As enterprise AI matures beyond pilots and prototypes, the conversation around governance is shifting from control to ownership: who decides what stays centralized, and who owns the context that drives decisions. In the concluding installment of a three-part series, Arun Nandi, Chief Data Officer (CDO) at Carrier Global, joins Or Zabludowski, Co-Founder and CEO of Flexor, to unpack how enterprises should structure their AI governance models.

Across the conversation, four themes stand out:

• Governance structure: Nandi centralizes functions like observability, identity and access management, and agent lifecycle management, while leaving context ownership with domain owners.
• Knowledge commoditization: Domain knowledge once locked inside teams and departed employees is increasingly accessible through large language models, which Nandi calls a “great equalizer.”
• Builder culture: Referencing Jack Dorsey’s builder, directly responsible individual, and player coach archetypes, Nandi argues organizations need more builders, including senior leaders writing code again.
• Executive buy-in: Zabludowski and Nandi agree that working prototypes now convince stakeholders faster than slide decks, compressing timelines and raising expectations for AI teams.

Together, these themes describe a broader shift toward governing the people and processes that build AI. Nandi treats centralization and decentralization as complementary rather than competing forces. More enterprises are likely to face the same balancing act as AI moves from experimentation into daily operations.

Did you miss the previous episode?

On the move? Take this conversation with you. If you don’t have time to finish the video, you can stream the complete, uninterrupted audio edition of this episode on the go.