Mr. Nandagopal Seshagiri

Trust, Identity, and Access Management in the World of AI

Abstract:

As AI systems become increasingly embedded in critical infrastructure and enterprise workflows, the boundaries of trust, identity, and access are rapidly evolving. Traditional IAM models, designed around human users and static policies, are no longer sufficient in a world where machines generate identities, make decisions, and initiate actions autonomously.

In this talk, Nandagopal Seshagiri explores how the rise of AI agents, LLMs, and autonomous systems challenges the foundations of identity and access management. He examines what it means to establish trust when identities are synthetic, interactions are opaque, and decisions may be derived from probabilistic reasoning. The session covers key shifts needed in authentication, authorization, and trust modeling, including the role of continuous validation, verifiable provenance, and dynamic policy enforcement.

Whether you're building AI-enabled platforms or securing access in AI-driven environments, this talk offers a strategic perspective on adapting IAM to a future where trust is not just granted — it must be earned, contextualized, and auditable in real time.