We actively evaluate our products using AI-assisted methods as part of our structured security engineering program, in Commvault-controlled environments, under the same governance as every other form of testing.

Our program is model-agnostic and tool-agnostic by design, so we can incorporate new methods inside one consistent governance framework.

Yes. All evaluation happens on isolated hardware or Commvault-managed cloud infrastructure. Source code never leaves our boundaries, and every AI-generated finding requires human confirmation before action.

We’re investing in risk-based triage and remediation infrastructure so the response process can scales with discovery volume, not just the discovery itself.