OpenAI Launches Apple Messages Plug-In for ChatGPT to Draft, Edit, and Send Messages

Well, this seems interesting. ChatGPT’s latest update adds a new Mac integration: the ability to work with conversations from Apple’s Messages app. The feature works with iMessage, SMS, and RCS. ChatGPT for Mac now includes an Apple Messages plugin The new Apple Messages plugin lets ChatGPT read and search chats on a Mac, and prepare and send messages. OpenAI says it is available across all plans in the ChatGPT desktop app for macOS, with support in both ChatGPT Work and Codex. That access raises privacy and control questions, so OpenAI stresses that sending remains gated by user approval under the default settings. Users should take care before granting persistent approval. OpenAI’s plugin guide details the risks, how to revoke access, and a known issue involving tasks that disable approval prompts. Here’s more from the release notes: Use the Apple Messages plugin to read and search Messages chats on your Mac and prepare or send messages. It’s available on all plans in the ChatGPT desktop app for macOS. You can use the plugin in ChatGPT Work and Codex. By default, ChatGPT sends messages only after you approve the message and its recipients. See the plugin guide for persistent-approval risks, revocation steps, and the known issue with tasks that disable approval prompts. Separately, OpenAI details how the Messages feature works on its dedicated plugin site: The Apple Messages plugin is available on all plans in the ChatGPT desktop app for macOS. In Codex and ChatGPT Work, it can read and search iMessage, SMS, and RCS chats on your Mac and send messages on your behalf through the Messages app. It doesn’t let you interact with ChatGPT remotely through Messages, and it doesn’t work in regular ChatGPT chats. For now, the feature only works with ChatGPT on Apple silicon Macs and not Intel machines. This link should go directly to the Messages plugin for ChatGPT on the Mac. Site co-editing, shared thread snapshots, and more While the Messages plugin is the standout addition for Apple users, the release includes several other improvements. For example, pinned chats now stay in sync across the ChatGPT desktop app and iOS, so users can keep important conversations accessible between a Mac and iPhone. ChatGPT also now lets users share a read-only snapshot of a local Codex thread. These snapshots do not change as the original thread evolves. Links created from personal accounts are accessible to anyone who has the URL, while workspace links remain limited to members of that workspace. OpenAI says Codex redacts known secret patterns, but warns users to review snapshots because sensitive information could remain. Elsewhere, Site owners can invite active members of the same workspace to co-edit a Site where collaboration is available. Editors can update content and database data, save versions, and publish changes after the owner completes the first publication. Owners retain control over access, settings, analytics, and version restoration. Some Site owners can also change a ChatGPT-hosted URL without redeploying, with the old address redirecting to the new one. Finally, last week’s Computer History feature is expanding to Pro, Business, and Enterprise users in the EEA, Switzerland, and the UK. The optional feature requires Memories and remains off by default. You can read the full release notes for today’s update here. Do more with your Apple products

ChatGPT is getting a little more comfortable inside the Mac, with its latest update adding an Apple Messages integration. The feature lets users bring conversations from Apple’s Messages app into ChatGPT, giving the assistant another source of context without requiring users to manually copy and paste entire conversations. Another app joins ChatGPT’s growing ecosystem The integration is part of a broader push to make ChatGPT more useful as a desktop assistant rather than simply a chatbot sitting in its own window. OpenAI has been expanding the Mac app’s ability to work with other applications and information, including its Computer History feature, which can reference selected activity from supported apps and websites. Messages is an especially interesting addition because conversations often contain information that users may want to act on later. Instead of digging through a long thread, users can bring that context into ChatGPT and ask it to summarize a conversation, pull out important details, or help work with information contained within it. OpenAI’s broader desktop strategy is clearly heading in this direction. Its current Mac app combines ChatGPT with Work and Codex, while connected apps and plugins are designed to let the assistant work with external tools and information rather than keeping everything inside a standalone chat. But this is where things get personal Giving ChatGPT access to Messages is considerably more sensitive than connecting something like Google Drive. Messages can contain private conversations, personal details, addresses, financial information, and other information that users probably don’t expect an AI assistant to see. That’s why the permissions and privacy controls matter here. Apple’s existing ChatGPT integration already makes clear that users control when information is sent to ChatGPT, with Siri asking permission before sending files. The Messages integration therefore feels like a small feature with much bigger implications. ChatGPT is slowly moving from an assistant users talk to into one that can understand more of what they’re already doing on their devices. That’s undeniably useful, but the more of a user’s digital life it can access, the more important it becomes to know exactly what ChatGPT can see, when it can see it, and where that information goes.

ChatGPT can now control iMessage, which could raise Apple privacy concerns Thursday, August 20, 2026 Apple sued OpenAI last month, accusing the AI provider of copying information about its upcoming products Bloomberg OpenAI implemented on Thursday compatibility to control Apple Inc.’s iMessage service through ChatGPT, a measure that could raise privacy concerns for the iPhone maker. The new feature, available to ChatGPT on Mac, allows the OpenAI system to perform actions such as reading, composing, and sending text messages. It can also search messages and obtain summaries and other information about conversations from Apple’s Messages app. Apple has positioned itself as a strong privacy advocate and protects the content of its users’ messages. An OpenAI tool that could upload Apple customers’ data to the company’s network would likely raise questions. A representative for Apple did not respond immediately to requests for comment. This measure comes at a time of growing tension between the two companies. Apple sued OpenAI last month, accusing the AI provider of stealing information about its upcoming products. In response to questions about the new feature, OpenAI emphasized that it requires user consent and does not create an index of all messages. The plugin also runs locally on the Mac and uses OS tools such as AppleScript and Accessibility to work with the Messages app, according to OpenAI. Both in the ChatGPT app and in Apple’s operating system, there are several activation menus for this integration. When a user sets up the integration for the first time, a permissions screen appears in ChatGPT indicating that the app will access the device’s message history. Users must also adjust their privacy preferences in the Mac System Settings app. This includes enabling the Full Disk Access function. They must also grant ChatGPT permission to access their contacts’ names and automation tools. It is similar to the settings required to activate popular AI-based automation software, such as Codex Computer Use. OpenAI has recently implemented compatibility with data analysis in the user’s iPhone and iPad Health app, but Apple’s iOS and iPadOS operating systems have a feature that requires customers to explicitly approve that their data be shared with ChatGPT. ChatGPT already works with other Mac apps, such as the Xcode development suite, Notes, and Terminal, though these integrations are less likely to bother Apple’s privacy and marketing teams than an unofficial Messages integration. The new OpenAI feature also threatens to intrude on the territory of the future Siri AI, which has the ability to analyze users’ messages and other content on their devices. This capability is seen as a potential edge over third-party chatbots. In the past, Apple has tried to block similar features. A recent example is when it shut down an app called Beeper Mini that helped bring iMessage to Android.

OpenAI today said there's a new Apple Messages plugin for ChatGPT, allowing ChatGPT on Mac to search messages, draft messages, analyze conversations, and send replies to people. The Apple Messages plugin can be installed from the ChatGPT Plugins interface, and it is listed under Public. OpenAI suggests users try these prompts: - Check my calendar and reply to [name] with a few times I'm free for dinner next week - Suggest follow ups from yesterday in messages - Find birthdays in @messages and add them to my calendar - Find potential spam messages I can delete ChatGPT can access iMessages, SMS messages, and RCS messages. Sending a message requires user approval by default, and OpenAI warns against granting persistent approval. For the Messages integration to work, ChatGPT needs Full Disk Access in System Settings, as well as access to contact names and automation tools. Apple's involvement in the plugin's development is unknown, but it uses AppleScript and Accessibility settings to function on a Mac. Apple and OpenAI are currently facing off in court after Apple sued OpenAI over theft of trade secrets. Messages integration is available in ChatGPT Work and Codex in the desktop version of ChatGPT, but it does not work in regular ChatGPT chats. It is available for all ChatGPT users who have an Apple silicon Mac.

OpenAI added an Apple Messages plugin to the ChatGPT desktop app for macOS in release notes and its own plugin guide flags a known issue with tasks that disable approval prompts. The plugin reads, searches, and sends iMessage, SMS, and RCS chats. The plugin gives the macOS app direct access to conversations in Apple’s Messages app, where ChatGPT can read, search, and prepare outgoing texts. OpenAI lists it as available on all plans in the ChatGPT desktop app for macOS, with support in both ChatGPT Work and Codex. OpenAI has drawn two boundaries around it. The plugin does not let anyone drive ChatGPT remotely by texting it, and it does not function in ordinary ChatGPT chats. Hardware narrows the audience further.OpenAI’s plugin documentationrestricts the feature to Apple silicon Macs, which leaves Intel machines out. Everyday conversations just got easier with the new Apple Messages plugin.Search messages, catch up on conversations, draft and send replies—all with ChatGPT on your Mac.Now available in ChatGPT Work and Codex on desktop.pic.twitter.com/nicfZMuxZc OpenAI describes the send step as consent-gated. “By default, ChatGPT sends messages only after you approve the message and its recipients,” the company states in the release notes. That default comes with an exception OpenAI documents itself. The plugin guide covers the risks of granting persistent approval, the steps for revoking access, and a known issue with tasks that disable approval prompts. A task configured that way removes the one control most users assume is always on, a sharper problem than the permission questions raised byChatGPT’s advanced security mode. The documentation confirms the issue and stops there, with no trigger, scope, or fix described. NothingOpenAIhas published says how many users already run tasks configured this way, whether a patch has shipped, or whether a message sent under a suppressed prompt gets flagged afterward. Four questions sit unanswered in the release notes: Anyone who has already granted the plugin persistent approval can revoke it through the steps in OpenAI’s plugin guide, then audit any tasks set up to skip prompts. Reviewing recently sent threads in Messages helps reduce the risk that an outgoing message went unnoticed, though it does not rule that out. Codex picked up read-only thread sharing in the same update. A snapshot freezes a local Codex thread at the moment of creation and ignores later changes to the original, and OpenAI redacts known secret patterns before the link goes out. Links from personal accounts open for anyone holding the URL, while workspace links stay inside that workspace. Redaction has a limit OpenAI states plainly: sensitive information can survive the pass, so snapshots need a read-through before sharing. That caution sits alongside the company’s earlier work onsecret handling in Codex. Sites gained collaborative editing where owners invite active members of the same workspace. Editors can update content and database data, save versions, and publish changes once the owner completes the first publication, while owners keep control of access, settings, analytics, and version restoration. Some Site owners can also change aChatGPT-hosted URLwithout redeploying, and the old address redirects to the new one. One weekly briefing with the launches, AI developments, and breaches that matter. No filler. Access to a personal message archive changes both what an assistant can do and what it can spill. Reading and searching aMac’s iMessage historymakes ChatGPT useful for the retrieval people actually want, such as digging out an address someone texted months ago, and that same reach turns an unapproved send into something far costlier than a bad draft in a chat window. The approval prompt carries the entire safety model here, which is why a documented path to switching it off deserves more attention than any other line in the release notes. What happens nextdepends on how quickly OpenAI closes the task issue and how loudly it tells affected users. Anyone turning the plugin on now should leave the default approval behavior alone and recheck the plugin’s permissions after setting up any task, because that is the documented route to a send without a prompt. OpenAI named the problem in its own documentation, which sets the bar for how visibly it now has to close it, and its usage numbers put a large audience one toggle away from the answer.

If you’ve ever wanted to share all of your digital conversations with OpenAI, we have good news for you: The AI lab has just launched an Apple Messages plug-in for ChatGPT, allowing interested users to connect their Messages inbox with the chatbot. The benefits of doing this, OpenAI argues, are numerous. Users can use the plug-in to sort, analyze, or edit their messages directly from ChatGPT. The plug-in also works with Codex and ChatGPT Work, so users can use it professionally, not just personally. A brief commercial advertising the new plug-in shows a user asking the chatbot to suggest follow-up messages to their contacts based on the messages received the previous day. You can also ask ChatGPT to delete messages for you, draft and send messages on your behalf, or search for information buried deep in your message history. As with most things related to AI, this new feature raises some privacy questions. OpenAI told Bloomberg that the plug-in runs locally on a user’s machine and that it “doesn’t create an index of all someone’s messages.” Still, the specifics of what that means aren’t immediately clear. TechCrunch has reached out to OpenAI for more information. When it comes to message sending, OpenAI encourages users to keep an eye on what ChatGPT is doing and discourages turning on persistent approval, warning that doing so “removes your final chance to review a message before ChatGPT sends it as you,” the company writes.

Latest news ChatGPT for Macnow supportsApple’s Messages app through a new plugin that can read and search conversations, prepare replies, and send messages directly from a Mac. The integration works with iMessage, SMS, and RCS chats, giving ChatGPT access to conversations stored inside the Messages app when users grant permission. OpenAI says the Apple Messages plugin is available across all ChatGPT plans in the desktop app for macOS, although the feature currently requires an Apple silicon Mac and does not support Intel-based machines. Users canaccess the pluginthrough ChatGPT Work and Codex, while regular ChatGPT conversations do not support the Messages integration. By default, ChatGPT asks users to approve both the message and its recipients before anything gets sent, which gives users a chance to review the content before it leaves the Messages app. OpenAI also warns users to be careful when enabling persistent approval because that setting reduces the number of confirmation steps before messages are sent. Set us as a preferred source and our Apple reporting ranks higher in yourGoogle Searchresults andDiscoverfeed — one tap, no account changes. And you can edit messages before sending!pic.twitter.com/OPfTfUhXqX OpenAIexplainsthat the plugin works directly with conversations already available on the Mac, including iMessage, SMS, and RCS chats. However, users cannot use Messages as a remote way to control ChatGPT, so sending a text to the Mac will not trigger ChatGPT commands or start a remote session. The update also improves how ChatGPT works across Apple devices, as pinned conversations now sync between the macOS desktop app and the ChatGPT app on iPhone. This makes frequently used chats easier to access when moving between devices. OpenAI has also added read-only snapshots for local Codex threads, allowing users to share a fixed version of a conversation through a link. Personal account links can be opened by anyone with the URL, while workspace links remain limited to members of that workspace. The same release expands collaboration tools for ChatGPT Sites, including shared editing for eligible workspace members, URL changes without redeploying, and wider availability of Computer History for Pro, Business, and Enterprise users in the EEA, Switzerland, and the UK. Δ This site uses Akismet to reduce spam.Learn how your comment data is processed.

OpenAI has shipped an Apple Messages plugin for the ChatGPT desktop app on macOS, letting the chatbot read and search a Mac’s iMessage, SMS, and RCS conversations and prepare or send texts through the Messages app itself. The feature landed August 20, 2026, in the company’s release notes, and it works inside Codex and ChatGPT Work (the agentic surfaces OpenAI has been building out all year) rather than in ordinary ChatGPT chats. The plugin is available on all plans, but with a hard hardware boundary: it ships only in the Apple Silicon (arm64) build of the desktop app, so Intel Macs are out, and it does nothing in ChatGPT on the web or mobile, Codex CLI, or the IDE extension. After installing it from the Plugins tab, a user starts a new Codex or Work chat and asks it to find, summarize, draft, or send a message. ChatGPT requests the relevant macOS permissions before touching Messages, and the user reviews both the message and its recipients before anything goes out. That send approval is the product’s central design decision. By default, ChatGPT sends a message only after the user approves the content and the recipients — an “Allow once” choice per send. Users can instead select “Always allow sending to this chat,” which lets ChatGPT fire off future messages to that thread without asking again. OpenAI’s own plugin documentation actively discourages the persistent setting: “Keep per-send approval for chats that may contain untrusted or misleading instructions. Persistent approval removes your final chance to review a message before ChatGPT sends it as you.” The warning is prompt-injection hygiene in plain language: a text thread is untrusted input, and a chatbot with standing authority to send messages as you is a channel an attacker can steer. Users who enable persistent approval can revoke it under Settings, in the Computer Use section, by removing the chat from the always-allowed list. The documentation also flags a known issue: tasks configured with full access or otherwise stripped of approval prompts may leave Messages unable to display the send confirmation, and OpenAI’s remedy is to switch the task back to an approval mode. What the Plugin Actually Does, and Where It Stops The capability set is narrower than the phrase “ChatGPT can text for you” suggests. The plugin can read and search message history on the Mac, draft follow-ups and replies, and send them through Messages with approval. It cannot turn Messages into a remote control for ChatGPT — texting the chatbot from your phone is explicitly not part of the release — and it stays out of regular ChatGPT chats entirely. Analyzing threads and pulling information out of old conversations sit in the same read-and-search envelope, while sending requires approval. For managed deployments, workspace administrators can switch Apple Messages off through the existing Computer Use control, the same mechanism that governs other desktop-agent capabilities. That admin hook matters for the product’s enterprise posture: OpenAI has told investors that its enterprise revenue has overtaken the ChatGPT consumer business, and features that touch personal data on a work machine live or die by whether IT can gate them. OpenAI’s Desktop-Agent Buildout The Messages plugin is one tile in a larger mosaic. Since July 2026, OpenAI has folded Chat, Work, and Codex into a single desktop app, launched ChatGPT Work as its long-running task agent, replaced the app directory with a plugin directory, and added computer-use features that let Codex operate Mac and Windows applications with permission. Earlier this month it extended that consumer surface in a different direction with ChatGPT for Teens, a tier with default protections and learning tools. Plugins are the distribution layer for this strategy. The same universal directory serves ChatGPT and Codex across web, desktop, and mobile, with OpenAI-built plugins alongside workspace and personal marketplace entries, and connectors to services like Gmail, Google Drive, and Slack that authenticate separately and apply their own terms when data flows through them. Apple Messages is the deepest OS-level hook in that catalog so far: not a cloud connector to a third-party service, but an agent acting through the local messaging stack on the user’s own machine, with the send button gated by an approval prompt OpenAI is visibly reluctant to let users waive.

ChatGPT is moving from answering prompts to acting inside one of the most sensitive surfaces on your phone: your messages. OpenAI has launched anApple Messages plug-in for ChatGPT, letting users connect their Messages inbox to the chatbot. Once connected, ChatGPT can sort, analyze, edit, search, draft, delete, and even send messages on a user’s behalf.Advertisement That is a small interface change with a very big trust question attached. ChatGPT is no longer just helping you write a text. It is being positioned as an operator for the inbox itself. Everyday conversations just got easier with the new Apple Messages plugin. Search messages, catch up on conversations, draft and send replies—all with ChatGPT on your Mac. Now available in ChatGPT Work and Codex on desktop.pic.twitter.com/nicfZMuxZc — ChatGPT (@ChatGPT)August 20, 2026 The reported plug-in lets users bring their Messages inbox directly into ChatGPT, where it can help manage the content of conversations rather than simply respond to copied-and-pasted snippets. OpenAI is pitching the tool for personal use, but also for work: the plug-in reportedly works with Codex and ChatGPT Work. The examples matter. A promotional video cited in the report shows a user asking ChatGPT to suggest follow-up messages based on texts received the previous day. The same plug-in can also search through old message history, draft replies, send those replies, and delete messages. That makes this feel less like a writing assistant and more like an action layer for private communication. The inbox becomes searchable, editable, and executable through natural language. It fits a wider product pattern: tech companies are trying to make AI the control surface for things that used to require menus, filters, and manual decisions. Google is testing a similar shift in a very different context, withnatural language controls for tuning Discover. The message is the same: users are being trained to ask systems to do things, not just navigate them. Messaging is not like organizing a playlist or summarizing a document. It is intimate, messy, and often context-heavy. That makes OpenAI’s privacy framing central to whether this becomes useful or unsettling. OpenAI told Bloomberg that the plug-in runs locally on a user’s machine and “doesn’t create an index of all someone’s messages.” That is an important reassurance, but not the same as a full explanation of what is processed, when, and how users can audit the system’s actions. The more immediate product tension is approval. OpenAI encourages users to monitor what ChatGPT is doing and warns against enabling persistent approval, because it “removes your final chance to review a message before ChatGPT sends it as you.” That warning says a lot. The product is powerful precisely because it can act. But the moment an AI can send a message under your name, convenience and accountability collapse into the same button. For platforms and brands, the signal is bigger than texting. AI is moving deeper into the places where relationships are managed, not just where content is created. If users accept that bargain, messaging becomes less like a private archive and more like a task surface AI is expected to operate.

The chatbot is moving beyond drafts and suggestions, gaining the ability to act inside conversations while privacy questions remain unanswered. As noted byTechcrunch OpenAI has introduced an Apple Messages plugin for ChatGPT that allows users to connect their incoming messages to the chatbot. The tool makes it possible to work with conversations in the Apple Messages app directly through ChatGPT. With the new integration, users can organize and analyze messages, edit their text, search for information in old chats, and prepare replies for contacts. During a demonstration, OpenAI showed that ChatGPT can suggest how to continue a conversation based on the content of messages received the day before. The Apple Messages plugin supports deleting messages as well as creating and sending new texts on the user’s behalf. It has also been integrated with Codex and ChatGPT Work, so its capabilities can be used for both private communication and work correspondence. The feature effectively allows the chatbot not only to suggest drafts but also to perform actions in chats. Before a message is sent, users can review the generated text and edit it if necessary. The introduction of such a tool also raises questions about the privacy of personal and work chats. OpenAI told Bloomberg that the plugin processes data locally on the user’s device and does not create a general index of all their messages. However, the company has not yet disclosed the technical details of how this mechanism works. OpenAI advises users not to opt out of continuously confirming ChatGPT’s actions when sending texts. This review remains the final opportunity to read a message before the chatbot sends it to another person on the user’s behalf. ChatGPT’s integration with Apple Messages expands the possibilities for automating correspondence, but it requires careful oversight, especially when private data, work discussions, or important messages are involved.