AW AgentAIWatchPublic threat intelligence
Public sources only

INCIDENT REPORT

Formal Verification of the Signal Protocol Using AI

2026-11-17
unverified public report Unknown Unknown

This seems like a good thing: Google Deepmind researchers used agentic AI to mathematically prove, or “formally verify,” a number of the Signal protocol’s key cryptographic guarantees­a landmark development for the widely used cryptography and AI itself. Conducted as a sort of proof-of-concept project in just four weeks, the work is significant in two ways. It serves as further independent validation of the cryptography that protects massive quantities of the world’s communications on platforms like the Signal app, Google Messages, and WhatsApp. And it demonstrates Google DeepMind’s formal verification capabilities using a new AI framework called AlphaProof Code...

Risk rationale: agentic (+14) Source: schneier.com ↗
32Risk
24Confidence

← All incident reports