Ari
Head of product and engineering for OpenAI's computer-use agent.
Timeline (last 90 days)
Early computer-operating agent models could reliably start tasks, but then often got stuck on issues such as pop-ups, loading delays, or page structure changes.
In a task involving subscribing to a customized meal service, manual operation takes two hours, while Computer Use on GPT-6.1 Sol completes it in 15 minutes, about eight times faster than manual work.
On the security level, Ari recommends that developers grant only the website and app access permissions necessary for the task, request user confirmation before performing critical operations such as payments, and gradually build user trust through reliability design.
All Facts (3)
Early computer-operating agent models could reliably start tasks, but then often got stuck on issues such as pop-ups, loading delays, or page structure changes.
50%UnverifiedIn a task involving subscribing to a customized meal service, manual operation takes two hours, while Computer Use on GPT-6.1 Sol completes it in 15 minutes, about eight times faster than manual work.
50%UnverifiedOn the security level, Ari recommends that developers grant only the website and app access permissions necessary for the task, request user confirmation before performing critical operations such as payments, and gradually build user trust through reliability design.
50%