Arthur has launched its Agent Discovery & Governance (ADG) platform, building on its existing evals, guardrails and observability capabilities, Arthur said on 17 December 2025.
The platform discovers AI agents running across compute environments, builds an inventory of them and lets organisations apply governance controls.
Arthur said this applies whether agents were developed in-house or bought, and whichever AI provider powers them.
Co-founder and CEO Adam Wenchel said the rapid adoption of agents means CIOs now have to govern and secure the technology.
Arthur cited McKinsey research reporting that 80% of organisations already see risky behaviour from AI agents.
For regulated firms, an inventory of which agents are running, and where, is a starting point for applying model risk and security controls.