OpenAI just dropped Dots, its new fleet of always-on AI agents, aiming to revolutionize how we get stuff done online. Unveiled at DevDay 2026, these personalized digital helpers, depicted as customizable blobs, are designed to tackle multi-step projects for you.
Powered by OpenAI's cutting-edge GPT-6 Astra model, Dots stand out with their "always-on" capability. Unlike typical chatbots that require new prompts for every request, Dots continuously scan the web and work on assigned tasks in the background. They also tap into your connected apps for context and learn your preferences over time, becoming increasingly tailored to your needs.
Imagine a Dot proactively suggesting dinner options when it sees you're working late, complete with delivery details like GrubHub, or helping you launch a new website from scratch. That's the kind of seamless assistance Dots offer. Users can interact with their Dots through ChatGPT, Slack, and Microsoft Teams, with context carrying over across platforms. ChatGPT Pro users can also join a waitlist for iMessage and Android RCS integration.
The rollout begins today exclusively for ChatGPT's $100/month Pro subscribers. OpenAI typically rolls out premium features to this tier first before broader distribution. While users start with one Dot, the company plans to enable control over multiple agents simultaneously in the future. Dots are built with safety in mind, requiring explicit approval for sensitive actions like installing software or changing passwords, and offering a "Custom Rules" tool for users to set clear boundaries.
This move positions Dots as a direct competitor to Meta's recently launched Muse agent, which has already gained significant traction. With Dots now entering the arena, the competition for the most sophisticated personal AI agents is heating up, promising exciting advancements for everyday users.
