Whittaker warns against the privacy threat posed by agentic AI

During her speech at the AI for Good Summit in Geneva on 8 July, Meredith Whittaker, President of the encrypted messaging app Signal, raised concerns about the risks posed by the latest advance in artificial intelligence (AI) technologies, Agentic AI. Whittaker warns that while these systems promise convenience, they can gather huge amounts of sensitive user data, potentially compromising the privacy that Signal provides to its millions of users, including those in government and the human rights sector.

Whittaker explained that even simple tasks, like booking a restaurant, require an AI agent to access personal data from calendars, credit cards, contacts, and messaging apps like Signal. This creates a "detour" that could undermine Signal's robust privacy by breaching the "blood-brain barrier between the operating system and the application layer."

Whittaker stressed that all application-layer technologies, including platforms like Spotify, face a competitive threat from Agentic AI accessing proprietary data. To mitigate these risks, Whittaker is calling for developer-level opt-outs to block AI agents from accessing specific apps entirely. She also emphasised the need for open, verifiable agentic systems to enable rigorous security engineering, particularly for integration into sensitive government and military infrastructures.

Training Announcement: Freevacy offers a range of independently recognised professional AI governance qualifications and AI Literacy short courses that enable specialist teams to implement robust oversight, benchmark AI governance maturity, and establish a responsible-by-design approach across the entire AI lifecycle. Find out more.