Toby Ord: Gemini 4 Safety Report discloses DeepMind's internal AI-for-AI-R&D efficiency data
tobyordoxford · x · 2026-10-22
Discussing what METR's time-horizon task suite implies about slowing AI progress, Oxford philosopher Toby Ord noted that Google DeepMind ran a very similar analysis on its internal use of AI for AI R&D and published the graph on page 78 of the Gemini 4 Safety Report.
He argues this can be treated as real evidence suggesting that—if real AI R&D work is as hard as METR's task suite—we may be about 14 months behind where one might have expected. But the finding was surprising enough that he thinks it deserves sanity-checking.
More from AGI Musings
- ChatGPT, Claude and Grok all answer p(doom) with ~10%, an odd cross-model consensus — maksym_andr · 2026-10-22
- Will math papers admit "the author cannot verify these details" in acknowledgements? — pickover · 2026-10-22
- What MLB's ABS challenge system and NBA's sensor ball teach us about handing decisions to machines — lmoroney · 2026-10-22
- Dario Amodei's stark take: AI threatens identity and the tax base, not just wages — coherence · 2026-10-22
- Wiblin's plea to AI lab staff: reassess now whether your work delays or accelerates RSI — geoffreyirving · 2026-10-22
- Competition-driven AI progress could make science uncontrollable and incomprehensible — ShahabBakht · 2026-10-22