Graph AI for Visual, Actionable Enterprise Role Mining

Session Track: Graph + AI in Production

Session Time: 08:00 am - 08:30 am April 15

Session description

Traditional Identity and Access Management (IAM) systems struggle to analyze the complex access relationships across 20k to 30k identities and deeply nested permission hierarchies, often creating critical risk blind spots.

This presentation will demonstrate how the speaker and their team leveraged Graph AI to solve this challenge. The session will cover how to model complex identity-permission relationships as a graph, the design choices for schema optimization, and how to integrate graph visualization to make operational access decisions.

Attendees will learn how Graph AI can be applied to real-world identity governance, resulting in tangible benefits like a 30% query performance improvement and a significant reduction in role-management workload and operational access incidents.

Speaker

Michael Tai

Identity Access Management Senior Manager|Graph AI Innovator

Michael Tai is an IAM Senior Manager with a focus on applying Graph AI to enterprise security and governance challenges. He has extensive experience in designing scalable identity solutions and permission modeling for large-scale organizations. Most recently, he led the development of a Graph-based Role Mining AI tool that successfully visualized 30k+ identities, significantly reducing risk and operational overhead. He is passionate about democratizing complex access data through intuitive graph visualization.