METR TH2 Evaluation Finds AI Time Horizons Doubling While Research Judgment Lags
casrai.org datasets casrai.org
Aurekai listening
Listen to this summary
Original instrumental score; voice stays in front.
Ready to listen
This reads the text available on this page using your browser's voice. It is not a publisher-recorded narration; voice processing depends on your browser.METR’s TH2 evaluation finds frontier AI time horizons doubling every 110 days on verifiable tasks, while open-ended research judgment continues to lag.
More ways to use this
Open on another device
Scan to open this page. Your saved notes and account access are not included.
Zohran Mamdani tears Achilles while playing soccer against NYC press corps
Lifeway Research, says 2/3rd of Churchgoing Christians failed Bible quiz.