Security brief

OpenShift AI privilege-escalation flaw required an authenticated account

REVIEW

CVE-2025-10725 could turn low-privilege access into cluster administration; Red Hat rates the issue Important.

Published 29 Oct 2026 · Material update 29 Sep 2025 — last time the underlying evidence changed · Evidence: high

What happened

Red Hat’s advisory for CVE-2025-10725 describes a privilege-escalation flaw in OpenShift AI. An attacker with an authenticated, low-privilege account could obtain cluster-administrator access, with consequences for applications and data across the cluster.

Red Hat rates the issue Important, explaining that authentication is required. That prerequisite should remain visible alongside the high numerical severity score rather than being lost in an “unauthenticated takeover” headline.

The advisory lists security errata for affected OpenShift AI streams. Operators should match their deployed stream and operator build to those errata, apply the corresponding fix and review unexpected changes to cluster permissions.

Read the original source.

Next step

Apply the security erratum for your supported OpenShift AI stream.

What we know

Prerequisite
Authenticated low-privilege access.

What remains unknown

Open question
The sources reviewed do not establish whether your deployment was exploited.

Source references

Technical details

CVE
CVE-2025-10725