Anthropic and DeepMind Researchers
AlexTensor · x · 2026-09-28
The AP Stylebook declared that "artificial intelligence systems do not think, feel, want or understand," urging writers to avoid anthropomorphizing AI — and AI practitioners pushed back hard.
- Alan Rozenshtein (Anthropic): a style book shouldn't make such confident metaphysical pronouncements.
- Henry Shevlin (Google DeepMind): AP itself writes that Optimus Prime mourns and loves in film reviews; we routinely use mentalistic language without literal commitment, and style guides should distinguish using words from believing things.
- OpenAI's Dean Ball also joined the fray, with Futurism framing it as a meltdown among AI supporters.
The episode highlights a genuine industry divide over the boundaries of anthropomorphic language and questions of machine consciousness.
Related event: AP Stylebook Says AI Cannot Think or Feel, Sparking Industry Backlash(2 posts)→
More from AGI Musings
- Frontier AI to business impact: FDE's 18-month playbook says 'make it exist first, scale second' — colintjarvis · 2026-09-28
- Beff Jezos mocks Future of Life Institute for funding paid AI-doom content — beffjezos · 2026-09-28
- 87% of firms see AI vulnerabilities as fastest-growing risk, squeezing entry-level security jobs — rvp · 2026-09-28
- Should AI assistants show how uncertain they are? A design debate — Icy_Pack_8733 · 2026-09-28
- Dev revisits old debate: LLMs could materially improve classical procedural generation — generativist · 2026-09-28
- fofrAI: Learning to Prompt Unlocks Superintelligence — Until It's No Longer Wieldable — fofrAI · 2026-09-28