🔍 Read the full analysis: Signal Blocks Screenshots: A Look At Its AI Agent Concerns on ThorstenMeyerAI.com
Get business pricing on tech for your team
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
Signal says recent app versions add controls against screenshots and some on-device AI assistants reading chats. The protections work differently on iPhone and Android and do not cover every device, capture method or copy of a conversation. Signal is asking operating-system makers for a documented way for apps to restrict AI access.
In a post published **November 20**, Signal said it is restricting AI agents built into mobile operating systems from potentially capturing and storing Signal data. Its support page says the features also hide Signal content from the iPhone App Switcher or Android recent-apps list while protection is enabled, and block screenshots and screen recordings within the app. Signal says the setting does not change message encryption or who can access messages through other means.
On iPhone, the **“Block Siri AI”** setting is on by default for Signal **8.37 or later** running on **iOS 27 or later**. Users can set it to On, Off or “Off for 10 Minutes”; the temporary option turns protection back on automatically. Signal says users who try to take a screenshot will see a notice and can choose to disable the protection temporarily or leave it on.
On Android, **Signal 8.35.1 or later** can turn on Screen security the first time the app detects an AI assistant trying to read its screen. Users can also manage it under Settings, Privacy, App security. Signal says that if a user disables the feature from the warning prompt, it will turn back on after **90 days**; disabling it through settings does not prompt that reminder. The Android feature depends on Signal detecting the assistant attempt.
Limits on AI Access to Chats
The change addresses a gap between **message encryption in transit** and privacy after a message appears on a device. End-to-end encryption protects messages as they move between people, but does not by itself prevent software with screen access from viewing content displayed on an unlocked phone. Signal argues that captured screen data may leave the protections the app controls.
That concern matters beyond individual users. Companies deploying **screen-reading agents** may encounter more apps that restrict automated access, while users who rely on assistants for device tasks may find some actions blocked or require them to switch off protections. Signal’s controls are not a universal ban on AI access: they target specific screen-capture and on-device reading paths, and their reach varies by platform.
Signal is also pressing Apple, Google and Microsoft for a **documented app-level opt-out** from AI data access. A standard operating-system control could give app developers a clearer way to express privacy restrictions than platform-specific detection or screenshot blocking. The companies’ responses to that request were not included in the source material.
Products Worth Considering

LG 27GS60QC-B Ultragear 27-inch Curved Gaming Monitor (QHD, 180Hz, 1ms, 1000R)

Apple MacBook Air 13-inch Laptop with M5 Chip, 16GB Memory, 512GB SSD, Liquid Retina Display, and 12MP Center Stage Camera (Midnight)

Bose QuietComfort Ultra Wireless Noise-Cancelling Over-Ear Headphones with Spatial Audio, White Smoke
As an affiliate, we earn on qualifying purchases.
From Recall to Mobile Assistants
Signal has previously added screen protections in response to automated capture. On **May 21, 2025**, it enabled Screen security by default in Signal Desktop on Windows 11 to guard against Microsoft Recall. Signal’s account says Recall periodically captures the screen, extracts text using optical character recognition and stores it in a searchable on-device database.
The current changes target newer phone assistants. Apple described **Siri AI**, released September 14, 2026, as having “onscreen awareness” that can answer questions or take actions related to screen content. Google announced **Gemini Intelligence** on May 12, 2026, saying app automation can use screen or image context. Signal’s latest measures respond to the possibility that assistants can use visible app content, rather than to a change in Signal’s message encryption.
Signal’s president, **Meredith Whittaker**, called operating-system AI agents “an existential threat to Signal, and our ability to responsibly uphold our mission,” according to the company post. She also characterized Signal’s response as “incomplete.” Reuters reported the announcement on November 20. The supplied reporting also cites a separate Reuters report about Meta’s Muse agent accessing Signal messages through permissions granted during setup; the details and scope of that report are not established by the Signal support information summarized here.
Coverage Depends on Device
Signal says the protections do not cover **Mac computers**, where it cannot reliably block screenshots. On Android, the automatic response depends on the app detecting an assistant’s attempt to read the screen; the source does not identify every assistant, device configuration or access method that detection covers.
The controls also do not protect **notifications, files or the other person’s phone**, according to Signal’s published limitations. It is not clear from the supplied material whether every form of operating-system recording or third-party capture is blocked in all circumstances. Signal’s description should not be read as a guarantee that chats cannot be copied or seen once displayed.
The source material does not include a response from Apple, Google or Microsoft to Signal’s request for an opt-out, nor independent testing of how reliably the features work across devices. It also does not establish the full scope of the separate Reuters report concerning Meta’s Muse agent.
OS Makers Face Signal’s Request
Signal says it wants **Apple, Google and Microsoft** to provide a simple, documented way for apps to opt out of AI data access. No timetable or commitment from those companies is given in the source material. Until a shared control exists, the available protections remain tied to Signal’s app versions, operating systems and ability to detect screen access.
Users can check Signal’s privacy settings and decide whether to leave the protections on. Signal warns that switching them off may expose chats displayed on the device to AI assistants. The next developments to watch are any operating-system changes, updates to Signal’s supported-device guidance and further details about how reliably the protections work in practice.
Key Questions
What did Signal change?
Signal added controls intended to block screenshots and screen recordings in the app, hide chats from recent-app views and prevent some on-device AI assistants from reading the screen. The exact behavior differs between iPhone and Android.
Are the protections on by default?
Signal says **Block Siri AI** is on by default on supported iPhones running iOS 27 or later. On Android, Screen security activates automatically when Signal detects an assistant’s attempt to read the screen; users can also enable it themselves.
Can users turn the protections off?
Yes. On iPhone, the setting offers On, Off and Off for 10 Minutes. On Android, users can switch Screen security off in Signal’s privacy settings. Signal warns that disabling it may expose displayed chats to AI assistants.
Does this protect every copy of a Signal conversation?
No. Signal says the controls do not cover Mac screenshots, notifications, files or the phone of the person receiving the message. The company also calls its response incomplete, and the source does not establish that every screen-reading method is detected.
Source: ThorstenMeyerAI.com
