OpenAI has launched Dots, personal AI agents powered by its GPT-6 Astra model that keep working on a user's behalf around the clock. The company announced them at its DevDay event and began rolling them out in ChatGPT the same day. Dots looks to rival Grok Bots that was unveiled a few weeks ago by SpacexAI.
What is Dots?
OpenAI describes Dots as agents built to handle everything. Each runs on its own cloud computer, with a browser, and pursues user-set goals in the background with little oversight. It also learns a user's preferences from feedback over time. Users can launch a Dot from ChatGPT or Codex, OpenAI's coding agent.
Users start with a primary Dot that they name and personalise. OpenAI says it expects teams of Dots to work together for a user eventually. At launch each user gets one Dot, and multiple agents are planned for later.
What they can do?
Dots connect to more than 4,000 apps through OpenAI's plugin ecosystem. Examples OpenAI has given include turning recurring customer feedback into tested pull requests, revising launch material when a project's scope changes, and rerunning a scientist's analyses as new data arrives. Users choose which apps a dot can access and can set rules for when it acts alone and when it must ask for approval. Voice calls work at launch, but a dot cannot start a call to the user, and texting is planned for later.
Who gets it?
Dots are available to Pro and Business Premium users in eligible markets, with a gradual rollout. A beta is planned for admin-enabled Enterprise workspaces. Dots usage won't count toward plan allowances for the first month, and OpenAI says it will share usage terms for each plan afterward.
Specialist Dots for companies
OpenAI is also previewing specialist Dots that take on defined jobs inside organisations, with their own identity and credentials. Early internal testing covers procurement, invoice processing, customer support and contracting. OpenAI says it is working with Microsoft to bring them into Agent 365.
