The Agentic Turn

AI has moved from answering questions to taking action. Agentic systems now plan, research, decide, and execute on behalf of organizations — drafting, communicating, hiring, trading — without human approval at each step.

Most institutions are unprepared. They have no governance framework for delegating to AI agents, no accountability structures for when agents fail, and no shared language for what "oversight" even means in practice. The gap between what agentic AI can do and what organizations know how to govern is growing fast.

This talk maps the real shift: what agentic computing actually is, what it isn't, and what leaders must build now — before the defaults get set by vendors, regulators, or avoidable failure.

Who this is for

Executives, boards, legal and compliance teams, policy makers, and any organization actively evaluating or deploying AI systems — especially in regulated industries or public-facing institutions where accountability and oversight are non-negotiable.

Audience takeaways

• What "agentic AI" actually means — and how to tell the difference between marketing and real capability
• Where agentic systems create genuine value versus where they create unacceptable risk and liability
• What institutional governance for AI agents looks like in practice
• How to design oversight structures before something goes wrong
• The policy and regulatory landscape taking shape around autonomous systems — and how to get ahead of it

Sample titles

• The Agentic Turn
• When AI Acts: Governing Autonomous Systems
• Delegating to Machines: What Leaders Need to Know

Best formats

Keynote; executive briefing; workshop on AI governance and risk; fireside conversation with Q&A.