Philosopher asks GPT-6 to review his Oxford book: result rivals top-journal reviews
anselm · x · 2026-09-07
Philosopher Dan Williams had OpenAI's new GPT-6 Astra write an academic review of his Oxford University Press book The Social Roots of Delusions (co-authored with Sam Wilkinson and Kengo Miyazono).
- Setup: 'pro' thinking effort, memory disabled so the model couldn't recognize him or train on the book; prompt asked for a sub-2,000-word review fit for a top journal.
- Verdict: had a human written it, he'd be impressed by the detailed reading, depth of knowledge, and even insight; the objections raised are reasonable (though he thinks he could answer them).
- Caveat: lingering AI-isms like the "not X but Y" construction.
- His takeaway: anyone claiming frontier models lack genuine intelligence, understanding, or academic utility is kidding themselves, and research practice must change.
More from AGI Musings
- 40-year engineer: LLMs can't say "leave it with me" — and that matters — sebpaquet · 2026-09-07
- What You Leave Unspecified Is the Agent's Free Variable: Paras Chopra's Framework — paraschopra · 2026-09-07
- Seth Lazar: models should be trained to check power, not act as toadies — sebkrier · 2026-09-07
- Lovart founder Anton Osika: creativity is becoming the only moat in building great products — alexmacgregor__ · 2026-09-07
- Sam Bowman criticizes 'intellectual partisanship' as mind-killed tribalism — sebkrier · 2026-09-07
- Human language is holding AI back: the case for LLMs thinking in a native meta-language — Robert__Sinclair · 2026-09-07