OpenAI has launched Dots, a new class of always-on AI agents designed to take ongoing responsibility for tasks across connected apps, alongside a new $500-per-month ChatGPT plan.
OpenAI’s documentation says each dot has its own cloud computer and browser, can work across connected tools, maintain context and continue making progress between conversations. The agents are powered by GPT-6 Astra.
Dots are built for ongoing work rather than one-off prompts
Unlike a standard chat session, a dot can be assigned a broader goal and continue working toward it over time. OpenAI says dots can use connected apps, handle recurring work and return to users when decisions or approvals are needed.
The company is initially rolling out Dots to Pro users outside the European Economic Area, Switzerland and the United Kingdom, as well as Business Premium customers. Enterprise customers can access the feature through an administrator-enabled beta.
Launch follows a renewed focus on agent safety
The announcement comes after OpenAI paused work on another advanced tool-using model because of internal safety concerns. OpenAI also joined other major AI developers this week in signing a voluntary White House safety pact covering internal controls, independent audits and board-level oversight.
Associated Press reported that CEO Sam Altman presented Dots at OpenAI’s developer conference as part of the company’s move from answering requests toward software that can proactively carry out work on behalf of users.
The launch places OpenAI more directly in competition with other companies building persistent digital agents and marks another step toward AI systems that operate continuously rather than only when prompted.
Sources: OpenAI Help Center and deployment-safety documentation; Associated Press; Anadolu Agency.
