OpenAI Dots

Also known as: Dots, ChatGPT Dots, dots agent

OpenAI's always-on agents inside ChatGPT, each running on its own cloud computer. You give a Dot a goal, connect it to your apps, and it keeps working toward that goal around the clock, checking in only when it needs your approval.

Launched at OpenAI DevDay 2026 on September 29, Dots are persistent AI agents built into ChatGPT. Unlike a regular chat session that stops when you close the tab, a Dot takes an ongoing goal, hooks into your connected apps (calendar, email, Slack, Teams, and thousands more via ChatGPT plugins), and keeps making progress in the background. It has its own cloud computer and browser to work from.

The interaction model is different from prompting. Instead of asking a question and reading an answer, you assign a responsibility: 'monitor my inbox for anything urgent, draft replies, and flag anything that needs a decision.' The Dot handles the routine work and surfaces only what genuinely needs you. It learns your preferences over time and can proactively gather information before you ask.

At launch, Dots rolled out to ChatGPT Pro and Business Premium subscribers in eligible markets. Each plan includes one Dot at no extra cost. OpenAI has said it envisions teams of Dots working together on a user's behalf, which would shift the product from a chatbot into something closer to a lightweight staffing layer you configure once and supervise over time.

This definition is AI-generated and refreshed weekly. It may contain inaccuracies. Use your own judgment, especially for production decisions.

Related terms