Google reported a 0.0% attack success rate against Gemini 4 Carbon on Gray Swan's indirect prompt-injection benchmark, down from 0.7% for Gemini 4 Argon

Google DeepMind's frontier AI model, released on October 28, 2026, two weeks after Gemini 4 Argon's wide rollout, first only to selected cybersecurity defenders through Google's Fairwind Program; it ranked first on the Artificial Analysis Intelligence Index v5 with a score of 59.