Governing the Autonomous Enterprise: GRC in the Era of Agentic AI
Thursday, November 12, 2026
09:00–10:30 AM
Enterprises are moving rapidly from AI systems that predict to AI systems that act.
Agentic AI, meaning LLM-based agents that plan, call tools, hold memory, and delegate to other agents, is already being deployed in finance, healthcare, logistics, and customer operations. Governance, Risk, and Compliance (GRC) frameworks were designed for deterministic software and human decision-makers. They assume static permissions, point-in-time audits, and a clear line of accountability. Autonomous agents break all three assumptions.
This workshop presents a practical, research-informed framework for extending GRC to agentic systems. Participants will learn to (1) classify systems on a five-level autonomy spectrum and identify where existing controls stop working, (2) recognise five agent-specific risk categories absent from conventional risk taxonomies (agent drift, hallucination at scale, cascading delegation failure, shadow agents, intent misalignment), (3) design layered guardrail and control architectures (permissions, sandboxing, kill switches, escalation), and (4) map these to the EU AI Act, NIST AI RMF, and ISO/IEC 42001. Two hands-on exercises using a realistic case scenario let participants build an agent risk register and score their organisation on an agentic GRC maturity model. The session closes with open research questions for the information security community
Workshop at a Glance:
FORMAT: 90-minute interactive workshop. Four short talk blocks, two hands-on exercises, live poll, closing research discussion.
AUDIENCE: Mixed academic and industry: information security and AI researchers, CISOs, CTOs, GRC and risk leads, AI engineers. No coding required.
TAKEAWAYS: Agent risk register template, guardrail and control checklist, EU AI Act / NIST AI RMF / ISO 42001 mapping sheet, maturity self-assessment, research agenda.
Dr. Sam Mokhtari is a technology leader, author, and advisor with twenty years spanning academic research, enterprise consulting, hyperscale cloud, and executive leadership. As CTO of Rezilens, a Dubai-based GRC technology company, he leads engineering, platform, and agentic-AI product work at the intersection of cybersecurity, risk, and compliance. He has advised more than 100 organisations across the Middle East, Asia-Pacific, and Europe on adopting data, AI, and cloud at scale.
He holds a PhD in Electrical Engineering from Queensland University of Technology and serves as Industry Expert at the University of Dubai and Adjunct Fellow at Macquarie University, contributing to AI, cybersecurity, and information systems programmes aligned with enterprise practice.
Publications and Credentials
Author of three books on agentic AI, cloud AI infrastructure, and GRC in the agentic era.