GraySwan
An organization focused on AI safety red teaming. It constructs adversarial inputs to evaluate a model's ability to resist prompt injection attacks.
Timeline (last 90 days)
According to Google's results on the GraySwan benchmark, Gemini 4 Argon has the lowest attack success rate among all tested models.
GraySwan is an organization focused on AI security red teaming, and its benchmark evaluates models' ability to resist injection attacks by constructing large numbers of adversarial inputs.
All Facts (2)
According to Google's results on the GraySwan benchmark, Gemini 4 Argon has the lowest attack success rate among all tested models.
50%UnverifiedGraySwan is an organization focused on AI security red teaming, and its benchmark evaluates models' ability to resist injection attacks by constructing large numbers of adversarial inputs.
50%