The US Government Wants You—Yes, You—to Hunt Down Generative AI Flaws

Humane Intelligence, in collaboration with NIST’s ARIA program, announced an open qualifying round for a nationwide red-teaming effort to evaluate AI office productivity tools, culminating in an in-person event at CAMLIS where red and blue teams will test systems against the AI 600-1 profile from NIST’s AI RMF; the initiative seeks broad public participation—including nontechnical and underrepresented groups—to advance transparency, security, resilience, and ethical evaluation of generative AI systems through structured, real-world testing.