The system divides labor between human specialists who establish mathematical specifications and AI agents that write and check formal proofs. Using Gemini Carbon-powered agents, the framework generated over six million lines of Lean proof language before human experts audited flagged sections of code. This hybrid process verified compiled, device-level software in just four weeks instead of the multiple years normally expected for such intensive tasks. The framework also successfully proved the insecurity of code when tested with reintroduced legacy bugs.
Key points
AlphaProof Code is the first method to formally verify a widely deployed cryptographic pr…: AlphaProof Code is the first method to formally verify a widely deployed cryptographic protocol at this level of depth.
The system reliably identified vulnerabilities by proving that modified software containi…: The system reliably identified vulnerabilities by proving that modified software containing two reintroduced bugs was insecure.
Google DeepMind is open-sourcing the verification artifacts to assist in securing systems…: Google DeepMind is open-sourcing the verification artifacts to assist in securing systems like power grids and medical devices.
Sources checked by our analysis pipeline: 1
How Google DeepMind researchers built a new agentic framework that formally verified the code of Signal’s cryptographic protocol — long considered to be a grand security challenge
Today, we’re sharing a milestone in software security long considered to be a grand security challenge: Google DeepMind researchers guided AI agents to mathematically prove key security properties of the actual code underlying the Signal protocol, the technology that protects billions of private conversations in apps like Signal and Google Messages.
This breakthrough discovery was made using AlphaProof Code, a novel agentic framework for formal verification at unprecedented depth that combines human expertise with AI agents. Formal verification—mathematically proving code runs correctly without flaws—is the highest standard of security assurance; however, it is difficult and often labor-intensive.
Our work marks the first time AI has been used to formally verify a widely deployed cryptographic protocol at this depth. It's also the first time the implementation of a widely deployed protocol like this has been formally verified by any method, with or without AI.
The maker of the Claude chatbot is pushing ahead with its public debut despite operating at a loss and confronting wider market turbulence. This move tests investor appetite amid political concerns following the US midterm elections that a Democrat-led Congress could curb artificial intelligence development. The high-stakes debut will serve as a bellwether for whether massive capital investments in sector infrastructure can ultimately translate into sustainable corporate profits.
The high-profile listing will test investor appetite for artificial intelligence leaders that are delivering explosive revenue growth alongside steep bottom-line losses. Founded by former OpenAI researchers, the enterprise is confronting a more skeptical market environment shaped by interest rate pressures and regulatory scrutiny. The market debut on the Nasdaq represents a defining moment for commercial AI infrastructure financing.
The artificial intelligence firm is currently exploring plans to list its shares on the public market. If executed, the transaction would place the organization alongside the most substantial market debuts recorded globally. Financial specifics remain under wraps as strategic discussions continue.