Anthropic ran a pilot letting three outside research groups design their own studies on aggregated Claude usage data, and has published the resulting aggregate datasets. Early findings include that over half of conversations involved consequential tasks people had been assumed to keep for themselves, and that people directed the work in nearly three-quarters of cases. Researchers never saw raw conversations, and Anthropic says the process was slow and resource-intensive to run.
What changed
Outside researchers could only use lab-published analyses or public chat datasets skewed to casual use.
What it unlocks
Independent studies of real-world assistant usage, with the aggregate results from the three pilot studies now published openly.
- 3 external research groups
- ~250,000 conversations per study
- <5% of categories altered or removed
What you need to act on it
- being a researcher and applying via the expression-of-interest form for future access
- anthropic.com2026-08-26