OpenAI now lets ChatGPT Pro subscribers in the US text their dot, the always-on agent inside ChatGPT, through iMessage on iPhone or RCS on Android. The beta, announced on 28 October 2026, accepts text, photos and voice notes, and the dot texts back on its own when a task moves forward. Texting does not work in the Netherlands or elsewhere in Europe yet.
Dots launched at OpenAI’s DevDay on 29 September as agents that keep working after you close a chat, each with its own cloud computer and browser. Until now you reached your dot in the ChatGPT app or through Slack and Teams. Texting puts it in the Messages app on your phone, next to your other conversations.
A verified phone number turns your dot into a contact in Messages
Setup starts in ChatGPT: open your dot’s profile and choose Texting. Digital Trends describes the next steps. You verify your phone number, save the contact your dot sends you and reply to it. On Android, RCS (the successor to SMS) must be switched on in your messaging app.
Conversations are one to one. Group chats, phone calls and FaceTime are not supported, although you can still call your dot inside ChatGPT. Your dot keeps the same memory and plugins wherever you reach it, so a task you start by text can continue in the app.
Approvals still happen in the ChatGPT app, not in the text thread
Your dot decides where to send an update, based on your conversations and connected channels. You can tell it to text only about urgent matters and keep routine updates in ChatGPT. When a task needs your approval, the dot texts a link that opens the request in the app.
Custom Rules in ChatGPT let you allow, block or require approval for specific actions. Sending approvals back to the app keeps that check in one place, instead of a quick yes in a chat thread. OpenAI also says only you can text your dot, your dot can only text you, and it gets no access to your other conversations. Engadget reported that OpenAI ran what it calls extensive penetration testing before the launch.
Linq carries the messages, and the release note says nothing about what it keeps
OpenAI uses the messaging company Linq to deliver the texts, Engadget reported. OpenAI’s release note does not say what Linq can see, where it processes messages or how long it keeps them, and the coverage so far does not answer it either. For an assistant that receives photos, voice notes and questions about your work, that is the first point a privacy officer will want settled.
Can you text a ChatGPT dot from Europe?
No, not yet. Texting is a beta for Pro users in the US, and OpenAI has given no date for other countries or plans.
Dots themselves are coming closer. At launch, personal Pro accounts in the EEA, Switzerland and the UK were excluded, while Business Premium workspaces got dots in all supported ChatGPT regions, The Next Web reported. On 28 October, at OpenAI’s DevDay event in Berlin, developer experience lead Romain Huet said dots should become more widely available in Europe in a few weeks, after talks with regulators. The Decoder reported his remarks.
A dot in your text messages should not hold your work plugins yet
For organisations, the change is about where agent work happens. A dot can connect to more than 4,000 apps through plugins, OpenAI said at launch. Once that dot lives in a personal Messages app, instructions and results travel through a phone thread, outside any company workspace.
Our advice to the organisations we work with is to start where an admin is in control. In Enterprise, Edu and Healthcare workspaces, dots stay off until an admin enables them, and Business Premium workspaces already have them. Pick one recurring task, such as tracking supplier updates or preparing a weekly report, and set Custom Rules so the dot asks before it sends, shares or buys anything. Tell staff now that a personal Pro dot gets no work mail, calendar or file plugins, so that rule exists before dots and texting reach Dutch phones.
For more information, visit the official announcement of Text your dot on OpenAI’s ChatGPT What’s new page.
