OpenAI unveils Dots, always-on personal agents, at DevDay

OpenAI unveiled Dots, a lineup of always-on personal AI agents, at its DevDay conference in San Francisco on September 29, Engadget reported. Each Dot runs on its own isolated cloud infrastructure with a virtual browser. Pro, Business, Premium, and Enterprise subscribers get one primary Dot at launch, with the ability to control multiple Dots planned for later.

CEO Sam Altman framed the agents in the keynote as "like an AI helper that always has your back, inspired by the cool versions of what we all watched in movies growing up." He described the delegation model bluntly: "You can delegate ambitious pieces of work the way you would to a high agency engineer or a chief of staff that you work with."

OpenAI's pitch is that the agents complete work "the way you would do it, sometimes before you even think to ask." In the company's launch example, one early tester's Dot noticed an invoice had never been sent to a publication, prepared it, and sent it after approval. Dots also plug into Slack and Microsoft Teams so users can converse with them outside ChatGPT.

On safety, OpenAI put read-only restrictions on proactive research features, blocking a Dot from sending messages, modifying content, or exercising unauthorized system control. Sensitive actions such as changing a password stay with the user, managed through ChatGPT's existing controls and custom approval rules.