Signal, the messaging app known for its encryption and privacy protections, is sounding the alarm about AI agents built into computer and smartphone operating systems.
In a blog post published Friday, Signal President Meredith Whittaker announced new protections to prevent AI assistants from accessing users’ private messages. On iOS 27, Signal will block screenshots by default, while on supported Android devices, it will automatically enable its existing Screen Security feature when it detects an AI assistant attempting to read the screen. Signal acknowledged that these protections won’t work on every Android device and that it currently has no reliable way to block AI assistants from capturing screenshots on macOS.
The move comes as tech companies are increasingly integrating AI agents into their operating systems, including Google’s Gemini Intelligence and Apple’s Siri AI. These agents are designed to connect with users’ apps and handle tasks for them, like responding to messages, booking appointments, shopping, and managing calendars. But Signal warns that giving these agents access to sensitive information could expose private conversations and pose an “existential threat” to the company.
“Now, in service of AI agentic capabilities, your sensitive data is being captured, stored, and made accessible to whatever local database or remote server an OS-level AI agent is using at that moment,” Whittaker wrote.




