Varonis Launches Agent Intent-Based Access Control to Mitigate AI Risks
Event summary
- Varonis introduced Agent Intent-Based Access Control (IBAC) on August 3, 2026.
- The new capability is integrated into Varonis Atlas and monitors AI agent behavior for policy compliance.
- Agent IBAC detects intent drift, evaluates full sessions, and can quarantine identities or sessions post-violation.
- The feature supports existing AI tools like Claude Code, Cursor, GitHub Copilot, and Microsoft Copilot Studio.
The big picture
Varonis' new offering addresses growing concerns around AI agent risks, particularly as non-human identities require broad data access. The launch reflects a broader industry shift toward securing AI from code to runtime, with enterprises increasingly prioritizing safeguards against unintended or malicious AI behavior. This move positions Varonis as a key player in the evolving landscape of AI security and governance.
What we're watching
- Adoption Pace
- How quickly enterprises will integrate Agent IBAC into their existing AI workflows.
- Competitive Response
- Whether competitors will introduce similar safeguards for AI agent access control.
- Regulatory Impact
- The extent to which Agent IBAC aligns with or influences emerging AI governance frameworks.
Related topics