Signal Blocks Screenshots by Default to Stop Siri and Gemini AI Agents From Reading Your Private Chats
The encrypted messaging service says built-in assistants such as Siri and Gemini pose an 'existential threat' to its privacy protections, though the measure has limits.

NEW YORK — Encrypted messaging app Signal is blocking screenshots on smartphones by default. The goal is to stop AI agents built into phone operating systems from reading users' private chats. Signal warned that the tech industry's push into AI is threatening its ability to protect privacy.
Encrypted messaging service Signal said on Friday it would restrict screenshots to stop AI agents from reading private chats on phones, warning that the tech industry's push to embed AI in consumer devices was eroding its ability to guarantee privacy.
Signal is most worried about AI built directly into phone software. Signal said Apple's Siri AI and Google's Gemini Intelligence are especially concerning because they are deeply built into devices.
"This is an existential threat to Signal, and our ability to responsibly uphold our mission," the company said in a blog post.
How the change works
The restriction covers agentic systems built into iOS 27, and Android versions where Signal can detect an AI assistant trying to read the screen. When Signal detects a screenshot attempt or an AI assistant trying to read the screen, users see a pop-up with an option to disable the block. The exact behavior depends on the device.
Users can disable the protection at Settings > Privacy > App Security, using "Block Siri AI" on iOS or "Screen security" on Android.
Signal President Meredith Whittaker explained the risk. "In order to gain access to information about you, these OS AI agents 'scan' the information that's on your screen," she said.
Signal said turning the protection off is risky. "However, doing so will potentially expose your (and your contacts') Signal content to these AI agents, and break our privacy guarantees," the company said.
Agents can store captured Signal content locally or on remote servers, outside the app's control, which could expose every conversation participant's privacy.
An admittedly partial fix
Signal admits the measure has limits. It conceded that the protection is "incomplete." Blocking screenshots does not stop other methods, such as an agent retrieving content already stored on the device's file system. On macOS, Signal said it could not reliably offer such protections.
"Right now, these agentic AI integrations are changing this environment in a way that's making it increasingly hostile to privacy," Signal said.
Signal called on the biggest operating system makers to act. It urged Apple, Google, and Microsoft to give developers more concrete ways to block agents, and asked these companies to explain how their agents work so developers can see what data agents access. Signal also asked for an easy, well-documented way for apps to opt out of AI data access, including screen reading, file system access, notifications, APIs, and OS frameworks, without major UX compromises.
Whittaker said accessibility settings meant for people should not act as "a skeleton key that provides access to AI agents."
The Meta Muse episode
The move follows reports that an AI agent from Meta Platforms was reading Signal messages. Reuters reported that Meta's Muse agent accessed Signal messages despite Meta's own "Connector" tool. Muse relied on broad OS-level permissions granted during setup.
On macOS, it used accessibility and screenshot tools to capture Signal screens, extract text, and recreate the content within the Muse app. On Android, it read messages by abusing push notification access. In one case, Muse reminded a Meta staffer to reply to a Signal message from her sister on her personal phone. An unnamed employee posted internally: "This seems like an extreme privacy violation."
Meta said agents need to see and interact with users' computers to work on their behalf, and that Muse accesses other apps only through permissions users explicitly grant.
Tech companies respond
Some platform makers have started adding controls. Apple released iOS 27.2 with an "Excluded Apps" option, letting users control which apps Siri, Search, and suggestions can access. Apple also previewed macOS controls for more granular, AI-aware app data permissions. Google Gemini lets users disconnect apps through its Connected Apps settings.
A familiar fight
Signal has taken this step before. The approach mirrors what Signal implemented last year for its Windows desktop app, where it blocked Microsoft's Recall — a feature that captured periodic screenshots of user activity — from storing images of Signal messages by default.
Other companies are also limiting AI agents. Signal joins United Airlines and Amazon in restricting AI agents' access to their platforms. United is blocking third-party agents from booking flights, while Amazon has barred Meta's Muse.
The dispute reflects a growing tension in consumer technology. AI assistants become more useful the more they can see, but privacy-focused apps say that access weakens the protections their users depend on.
© Copyright 2026 IBTimes AU. All rights reserved.

