As AI systems evolve from simple scripts to autonomous decision-makers, a new kind of insider risk is emerging. In this AI Journal feature, Rosario Mastrogiacomo, Chief Strategy Officer at SPHERE, examines how these AI agents—once built to streamline workflows—are beginning to operate with a level of independence that challenges traditional governance, accountability, and security frameworks.
The article explores how enterprises are losing visibility over these “AI identities,” which can act, reason, and adapt without explicit human oversight. It calls for stronger governance, transparency, and lifecycle management to ensure autonomy doesn’t turn into an uncontrolled threat operating from within the organization.
