Apple has long marketed itself as the tech industry’s privacy champion, but a newly introduced feature may test that reputation. OpenAI and Apple have rolled out a capability that lets ChatGPT read, write, and send iMessages on Mac computers, according to a report from Bloomberg. The integration allows the AI assistant to search conversations, summarize message threads, and even compose and dispatch texts — all within Apple’s Messages app.
The feature, which is available through ChatGPT on the Mac, represents a significant step in AI’s reach into personal communications. But it also raises fresh questions about how much of a user’s private messaging data could end up in the hands of a third-party AI service — a concern made sharper by the fact that Apple is currently suing OpenAI for allegedly stealing hardware trade secrets.
What the Integration Does
According to Bloomberg’s reporting, the new tool works by letting ChatGPT take direct action inside Apple’s Messages app. That includes pulling up past conversations, generating summaries of message content, and drafting or sending new texts on the user’s behalf. The AI can also search through message history, effectively giving it broad access to a user’s personal communications.
On its face, the feature is similar to other AI-powered automation tools that have appeared on Macs in recent months. ChatGPT already integrates with other Apple applications, including Xcode, Notes, and Terminal. But those connections are generally considered less sensitive than a link to Messages, where intimate and often confidential exchanges take place.
OpenAI was quick to note that the integration is not a free-for-all data grab. The company said the feature requires explicit user consent and does not build an index of all of a person’s messages. The plug-in runs locally on the Mac, meaning the AI processes data on-device rather than sending it to cloud servers by default. It relies on existing operating system tools, such as AppleScript and Accessibility, to interact with the Messages app.
Opt-In Steps and Permissions
Setting up the integration involves several steps, each designed — at least in theory — to put the user in control. When the feature is first activated, a permissions screen in ChatGPT notifies the user that the app will access on-device message history. Users also have to manually adjust privacy settings in the Mac’s System Settings, including enabling Full Disk Access, which is a significant level of permission that grants the app broad visibility into files and data on the machine.
Additionally, the user must grant ChatGPT permission to access contact names and automation tools. That process is similar to what’s required for other AI-based automation software, such as Codex Computer Use, which also runs on Macs. In other words, while the setup is onerous, it’s not unique in the AI landscape.
This is not the first time OpenAI has sought deeper access to Apple’s ecosystem. Earlier, the company added support for analyzing data from a user’s iPhone and iPad Health app. In that case, Apple’s iOS and iPadOS systems require customers to explicitly approve sharing that data with ChatGPT, giving Apple a degree of oversight over how sensitive health information is used.
Apple’s Legal Battle With OpenAI
The timing of the integration is awkward, to say the least. Apple filed a lawsuit against OpenAI in federal court in Northern California, alleging systematic theft of the iPhone maker’s intellectual property. The complaint claims that OpenAI obtained Apple’s proprietary information across multiple levels of its organization.
In court documents, Apple pulled no punches. “This much is clear, however: at every level, from members of its Technical Staff to its Chief Hardware Officer, and in coordination with business partners, OpenAI has been stealing Apple’s trade secrets and confidential information,” the company stated.
The lawsuit marks a dramatic shift in the relationship between the two companies, which have also collaborated on AI features in recent years. But the legal fight has not stopped the commercial partnership from advancing — at least not when it comes to products like the new Messages integration.
Privacy Concerns and the Bigger Picture
Apple has built its brand around protecting user privacy. The company has consistently emphasized on-device processing and end-to-end encryption for iMessage, and it has been guarded about the contents of customer messages. A tool that could send Apple user data to OpenAI’s network — even with consent — is likely to draw scrutiny from privacy advocates and regulators alike.
The new feature also pits ChatGPT against Apple’s own upcoming Siri AI. Apple is reportedly developing a more advanced Siri that can search through messages and other content stored on a user’s device. That capability has been viewed as a potential advantage over third-party chatbots because it keeps data on-device. If ChatGPT can now do similar things on Macs — and potentially as well or better — it could undercut that advantage.
For now, users who want to enable ChatGPT in their Messages app must jump through several hoops and explicitly consent to the access. But the very existence of the feature highlights the growing tension between the convenience of AI and the privacy expectations that Apple has cultivated for years. As the integration rolls out, observers will be watching closely to see how both companies navigate the fallout — and whether the legal battle over trade secrets spills over into the consumer AI space.
Source: www.breitbart.com — https://www.breitbart.com/tech/2026/08/24/ai-privacy-nightmare-apple-welcomes-chatgpt-into-imessage-to-read-and-write-text-messages/