Employees are increasingly bypassing formal IT channels to build agents via low-code studios and SaaS-native tools. These automated workers often hold sensitive credentials and data access without undergoing security reviews. While traditional identity and endpoint tools like Okta or CrowdStrike can monitor behavior, they lack the business context required to distinguish between critical infrastructure and security liabilities. Shreyas Sadalgi, CEO of Amplifier Security, notes that the platform aims to reduce risk without triggering the productivity-killing disruptions often caused by bulk revocation.
Amplifier Security Launches AIPM to Manage Shadow AI Agents
With 82% of organizations reporting the discovery of unauthorized AI agents in the last year, Amplifier Security is launching Agent Intent & Posture Management (AIPM). The tool forces a bridge between unmanaged AI workflows and corporate oversight by requiring employees to verify the purpose of the agents they build.
AIPM automates the governance lifecycle by discovering agents directly on endpoints and triggering a verification flow through Slack or Microsoft Teams. Owners are prompted to confirm their agents’ intended use, which is then logged alongside risk indicators and behavioral data. This allows security teams to make informed decisions on whether to restrict or retire specific tools based on actual work dependencies rather than blind policy enforcement. By integrating with existing GRC and audit infrastructure, the platform provides a defensible record of AI governance for organizations struggling to keep pace with decentralized innovation. AIPM is currently available in private preview.
Comments (0)
No comments yet. Be the first!