What are OpenAI dots?

Updated September 29, 2026

Dots are always-on AI agents from OpenAI. A dot learns your goals and standards, keeps working toward them in the background, and brings you finished work to review, sometimes before you think to ask.

TL;DR
  • Powered by GPT-6 Astra.
  • Each dot has its own cloud computer and browser and can run 24/7.
  • Connects to 4,000+ apps through ChatGPT plugins.
  • Reachable in ChatGPT, Slack and Microsoft Teams, by chat or voice call.
  • Learns your preferences from feedback over time.

How dots differ from ChatGPT

ChatGPT responds when you open a conversation. A dot is persistent: it holds several projects at once, keeps context across every channel you use, and can message you with progress, questions or decisions. You do not need to manage separate threads or direct each step.

What a dot can do

You can open your dot's computer at any time to inspect its work.

Examples

Safety and control

Dots follow built-in rules for when to act and when to ask. You can add Custom Rules to allow, require approval for, or block specific actions, and follow everything, including background work, in Activity View. Sensitive tasks such as changing a password always stay with you. Setting rules →

Specialist dots

For organizations, OpenAI previewed specialist dots with their own identity, credentials and system access for defined roles, such as procurement, invoice processing or customer support. They start as enterprise pilots, and OpenAI is integrating them with Microsoft Agent 365.

Known limitations

OpenAI notes that dots can still make mistakes and that consequential work should be reviewed. An early hands-on review from Every reported permission issues, dropped messages and trouble connecting to the in-app browser.

See pricing and availability

Sources