AI Agents Need Identity Governance Now

SailPoint's Mo Joueid on Treating AI Agents as First-Class Identities Tony Morbin (@tonymorbin) • March 16, 2026

AI agents are entering enterprise environments faster than governance frameworks can keep up. Many organizations discover agents not through planned rollouts but through breaches or audit findings, said Mo Joueid, principal solution consultant at SailPoint.

See Also: SASE and Zero Trust: The Backbone of Integrated Security (eBook)

Agents can't be managed like human identities. They interact with multiple systems, make autonomous decisions and operate across platforms simultaneously, creating identity patterns that traditional controls weren't built to handle. Joueid recommends classifying agents by their autonomy level, distinguishing a read-only adviser from a task executor with privileged access, to help organizations gauge risk before complexity gets out of hand.

"Ownership is not just a governance concept, it's a fundamental control," he said.

In this video interview with Information Security Media Group at ...