Skip to content
VECTOR WIREAI INTELLIGENCE
UTC
Home

Signal restricts screenshots on mobile to block Gemini and Siri AI from reading chats

Screenshot restrictions apply by default on Android and iOS 27 to block Gemini and Siri AI screen access.

Signal is restricting screenshots by default on Android and iOS 27 to prevent AI agents built into mobile operating systems from capturing and storing private messages1,2. The encrypted messaging platform described the protections as "incomplete" and called on Apple, Google, and Microsoft to give developers stronger tools to opt apps out of AI data access3.

Signal president Meredith Whittaker said in a blog post on November 20 that agentic AI capabilities analyzing on-screen information are "undermining Signal's ability to independently guarantee privacy". She added that 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".

What Signal is blocking

The restrictions target Google's Gemini Intelligence and Apple's Onscreen Awareness with Siri AI, agentic features now integrated deep into their respective operating systems4. On Android, Signal automatically enables its Screen security setting when it detects an AI assistant attempting to access the screen. On iOS 27, Signal blocks screenshots in the app and gives users an option to toggle the protection.

Signal had already taken a similar step on Windows last year, using screenshot blocking to prevent Microsoft's Recall from taking and storing screenshots of Signal messages on Windows 11 by default.

The platform said it is currently unable to implement these protections on macOS, citing a lack of clearly documented APIs for screenshot protection. Signal's only recommendation for macOS 27 users is to manually add Signal to the Excluded Apps section in System Settings to protect message content from Siri AI.

The Meta Muse incident

Meta's Muse AI agent accessed Signal conversations on an Apple computer and an Android smartphone by using operating system-level permissions, according to a Reuters report cited by multiple outlets5. On macOS, Muse used accessibility features and screenshot tools to navigate Signal, capture chats, extract their text, and reproduce the messages within the Muse application. Apple released an operating system update this week that closed the specific access route, according to the same reporting. Meta told Reuters that AI agents need access to users' screens and devices to perform tasks on their behalf and that Muse accesses other applications only through permissions explicitly granted by users.

Broader developer demands

Signal said it spent considerable developer time over the past few months conducting forensic investigations across Android and iOS to learn what happens when AI systems access application data. The company called on OS providers to create better documentation and technical specifications around how AI agents function, particularly regarding data access and processing. Signal said developers should have an easy, well-documented way to opt their apps out of AI data access, covering screen reading, filesystem access, notifications, and other OS frameworks, without undermining core utility.

Signal described these agentic AI integrations as changing the operating-system environment in a way that is increasingly hostile to privacy. Signal noted that some major OS providers, including Google and Apple, are engaging with it on how these systems should better interface with privacy-sensitive apps.

ANALYSIS The move extends a pattern Signal established with its Recall block on Windows to the two dominant mobile platforms, but the macOS gap and the platform's own "incomplete" characterization underscore the limits of application-level defenses against OS-level AI features.

Users can disable Signal's screenshot protection within the app's settings.

Was this useful?

The day’s vector, in your inbox.

Vector Wire Daily: one dispatch each publishing day.