Before she knew what a governance framework was, the speaker watched her father Billy Diamond run one. He managed Fats Domino, Shirley Goodman of Shirley and Lee, Smiley Lewis, and the 5-4 Ballroom simultaneously, under the scrutiny of an industry that was actively trying to exploit the artists he represented. He did not call it portfolio governance. He called it making sure nobody got taken advantage of on his watch. That is still the most accurate definition of AI governance she has ever heard.
The speaker was a program and project manager long before she held a single certification. She ran King County programs, directed a $386 million federally audited AI-integrated portfolio for the City of New Orleans, and maintained 100% compliance across three years of continuous US Treasury oversight. What kept that portfolio clean was not the governance document. It was the architecture — the specific design decisions that made every AI-assisted output traceable, every human accountability chain documentable, and every automated decision correctable before a federal monitor ever saw it.
This session presents the AI governance framework built from that experience, translated into the enterprise data strategy language that this conference's audience of senior data leaders, AI practitioners, and public sector technology executives will recognize and immediately apply. It addresses the three failure modes that cause AI governance initiatives to collapse in public sector and enterprise environments: governance added as an afterthought, human oversight treated as optional, and compliance frameworks designed for yesterday's regulations rather than today's AI realities.

Key Takeaways
• How to architect a human-in-the-loop accountability chain that satisfies federal audit requirements and scales across program areas, drawn from a $386M public sector AI portfolio with zero audit findings
• A practical AI governance integration framework for portfolio management offices and enterprise data teams, grounded in real execution rather than theoretical best practice
• The three structural failure modes that cause public sector and enterprise AI governance initiatives to collapse under scrutiny, and the design decisions that prevent each one
• How to build compliance infrastructure that protects both the organization and the communities it serves, with specific patterns for regulated industries, government, healthcare, and financial services