Google DeepMind's AI agents have demonstrated new methods that reduce the computational cost of attacking obsolete 1024-bit RSA cryptosystems, though modern encryption protocols using 2048-bit or larger keys remain secure. The research highlights AI's evolving
Google has rolled out its advanced AI model, Gemini 4 Argon, to trusted cyber defenders through its Fairwind Program. The model shows significant leaps in autonomously discovering, validating, and patching critical software vulnerabilities.