White Paper: Operationalizing Zero Trust for AI Systems

Artificial intelligence—particularly generative and agentic AI—is rapidly reshaping how federal agencies operate, deliver services, and execute mission objectives. Across civilian, defense, intelligence, and research environments, agencies are moving from experimentation to scaled adoption. This acceleration is introducing meaningful gains in efficiency, responsiveness, and analytical capability. At the same time, it is fundamentally altering the cybersecurity landscape.

This whitepaper synthesizes insights from a not-for-attribution federal roundtable involving senior leaders and practitioners. A central conclusion emerges: the nature of risk is shifting. Cybersecurity is no longer limited to protecting data and systems. It now extends to governing the behavior of autonomous and semi-autonomous systems operating at machine speed.