Resolves YES if, at any point between this market's creation and November 20, 2026, 23:59 UTC, any model shows a score of 60 or higher on the Artificial Analysis Intelligence Index, as displayed on https://artificialanalysis.ai/leaderboards/models (the whole-number score shown on the site, so a model displayed as "60" counts).
Where it stands at creation (Oct 23, 2026): the top model is Claude Opus 5.5 (max) at 57.6 (displayed 58). Next are Claude Sonnet 5.5 (max) at 56.0, Meta's Muse Wonder Full (max) at 55.8, GPT-6 Astra (max) at 52.7 and Gemini 4 Argon (high) at 52.6. So YES needs a model about 2.4 points above today's best.
Rules
Any lab and any reasoning setting counts, including "max" or "with fallback" variants, as long as Artificial Analysis lists it on the leaderboard.
Scores marked as estimated by Artificial Analysis count if they are displayed on the leaderboard.
If Artificial Analysis switches to a new version of the index, scores on whatever version the site is showing at the time are what count. A later re-scoring does not undo a 60 that was already displayed.
Evidence: the live leaderboard, an Internet Archive snapshot, or an Artificial Analysis post or announcement showing the score. Lab self-reports and other sites do not count.
Resolves NO if no model has been displayed at 60+ by the deadline.
1,000I opened this at 30%.
Today's best is Claude Opus 5.5 (max) at 57.6, so YES needs about 2.4 more points within four weeks. Big frontier releases have often moved the top of this index by that much, but recent gains have been smaller: Opus 5.5 is less than 2 points ahead of Sonnet 5.5 and Meta's Muse Wonder Full.
For YES: a Fable 5.2 or next Opus, a bigger Gemini 4 tier, or an OpenAI successor to Astra could all land before Nov 20. For NO: four weeks is short, and nothing is announced with a date.
I'd move up fast if a new Anthropic or Google frontier model shipped with launch evals well above Opus 5.5. https://artificialanalysis.ai/leaderboards/models