OpenAI is bringing its Dot AI agents to everyday messaging apps, allowing eligible ChatGPT Pro users to communicate with their personal AI assistants through Apple iMessage and Android’s Rich Communication Services (RCS).
The feature began rolling out in beta in the United States on October 28, 2026, making it possible to delegate tasks and receive updates without opening the ChatGPT app every time.
The new feature allows users to send their Dot agents text messages, photos and voice messages. Users can ask an agent to handle a task, monitor something important or provide updates when new information becomes available. The agent can then send messages proactively when it has relevant information to share.
By popular request, you can now text your dot. Give it a task to take off your plate or something to keep an eye on, using text, photos, or even voice notes. It will text you proactively with relevant updates. Safe, secure texting via iMessage or RCS. You’re the only one who can text your dot, and your dot can only text you. Rolling out now as a preview for Pro users in the US.
— ChatGPT (@ChatGPT) October 28, 2026
Dot agents retain the same memory and connected plugins across ChatGPT and supported messaging channels, allowing conversations to continue without starting from scratch. Users can also influence how their agents communicate, such as requesting text messages for important developments while keeping routine updates inside ChatGPT. If an action requires approval, the agent can send a link so the user can review the request in ChatGPT.
Getting started requires opening the Dot’s profile in ChatGPT and selecting the Texting option. Users must then enter and verify their phone number and save the contact provided by their agent. On iPhones, iMessage must be enabled, while Android users need to have RCS enabled in their messaging application. The feature is available only to accounts included in the limited beta rollout.
OpenAI is working with Linq, a third-party messaging provider, to handle the sending and receiving of messages. The company says it has conducted extensive penetration testing with its advanced AI models and external security firms as part of its work on the integration.
Privacy is another important part of the rollout. OpenAI says a Dot can communicate only with its owner, and the owner is the only person who can text that agent. Connecting a Dot to messaging does not give the agent or OpenAI access to the user’s other text conversations. However, because messages pass through a third-party provider, users should review the applicable privacy terms and avoid sending sensitive information unnecessarily.
The initial release also comes with several limitations. Messaging supports one-to-one conversations, but group chats, phone calls and FaceTime are not supported through the texting feature. Users can still call their Dot through ChatGPT itself. Access is restricted to eligible Pro subscribers in the United States, and not every Pro account may receive the beta immediately.
The update builds on OpenAI’s broader effort to make Dot agents more useful as persistent personal assistants. By extending access to familiar messaging applications, the company is making it easier for users to assign tasks, share information and receive timely updates without keeping ChatGPT open throughout the day.
More details are available in OpenAI’s official ChatGPT release notes and its guide to messaging your Dot.
If this article helped you, please consider supporting our work. Every small contribution keeps Abijita.com independent and running.
