OpenAI has expanded ChatGPT’s Dot agent to iMessage and RCS, so users can now text their Dot instead of opening the ChatGPT app. Announced on October 28, 2026, through ChatGPT’s own account on X, the expansion lets people give their Dot a task or something to keep an eye on using text, photos or voice notes, and receive proactive updates by message.
From Inside The App To The Messages App
Until now, Dot was limited to the ChatGPT experience on mobile, desktop and the web. That meant interacting with a proactive agent required opening the main ChatGPT app, even for a quick task or a check-in. Texting removes that step and puts the agent where many people already spend their day: the Messages app.
The pitch, in OpenAI’s own words, is simple: give it a task to take off your plate, or something to keep an eye on, and it will text you proactively with relevant updates. A Dot that can be reached by text becomes more like a personal assistant on call, and less like a tool you remember to open. That shift is the whole point of a proactive agent: it should meet you in the flow of your day, not wait for you to come to it.
Setup is deliberately short. To enable texting, users tap on their Dot’s profile and select Texting. The Dot keeps the same context and plugins wherever it is reached, so a task started in the app can continue by message, and vice versa. OpenAI also stresses a privacy boundary: texting does not give the Dot, or OpenAI, access to the user’s other text conversations.
What Text, Photos And Voice Notes Unlock
The three input types cover different kinds of tasks. Text suits a straightforward instruction or a question to monitor. Photos let a Dot see something the user is looking at, from a document to a scene that needs identifying. Voice notes suit hands-free moments, when typing is impractical but a quick spoken instruction is not.
Proactive updates are the other half of the feature. A Dot is not only a responder; it is meant to watch for developments and report back when something relevant happens. By message, that loop feels natural: the user sends a task, goes about their day, and receives a text when there is something worth knowing.
OpenAI describes the texting itself as safe and secure via iMessage or RCS, without spelling out the full technical detail in the announcement thread. Users should therefore treat the announcement as the starting point, not the complete privacy picture, and check OpenAI’s help documentation for how messages are handled before handing a Dot sensitive tasks.
Limits And Open Questions
The announcement thread does not spell out full availability, pricing tiers or regional rollout in detail, and that gap matters for readers trying to use the feature today. Some reports describe the texting expansion as a limited beta, which would mean not every Dot user sees it at once. Until OpenAI publishes fuller availability details, the safest reading is that the feature is rolling out, and users who do not see the Texting option yet may need to wait for their account to be included.
There is also a device question worth flagging as speculation, not fact. Bringing Dot to Messages could in theory open a path to the Apple Watch, where messages are a natural interface, but OpenAI has not announced Watch support, and that possibility remains untested. Readers should not buy a device or change a plan on the strength of that speculation.
The expansion also sits inside a busy week for ChatGPT’s Apple integrations. OpenAI recently added iCloud Mail and Calendar plugins, letting ChatGPT search mail, catch up on inboxes and update calendar events, and the Dot expansion extends that same idea: ChatGPT should work through the tools people already use. The comparison with other voice and agent surfaces is instructive, from ChatGPT’s iCloud mail and calendar plugins to Perplexity’s voice mode that listens while it talks.
Why It Matters
Dot matters because the hardest problem for personal agents is not intelligence, but presence. An agent that only works inside one app is easy to forget, however capable it is. An agent that can be texted, that remembers context across surfaces, and that reports back unprompted, has a chance to become part of daily routines rather than a demo visited occasionally.
That presence also raises the stakes. A Dot with access to tasks, photos, voice notes and proactive monitoring holds a detailed picture of what its user cares about, and the boundary OpenAI draws around other conversations will be tested as the feature expands. If texting proves reliable and restrained, it could become the most natural way to use a personal agent; if it becomes chatty or intrusive, users will simply mute it. The Messages app is a demanding place to earn a permanent spot.
Background
Dot is OpenAI’s proactive, always-on agent for ChatGPT, designed to take on tasks and monitor developments on a user’s behalf rather than only answering questions when asked. The move to iMessage and RCS extends that agent from inside ChatGPT into the messaging apps where everyday communication already happens.
Source: 9to5Mac report
0 Comments