CVE-2025-12805: A flaw was found in Red Hat OpenShift AI (RHOAI) llama-stack-operator

This month

HIGH 8.1WATCHLISTDOWNGRADE

CVE-2025-12805 is a high-severity flaw (CVSS 8.1) affecting openshift ai. Below: why it ranks where it does, which versions are affected, what to do, and the exact commands.

A flaw was found in Red Hat OpenShift AI (RHOAI) llama-stack-operator. This vulnerability allows unauthorized access to Llama Stack services deployed in other namespaces via direct network requests, because no NetworkPolicy restricts access to the llama-stack service endpoint. As a result, a user in one namespace can access another user’s Llama Stack instance and potentially view or manipulate sensitive data.

Why this priority

High severity
Public exploit reference
On your watchlist

No fix is published yet: follow the downgrade or mitigation steps.

Risk facts

CVSS score	8.1 (High)
Attack vector	Network (remote)
Attack complexity	Low
Privileges needed	Low
User interaction	None
EPSS exploit chance	0.4% (top 67%)
CISA exploited list	No
Weakness	CWE-653
Published	2026-03-26
Last modified	2026-06-17

Publisher (CNA): Red Hat, Inc. · Vendor: redhat

Last updated 2026-06-17 · Sources: CVE Program, NVD, CISA KEV, EPSS and vendor advisories · How we score

Am I affected?

Timeline

2026-03-26 22:16 UTCPublished CVE record published
2026-06-17 08:32 UTCUpdated Record last modified

redhat openshift ai DOWNGRADE

Affected: 2.25

Fixed in: no fixed release known

Safe to downgrade to: a release before 2.25

1. Check installed version

apt-cache policy openshift-ai || rpm -q openshift-ai || openshift-ai --version

2. No fixed release: DOWNGRADE to a release before 2.25 (Debian / Ubuntu)

apt-cache madison openshift-ai sudo apt install openshift-ai=<VERSION_BEFORE_2.25> sudo apt-mark hold openshift-ai

2. No fixed release: DOWNGRADE (RHEL / Amazon Linux / Fedora)

dnf --showduplicates list openshift-ai sudo dnf downgrade openshift-ai-<VERSION_BEFORE_2.25> sudo dnf versionlock add openshift-ai

Downgrade risk

Downgrading can break dependent software and may reopen older bugs. Read the distro advisory first and test on a non-production host.

After downgrade

Restart the service, and keep the package held until a fixed release ships.

Verify at the official source NVD · CVE Program · Report a problem with this page
Scores and fixes here are computed automatically from those and other public sources. Confirm with the vendor advisory before changing production systems.

Quick answers

How do I fix CVE-2025-12805?

openshift ai: downgrade or apply the listed mitigation. Exact commands for each OS are on this page.

Is CVE-2025-12805 being exploited?

CVE-2025-12805 is not on the CISA Known Exploited Vulnerabilities list at the time of the last update. Check again, as this can change.

What does CVE-2025-12805 affect?

CVE-2025-12805 affects openshift ai. Severity is high (CVSS 8.1). A flaw was found in Red Hat OpenShift AI (RHOAI) llama-stack-operator. This vulnerability allows unauthorized access to Llama Stack services deployed in other namespaces via direct network requests, because no NetworkPolicy restricts access

How urgent is CVE-2025-12805?

Priority 47/100. High severity; Public exploit reference; On your watchlist

Software pages

OpenAI and LLM tools vulnerabilities · Red Hat OpenShift vulnerabilities

Related CVEs

CVE-2026-5483 HIGH 8.5 UPGRADE A flaw was found in odh-dashboard in Red Hat Openshift AI
CVE-2026-42271 HIGH 8.8 EXPLOITED UPGRADE UPGRADE BerriAI LiteLLM Command Injection Vulnerability
CVE-2026-15154 MEDIUM 6.5 MITIGATE A flaw was found in `guardrails-detectors`, a component of Red Hat OpenShift AI
CVE-2026-48710 MEDIUM 6.5 EXPLOITED UPGRADE UPGRADE Kludex Starlette HTTP Request/Response Smuggling Vulnerability
CVE-2023-54365 HIGH 7.5 UPGRADE UPGRADE Traefik before 2.10.5 and 3.0.0-beta4 is affected by a denial-of-service vulnerability in HTTP/2 request handl

References

← Back to latest CVEs