The Verge

AI, Security & Privacy

Google Launches Gemini 3.5 Flash Cyber, a Lower-Cost AI Security Model to Rival Anthropic’s Mythos

Google’s July 2026 Gemini 3.5 Flash Cyber targets lower-cost vulnerability finding, running via CodeMender where it can be called multiple times at high speed and low cost. On CyberGym it showed “competitive performance” up to 5 invocations; on V8 it confirmed 55 issues vs 47 (Gemini 3.5 Flash) and 36 (Opus 4.6), including 10 no other model found.

AI, Security & Privacy

1Password Integrates With Claude to Let AI Fill Login Credentials on Your Behalf

1Password’s July 2026 “1Password for Claude” lets Anthropic’s Claude complete multi-step tasks using stored logins after per-task biometric approval. Its “zero-exposure” channel injects credentials Claude can’t view and it scans pages after autofill. Available now on Mac for business, family, and individual plans, with both 1Password and Claude desktop/browser extensions.

Scroll to Top