EP 23 - Red teaming AI governance: catching model risk early

AI systems are moving fast, sometimes faster than the guardrails meant to contain them. In this episode of Security Matters, host David Puner digs into the hidden risks inside modern AI models with Pamela K. Isom, exploring the governance gaps that allow agents to make decisions, recommendations, and even commitments far beyond their intended authority.

Isom, former director of AI and technology at the U.S. Department of Energy (DOE) and now founder and CEO of IsAdvice & Consulting, explains why AI red teaming must extend beyond cybersecurity, how to stress test AI governance before something breaks, and why human oversight, escalation paths, and clear limits are essential for responsible AI.

The conversation examines real-world examples of AI drift, unintended or unethical model behavior, data lineage failures, procurement and vendor blind spots, and the rising need for scalable AI governance, AI security, responsible AI practices, and enterprise red teaming as organizations adopt generative AI.

Whether you work in cybersecurity, identity security, AI development, or technology leadership, this episode offers practical insights for managing AI risk and building systems that stay aligned, accountable, and trustworthy.