Tope Iluyomade has been working in AI since 2004, starting with robotics and reinforcement learning, and has spent two decades advising and leading technology inside large enterprises as both CIO and CISO before moving into executive advisory work. In this conversation, he breaks down why most AI governance fails at the cultural level rather than the technical one, how the CIO-CFO conversation has fundamentally shifted from deployment updates to P&L accountability, and where human judgment still has to override what an AI model would otherwise optimize for.
**Topics Discussed:**
- Why AI belongs in the decision loop, not the tool stack
- Two-layer AI portfolio model: foundational access for all, just-in-time specialization with proof of outcomes required before renewal
- How CIOs are shifting the CFO conversation from deployment updates to P&L accountability
- The "AI flight school" model for CIO and CHRO co-designing enablement curriculum
- The greedy algorithm failure mode: when AI optimization overrides human judgment
- Data masking and tokenization as the practical middle ground on AI data risk
- Legal and facilities: the two functions still largely untouched by AI, and why
- Why operators are replacing executive committees in AI experiment sign-off
- SaaS split: ecosystem winners versus commoditized casualties, and how MCP architecture is shifting the build-versus-buy math
Tope IluyomadeGuestAI Impact HostHost
