---
id: CVE-2026-18951
title: >-
  A flaw was found in the Red Hat OpenShift AI (RHOAI) overlay for the training
  operator
summary: >-
  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 …
severity: high
cvss: 8.8
cvssVector: 'CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H'
cwe:
  - CWE-284
vendor: Red Hat
product: rhoai/odh-training-operator-rhel9
affected:
  - rhoai/odh-training-operator-rhel9 (all versions)
  - rhoai/odh-training-operator-rhel9 (all versions)
  - rhoai/odh-training-operator-rhel9 (all versions)
  - rhoai/odh-latency-predictor-prediction-rhel9 (all versions)
  - rhoai/odh-latency-predictor-training-rhel9 (all versions)
published: '2026-08-10'
updated: '2026-09-21'
sourceUpdated: '2026-09-21T11:17:08.520'
source: NVD
sourceUrl: 'https://nvd.nist.gov/vuln/detail/CVE-2026-18951'
references:
  - url: 'https://access.redhat.com/errata/RHSA-2026:53262'
    label: secalert@redhat.com
  - url: 'https://access.redhat.com/errata/RHSA-2026:53263'
    label: secalert@redhat.com
  - url: 'https://access.redhat.com/errata/RHSA-2026:60367'
    label: secalert@redhat.com
  - url: 'https://access.redhat.com/errata/RHSA-2026:60520'
    label: secalert@redhat.com
  - url: 'https://access.redhat.com/security/cve/CVE-2026-18951'
    label: secalert@redhat.com
  - url: 'https://bugzilla.redhat.com/show_bug.cgi?id=2511187'
    label: secalert@redhat.com
  - url: >-
      https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-18951.json
  - url: 'https://www.cve.org/CVERecord?id=CVE-2026-18951'
  - url: 'https://nvd.nist.gov/vuln/detail/CVE-2026-18951'
tags:
  - nvd
  - cve.org
  - csaf
  - vex
  - red-hat
ssvc:
  exploitation: none
  automatable: 'no'
  technicalImpact: total
  timestamp: '2026-08-12T15:36:35.639296Z'
epss: 0.00617
epssPercentile: 0.48225
ingestedAt: '2026-09-21T10:35:15.461Z'
patched:
  - openshift_ai 3.3
  - openshift_ai 3.4
  - openshift_ai 3.5
---

## Overview

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.

## Remediation

Refer to the linked advisories for vendor-supplied fixes and affected version ranges.

## Vendor advisories

- **RHSA-2026:53263** · Red Hat · fixed in: Red Hat OpenShift AI 3.3 · released 2026-08-11 · [advisory](https://access.redhat.com/errata/RHSA-2026:53263)
- **RHSA-2026:60520** · Red Hat · fixed in: Red Hat OpenShift AI 3.4 · released 2026-08-27 · [advisory](https://access.redhat.com/errata/RHSA-2026:60520)
- **RHSA-2026:53262** · Red Hat · fixed in: Red Hat OpenShift AI 3.4 · released 2026-08-11 · [advisory](https://access.redhat.com/errata/RHSA-2026:53262)
- **RHSA-2026:60367** · Red Hat · fixed in: Red Hat OpenShift AI 3.5 · released 2026-08-26 · [advisory](https://access.redhat.com/errata/RHSA-2026:60367)