Looking back · SecondSource deep dive of October 26

What drives progress in AI that operates computers: on the first OSWorld test, Sonnet 4.6 has reached 72.5%, level with the 72.36% human baseline, while on the longer-workflow OSWorld 2.0 the best score in a June paper this year is only 20.6%.

Concept illustration in two side-by-side panels. On the left, under the large heading "OSWORLD", a framed window shows a person seen from behind sitting in a brown chair at the left, with a counter holding a stack of papers in a tray, a card lettered "SONNET 4.6" and a second stack of papers in a tray; on the right, an open book topped by a black band lettered "OSWORLD 2.0" shows a tall stack of books on its left page, a card lettered "CLAUDE OPUS 4.8" in the middle, and two red books in a tray in front of a pale pink rectangle on the right page. Printed on the image: 20.6% OSWorld 2.0, top score in a June paper.

Which computer-use work gets automated first depends on whether it can be turned into a test environment that can be reset and rerun repeatedly, and the scores are leaderboard or paper figures, not SecondSource measurements. OSWorld 2.0 counts a task only if the whole task is finished, and the top score in the June paper belongs to Claude Opus 4.8. The two versions differ in models, measurement dates and step settings, so each score can only be read against its own human baseline. SecondSource's bet is that OSWorld 2.0 will climb the path OSWorld took: the official leaderboard's top completion rate reaches 60% or more before June 30, 2027. If it is still below 60% by then, task length matters more than rerunnability in setting the pace of progress. (card). How we checked →

Source and checks

This item ran in the brief of Nov 7, 2026, alongside 13 other items. How we checked it, the quote checks and what was still missing that day are on that issue's How we checked page.

Cite this item

SecondSource (2026). "What drives progress in AI that operates computers: on the first OSWorld test, Sonnet 4.6 has reached 72.5%, level with the 72.36% human baseline, while on the longer-workflow OSWorld 2.0 the best score in a June paper this year is only 20.6%.." https://secondsource.io/en/news/2026-11-07-5/. Accessed access date.