DayNews.aiWeb UI agents still far behind humans
A new benchmark reveals that AI agents struggle with everyday web UI tasks, taking 3.7x longer than humans and varying wildly based on how they perceive the screen.
DayNews.aiDayNews.aiDayNews.aiA new benchmark reveals that AI agents struggle with everyday web UI tasks, taking 3.7x longer than humans and varying wildly based on how they perceive the screen.