Reading view: this page was generated at build time from one reviewed JSON file. Enable JavaScript for search, filters, sorting, and pagination.
Partially achieved — as of Nov 08, 2026, only part of this task has been demonstrated. A partial demonstration is not a completed result, and key steps still need a human.
Dernière révision Nov 08, 2026
01 / THE SHORT ANSWER
« Evidence published » est la date de publication de la source publique, jamais la date d'une première réalisation. « As of » correspond à la dernière revue manuelle de cette fiche.
02 / THE TASK
AI Index reports 50.6% top-model accuracy on ClockBench, versus 90.1% for humans.
AI Index reports 50.6% top-model accuracy on ClockBench, versus 90.1% for humans.
03 / STATUS
Démonstration partielle, pas un résultat complet.
Frontier multimodal models demonstrates part of this task, but key steps still require a human. No evidence publication date is recorded for a completed result.
Only part of the task is demonstrated; key steps still require a human.
04 / EVIDENCE
For Not Yet entries, a source may describe a limitation or an evaluation rather than a breakthrough. Links are editorial references, not endorsements.