Event Date

December 10, 2026

Registration

Session Summary

AI runs on data. That is precisely why AI implementation keeps landing on the data team’s desk, right after the data quality problem, right before the compliance problem, and often without anyone ever deciding it should live there. It is the easiest place to put it. It is rarely the right place to put all of it.

This session examines what happens when organizations collapse AI implementation into data governance simply because data governance is the structure they already have. Data leaders can quickly become accountable for model behavior, decision outcomes, human use, regulatory interpretation, and downstream consequences they were never given the mandate, expertise, or authority to control. AI may be described as cross-functional in the strategy deck while being managed as a data function in practice.

During this working session, participants will take an AI use case and separate the distinct responsibilities involved in implementation: the data itself, including quality, lineage, provenance, access, and stewardship; the model, including selection, evaluation, limitations, and monitoring; the business process and purpose; the role of human judgment; legal and risk responsibilities; and executive ownership of what the organization is ultimately authorizing AI to do. The distance between responsibility and authority is where AI implementations can quietly fail, even when the data is accurate, the model performs within tolerance, and no individual control appears to have malfunctioned.

Built for participants ranging from early-career analysts to senior data leaders, this is a working session rather than a three-hour lecture. I will use structured exercises, live case mapping, and group discussion so both the in-person and virtual participants can actively work through the concepts and apply them to situations they are likely already encountering inside their organizations.

Speaker Bio

Leigh Felton is President and Chair of the AI for Job Security Foundation and creator of the Co-Adaptive Anthropotechnical Environments (CAAE) framework, which examines AI as an environment that reshapes human behavior, organizational authority, decision-making, and accountability rather than simply as a tool operating inside an existing process. Her work focuses on where responsibility for AI implementation actually lands inside institutions, how authority shifts when AI enters a decision process, and what organizations fail to see when those boundaries are left undefined.

Her perspective is grounded in nearly two decades of leadership across technology, public-sector transformation, and organizational strategy, including Responsible AI leadership at Microsoft. At Microsoft, she served in senior leadership roles including Chief of Staff and Head of Responsible AI Enablement, where she helped build and scale the company’s Responsible AI Champs model to translate Responsible AI principles into organizational practice. She previously held Chief of Staff and senior leadership roles at Mozilla and within Washington State government. At Microsoft, she served in senior leadership roles including Chief of Staff and Head of Responsible AI Enablement, where she helped build and scale the company’s Responsible AI Champs model to translate Responsible AI principles into organizational practice. She previously held Chief of Staff and senior leadership roles at Mozilla and within Washington State government.