In the era of artificial intelligence, organizations face a surge in machine identities. According to the 2026 Identity Security Landscape report by Palo Alto Networks Inc., companies manage an average of 109 machine identities per human. Projections indicate that AI agent identities will increase by 85% over the next 12 months.
Research from SailPoint Technologies Inc. highlights significant security gaps, noting that 97% of AI agents have access to sensitive data, while only 21% of organizations feel confident in managing these risks. A recent proof-of-concept at a Fortune 500 company identified over 10,000 unknown AI agents, underscoring the rapid spread of shadow IT.
To address these challenges, SailPoint launched Agentic Fabric in May, which reached general availability in August. The solution, integrated into the Atlas platform, provides discovery tools for AI agents, maps them to human owners, and includes features such as browser sensors, data redaction, and a centralized kill switch for unauthorized agents.
Industry experts emphasize that containment methods like sandboxes are insufficient without robust identity governance. SailPoint CEO Mark McClain stresses the necessity of identifying data access, human accountability, and revocation mechanisms for every agent. The company plans to discuss these security strategies at the Navigate conference in Austin.
