SAN FRANCISCO: OpenAI has introduced Dots, a new type of persistent artificial intelligence agent designed to carry out tasks on behalf of users and continue working between conversations.
The company announced Dots during its DevDay 2026 developer conference, describing them as always-on agents that can take responsibility for ongoing work. Dots are powered by GPT-6 Astra and operate through their own cloud computer and browser.
Unlike a conventional chatbot, a Dot can be given a goal and allowed to make progress on tasks over time. It can determine what needs to be done next and return results to the user for review, while asking for input when a decision requires the user’s judgment.
OpenAI said Dots can work with connected applications and tools, allowing users to incorporate them into existing workflows. The company also said Dots can handle recurring work and follow up through services including ChatGPT, email, Slack and text messaging.
Users can interact with Dots through ChatGPT and set controls defining what the agents are allowed to do. OpenAI has also introduced additional safeguards for persistent and long-running tasks.
According to OpenAI, each Dot has its own cloud computer and browser. The system is designed to maintain context and learn from ongoing collaboration so it can better handle recurring responsibilities.
The company said privacy and security protections are built into the system, including controls that allow users to determine the scope of an agent’s actions.
OpenAI is gradually rolling out Dots, so availability and specific features may vary by account and plan.