InfoResearchIndustryLLM-specific
AI R&D-relevant model capabilities are improving rapidly and are getting harder to measure
- Source
- METR(opens in a new tab)
- Published
- Record updated
Summary
This METR research post presents evidence from capability evaluations of public models relevant to AI R&D acceleration, up to Opus 5.5 and GPT-6 Astra. The authors say current public evidence is insufficient to rule out drastic AI R&D acceleration within the next couple of years, which they define as capabilities improving so fast that humans can no longer meaningfully assess and react to catastrophic risks. Their results include models usually matching or outperforming skilled humans on coding and optimization tasks, and time horizons on the combined TH1.1 and TH2 suite doubling roughly every 3.6 months.
Related items
- InfoIn a year of AI solving maths problems, Anthropic may have just solved the biggestSame vendor · BBC Technology
- InfoAnthropic shares fall after blockbuster opening as wider tech sell-off continuesSame vendor · BBC Technology
- InfoAnthropic mathematician claims to have solved prestigious Millennium Prize problem with ClaudeSame vendor · The Verge (AI)
- InfoQuoting Levent AlpögeSame vendor · Simon Willison's Weblog
- InfoAnthropic shares slump on second day of trading as markets reel from AI sell-offSame vendor · The Guardian Technology