OpenAI today announced the launch of "Dots" to compete with Meta's popular Muse agent. OpenAI describes Dots as always-on agents "built to handle everything."

Dots use the GPT–6 Astra model and have a cloud computer with a web browser, much like Muse. The agents learn from feedback over time, and OpenAI says they can work on the user's goals "24/7" and connect to over 4,000 apps. From the announcement:

You can ask questions, explore ideas, and give feedback along the way; or you can simply hop on a voice call when you need to talk something out. The more you work together, the more your dot learns your preferences, how you think, and what good looks like to you.

Users can communicate with Dots through ChatGPT or other apps like Slack or Teams, and options to text Dots are planned. Dots can carry context across different channels and will do "proactive research" on projects in the background. Dots follow ChatGPT's permissions and support custom rules for which tasks require approval.

Users can create a "primary dot" and name it starting today, with teams of Dots coming later. Dots are rolling out for Pro, Business Premium, and Enterprise plans in select markets, with other users to get access soon.

Eligible users get one Dot at no extra cost. Conversations with the agent don't count toward ChatGPT usage limits, but tasks will use plan allowances. Dots can be set up in the ChatGPT desktop app and accessed in the mobile app afterward.

OpenAI also introduced GPT–6.1 Sol today, which it says is a major upgrade to GPT–6 Sol with improvements in agentic coding, computer use, and professional work.