JumpCloud Inc. has expanded its Agentic IAM service to provide IT teams with management tools for autonomous AI agents. Each agent is now assigned to a specific human owner, with the platform tracking the lifecycle of every agent from creation to retirement. Unassigned agents are blocked from execution.
The update introduces four key capabilities, including unique identity assignment for agents, which allows administrators to audit actions or revoke access independently of the deploying employee. The platform also routes Model Context Protocol servers through the JumpCloud AI Gateway to monitor unofficial channels.
New security features include the extension of device trust and access policies to AI tools, alongside a privileged access management server expected by the end of the fourth quarter. This server aims to replace shared API keys, allowing privileged access rules to expire automatically once a specific task is completed.
According to a July survey by JumpCloud, only 21% of organizations have implemented governance for nonhuman identities, despite these entities outnumbering human employees in 83% of surveyed firms. The company previously launched its Agentic IAM service in April.
Based in Louisville, Colorado, JumpCloud has secured $408 million in funding across 10 rounds. The company's last major investment was a $159 million Series F round led by Sapphire Ventures five years ago, which valued the business at $2.56 billion.
