A vulnerability in the RHOAI training-operator allows users with standard edit or admin roles in any Kubernetes namespace to escalate privileges. By creating training jobs, attackers can impersonate service accounts, access the host filesystem, and potentially execute arbitrary code remotely. This issue is due to the aggregation of training job permissions onto native Kubernetes edit and admin ClusterRoles combined with unrestricted PodTemplateSpec passthrough. The vulnerability has a high severity CVSS score of 8.8. Red Hat has released updated images for Red Hat OpenShift AI to address this issue and provides documentation for upgrading clusters to mitigate the risk.