Description
A flaw was found in the Red Hat OpenShift AI (RHOAI) overlay for the training operator. The RHOAI overlay incorrectly aggregates `trainjobs` management permissions into the native Kubernetes `edit ClusterRole`. This allows any user with `edit ClusterRole` permissions in a namespace to create, modify, and delete `TrainJobs`. When combined with a separate vulnerability (TRN-01) that permits arbitrary pod configurations, a remote attacker with namespace editor privileges could exploit this to escalate privileges, potentially leading to arbitrary code execution.
Affected products
Version statuses are supplied by the assigning organization. Consult the original record for detailed ranges and platform conditions.
Red Hat Red Hat OpenShift AI 3.3
Default version status: affected
1785188461 to < *unaffected
Red Hat Red Hat OpenShift AI 3.3
Default version status: affected
1790137153 to < *unaffected
Red Hat Red Hat OpenShift AI 3.4
Default version status: affected
1784814352 to < *unaffected
Red Hat Red Hat OpenShift AI 3.4
Default version status: affected
1787361677 to < *unaffected
Red Hat Red Hat OpenShift AI 3.5
Default version status: affected
1786552271 to < *unaffected
Red Hat Red Hat OpenShift AI 3.5
Default version status: affected
1786552250 to < *unaffected
References
- https://access.redhat.com/errata/RHSA-2026:53262 ↗
- https://access.redhat.com/errata/RHSA-2026:53263 ↗
- https://access.redhat.com/errata/RHSA-2026:60367 ↗
- https://access.redhat.com/errata/RHSA-2026:60520 ↗
- https://access.redhat.com/errata/RHSA-2026:73987 ↗
- https://access.redhat.com/security/cve/CVE-2026-18951 ↗
- https://bugzilla.redhat.com/show_bug.cgi?id=2511187 ↗