Skip to content
LIVE
The Executives BriefThe Executives Brief

Signal blocks Siri and Gemini from reading your private messages

Meredith Whittaker's default screenshot block on iOS and Android forces a reckoning over who owns the data AI agents scan, and Apple, Google, and Microsoft are not off the hook.

ByLama Al-RashidTechnology Correspondent, The Executives Brief
·4 min read
Signal blocks Siri and Gemini from reading your private messages
Executive summary

Signal CEO Meredith Whittaker has made Signal's iOS and Android apps block Siri, Gemini, and other AI agents from screenshotting messages by default, citing OS-level agents that scan screen data and can send private messages to third-party servers. For decision-makers, the move signals that platform owners Apple, Google, and Microsoft have not given developers workable tools to stop AI agents from undermining app-level privacy, turning agent access into a board-level trust and competitive risk.

Signal CEO Meredith Whittaker just made the encrypted messaging app a no-go zone for the AI agents that Apple and Google are pushing into their operating systems. Updates to Signal's iOS and Android apps now block Siri on iOS and Gemini and other agents on Android from taking screenshots by default, unless the user has explicitly granted permission. That matters because AI agents use optical character recognition to pull text out of screenshots. If an agent can screenshot Signal, it can read messages that were meant to stay private. Whittaker says users who want to screenshot their own messages can disable the protection in Signal's settings menu, but she warns that doing so risks exposing your Signal content, and the content sent by your contacts, to AI agents.

The trigger is not hypothetical. In a blog post, Whittaker wrote that Google and Apple are integrating agentic capabilities like Gemini Intelligence with Google Gemini and Onscreen Awareness with Siri AI deep into their operating systems. "In order to gain access to information about you, these OS AI agents 'scan' the information that's on your screen," she wrote. "In doing so, your Signal data is captured, then made available to these agentic systems in order to allow them to answer your questions and do your bidding." The result, she says, is that private messages ingested by these agentic systems may end up on someone else's server where they are no longer private.

Signal is not new to this fight. The app added screenshot blocking last year to defend against Microsoft Recall, and Whittaker is now extending that logic to the mobile agents baked into iOS and Android. But she concedes screenshot blocking is only a partial solution, since AI agents may have other ways to access application data through the device file system. "We want to be very clear: it is a serious problem that Signal - and application developers beyond - lack robust tools to prevent AI agents from undermining application-level privacy," she said. "Even where we are able to detect and block AI data access, Google, Apple, Microsoft, and others do not provide ways for developers to easily restrict such AI data access without making drastic compromises to the core user experience. This is also the reason we are currently unable to implement these protections on macOS."

Apple is taking some action on macOS, though its motives look narrower. With macOS 27.3, currently in public beta testing, Apple will add more granular controls on two APIs: Device Control and Data Access, and Full Disk Access. The result will be more permission prompts for macOS apps that seek access to user data, plus the inconvenience of applying for an entitlement from Apple to obtain Full Disk Access for backup software and assistive technologies that use Accessibility APIs. "As AI agents become increasingly capable and autonomous, the risks associated with this level of access will grow substantially," Apple said when it introduced the Full Disk Access change. "We are committed to ensuring users clearly understand these risks before granting such access, so they can make informed decisions about their own data and privacy." Apple is also reportedly working on its own AI agent to compete with Anthropic and OpenAI. It remains to be seen whether Apple will use its operating system ownership as competitive leverage against AI firms that threaten its gatekeeping power, as it has done with web browsers. Siri can be turned off separately, and Apple's macOS changes appear aimed largely at third-party access rather than at limiting its own Siri service.

Signal's move lands in the middle of a broader backlash against agents that were supposed to serve users but increasingly look like data middlemen. AI agents can spirit data away to the cloud and come between customers and merchants, which wears out their welcome fast. Amazon has blocked shopping agents like Meta Muse from its e-commerce site, though not AWS Marketplace, while Walmart has invited third-party agents to interact through its AI Gateway. Anthropic has blocked subscription customers from using third-party harnesses. United Airlines has banned transactions from unauthorized AI bots. Similar barriers or limitations related to third-party AI agents have been erected by Target, Home Depot, Etsy, eBay, JPMorgan Chase, Charles Schwab, Citigroup, and Wells Fargo, among others. Meanwhile, a burgeoning industry in bot detection and mitigation has formed, led by the likes of Cloudflare, Akamai, and DataDome.

The web browser offers the cautionary tale. It is sometimes referred to as a user agent, based on the idealistic notion that it serves its user. That vision has been muddied by online advertising and data collection, and by default settings that often serve the interests of browser makers instead of promoting privacy. Now AI agents face the same scrutiny as people wonder whether their agents really serve them or someone else. After a year or two of agent vendors arguing for access to all digital data for the sake of personalization and context, the backlash has arrived.

For executives and boards, the Signal episode is a preview of the permission fights coming to every software category. If operating system agents can scan screens, then every app's data becomes a potential input, and every privacy policy becomes a liability if it cannot enforce boundaries at the device level. Developers currently lack robust tools to block AI data access without wrecking the core user experience, which means the burden falls on platform owners like Apple, Google, and Microsoft to build real controls. Companies deploying agents should expect harder questions from partners, regulators, and customers about what gets scanned, where it goes, and who can turn it off. Signal's default block is a warning shot: the agent era will not be won by access alone. It will be won by trust that can be verified.

Executive ActionsLocked

This story's Key Insights and Take-aways are locked.

Create a free account to unlock Executive Actions for one credit.

Register to Unlock

Always free for Executives Club members. Join the Club

More in Technology