AI security risks demand stronger identity governance

Rapid AI Adoption Creates New Vulnerabilities

A recent SailPoint survey reveals 82% of companies now deploy AI agents across multiple business functions. These autonomous systems perform tasks previously requiring human oversight, delivering unprecedented efficiency but also introducing new risks:

• 90% of companies report AI agents taking unintended actions
• 23% have experienced credential exposure from AI systems
• 60% report unauthorized access to privileged data

Identity Security Becomes Critical

SailPoint emphasizes that traditional access control is no longer sufficient. Organizations must now track:

• What is acting in their environment
• How systems are interacting
• Why access is being used

Current safeguards remain inadequate:

• Only 44% have formal AI governance policies
• Just 52% can track data accessed by AI agents

"The smarter AI gets, the more human its risks become," the article warns.

SailPoint’s Solution

The company offers:

1. Identity Security Cloud: Governs all identities at enterprise scale
2. Harbor Pilot: AI-powered security agent that:
   • Recommends access decisions
   • Flags anomalies
   • Automates repetitive tasks
   • Learns from context

Strategic AWS Partnership

Key developments include:

• AWS-based SaaS instances in Dubai, APAC, and Brazil
• FedRAMP-authorized solutions for U.S. government
• Integration with Amazon Bedrock for automated SaaS connector code generation
• Availability in