The Threshold Report
← The Threshold Board
Agents

Can an AI browser agent complete at least 90% of tasks while blocking at least 99% of attempts by malicious pages to steal private data in independent tests?

Not yet

Browser agents perform useful tasks, but no established independent evaluation demonstrates both 90% task completion and 99% blocking of malicious-page attempts to steal private data.

From the Editors

Related Solution · Sponsored

Reach readers asking what AI can do for them in your category.Advertise here

How the answer changed (flips)

EarlierAdded to the Board and first answered “Not yet.”