Report
The State of Identity in the AI Era
A survey of 161 security and identity leaders: 96% can't fully account for the AI agents already running in their environment.

AI agents are already doing real work in the enterprise, but the leaders responsible for them can't see them, can't stop them, and plan to fix it with tools they already own.
We surveyed 161 security and identity leaders, most at companies with more than 1,000 employees. The report is for CISOs, IAM and PAM teams and identity architects who own access for AI agents. It shows how many agents go unaccounted for, how fast teams can revoke one, how agents are managed today, where budgets are heading over the next 24 months, and what moving from standing to just-in-time access looks like.
What's inside
- How your peers see the threat.
- A benchmark for your own program.
- Where the market is investing.
- The readiness gap, quantified.
More from the library
The full library
White paper
Privileged Access Management for the Agentic Era
Only 4% of organizations keep a centralized, automated inventory of their AI agents. This paper sets out a five-layer standard for governing every AI agent as a privileged identity.

Webinar
The 2026 NHI & AI Breach Report and the new PAM playbook
Most of 2026’s non-human identity and AI agent breaches started with a valid credential. Venice Field CTO Chris Maroun and Lalit Choda, founder of NHI Mgmt Group, unpack the breaches and the 2027 PAM playbook.

Webinar
Architecting Zero Standing Privilege for AI agents
Zero Standing Privilege used to be the goal while “good enough” was the reality. AI agents end that compromise. Learn why it is now mandatory and what access management built for AI agents has to include.