Skip to main content
Close-up of a smartphone displaying the Signal app, with a robotic eye icon superimposed, symbolizing AI screen-reading preve

Editorial illustration for Signal Adds Feature to Block AI From Reading Your Screen

Signal Blocks AI From Reading Your Screen

• 2 min read

Meredith Whittaker has spent the past year warning that operating systems are turning into surveillance layers, and now her company is doing something about it inside its own app. Signal's president published a blog post this week laying out why the encrypted messenger needs a new toggle for Android and iOS users, one that cuts off the kind of screen access that Apple's Siri and Google's Gemini increasingly rely on to function as assistants. Those systems work by scanning whatever is visible on a phone, reading texts, parsing images, pulling context from apps that never agreed to share it.

Whittaker's complaint is that there's no real mechanism for a developer to say no. Apple and Google build the plumbing; app makers like Signal are left guessing at what their users' devices are quietly handing over.

The fix Signal shipped is blunt: block screenshots, and with them, the OS-level tools that use those captures to feed AI models. It's a workaround, not a negotiated standard, and Whittaker says as much in her post, pressing both companies to give developers actual controls rather than forcing this kind of defensive patch. The move puts a precise, named target, Apple and Google's assistant AI, at the center of an argument Signal has been making in general terms for months.

Signal is restricting the ability to take screenshots on Android and iOS devices in an effort to prevent AI agents from reading users' messages and other data.

Why this matters

Signal's move is a preview of a fight that's coming for every app that touches sensitive data. Apple and Google built Siri and Gemini to read what's on screen by default, which means the privacy burden now falls on individual developers to opt out on behalf of their users. That's a strange inversion: platform-level AI features are becoming an attack surface that app makers have to patch against, not a tool they asked for.

For developers building anything with end-to-end encryption or regulated data, this is a signal, no pun intended, that "AI-enabled by default" at the OS layer creates liability you didn't choose. For founders pitching on-device AI as a privacy win, Signal's pushback is a useful counterpoint: on-device doesn't mean consent-based, and screen-reading assistants blur that line fast. Expect more apps, especially in finance, health, and messaging, to ship similar toggles.

The real question for researchers is whether Apple and Google start offering granular, per-app opt-outs natively, or whether every privacy-conscious developer has to keep building their own workaround against their own platform.

Common Questions Answered

Why did Signal add a feature to block AI from reading your screen?

Signal's president Meredith Whittaker warned that operating systems are becoming surveillance layers, and the new toggle prevents AI assistants like Apple's Siri and Google's Gemini from scanning sensitive information visible on users' screens. By restricting screenshot access, Signal protects encrypted messages and other private data from being read by these AI agents that function by default by scanning whatever is visible on a device.

How do Apple's Siri and Google's Gemini access information on your screen?

Siri and Gemini work by scanning whatever is visible on a user's screen to function as intelligent assistants. These systems have been built to read screen content by default, which means they can potentially access sensitive information displayed in apps like Signal unless developers implement protective measures.

What is the privacy concern with platform-level AI features reading screen content?

Platform-level AI features like Siri and Gemini create a privacy burden that falls on individual app developers to opt out on behalf of their users, rather than users having control by default. This represents an inversion where AI features become an attack surface that app makers must patch against, rather than a tool they actively requested or implemented.

Why does Signal's move matter beyond just the messaging app?

Signal's decision to block AI screen access is a preview of a larger fight coming for every app that handles sensitive data. As platform-level AI features become more integrated into operating systems, developers across the industry will need to implement similar protective measures to safeguard user privacy and end-to-end encryption.

LIVE02:22Signal Adds Feature to Block AI From Reading Your Screen