Story
August 21, 2026

ChatGPT Can Now Text Through Apple Messages, but the Privacy Trade-Off Is the Real Story

OpenAI’s Apple Messages plug-in for ChatGPT on Mac promises to search, draft and send replies from users’ inboxes. The convenience is immediate, but so are questions about access, oversight and who controls intimate conversations.

ChatGPT’s latest Mac integration turns Apple Messages into an AI workspace, offering a faster way to manage conversations while asking users to hand a chatbot unusually close access to their private lives.

OpenAI announced the Apple Messages plug-in as a tool for ChatGPT on Mac, saying it can “search messages, catch up on conversations, draft and send replies.” The pitch is straightforward: users can surface old information, summarize ongoing threads and generate responses without manually combing through an inbox.

The broader rollout goes further. The plug-in can sort, analyse and edit messages, and it also works with Codex and ChatGPT Work, extending the feature from personal correspondence into professional use. A demonstration showed ChatGPT suggesting follow-ups based on messages received the day before; users can also ask it to delete messages, search message history, or draft and send texts on their behalf.

That is where the convenience argument collides with the privacy one. OpenAI told Bloomberg that the plug-in runs locally on the user’s machine and “doesn’t create an index of all someone’s messages.” But the available description leaves key practical questions unresolved, including precisely what information the system can access and how those local operations are handled.

OpenAI’s own guidance underscores the risk of giving the assistant too much autonomy. The company advises users to monitor what ChatGPT does and warns against persistent approval because it “removes your final chance to review a message before ChatGPT sends it as you.” The feature may make Messages more useful, but it also makes the final human check more consequential.