In this EY.AI podcast series, Abbas Godhrawala, Partner, Risk Consulting at EY India, explains why Agentic AI — systems capable of autonomous decision-making and real-time actions — demands a fundamentally different risk framework compared with traditional AI models.

Abbas outlines how the autonomous nature of Agentic AI introduces new layers of complexity: increased exposure to adversarial threats, unpredictable behavior across interconnected systems and reduced human intervention. He discusses why current governance models fall short and how organizations must rethink their oversight mechanisms to keep pace with rapidly evolving AI capabilities.

He also distills three foundational priorities from the EY Agentic AI risk framework that enterprises should embed immediately: enhanced observability, rigorous testing structures and strong human oversight to ensure accountability and system integrity.

Tune in to understand how a future-ready risk framework can help organizations safely scale Agentic AI.