A vulnerability in the training-operator component of Red Hat OpenShift AI allows authenticated users to bypass an immutability check on PyTorchJob resources. This flaw arises from incorrect object casting in the PyTorchJob ValidateUpdate webhook, enabling alteration of the spec.runPolicy.managedBy field after job creation. Exploitation could disrupt resource quota management or hijack job controller ownership, primarily causing a denial of service (DoS) condition. Currently, no effective mitigation or fix meeting Red Hat's criteria is available.