Is CVE-2026-17557 exploited in the wild?

No. Junglewise Threat Intelligence has no report of CVE-2026-17557 being exploited in the wild.

How severe is CVE-2026-17557?

Medium severity, with a CVSS score of 6.5, published 21 October 2026.

Is there a fix for CVE-2026-17557?

Yes: fixed in OpenShift AI 3.5. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

How is CVE-2026-17557 attacked?

It is attacked over the network.

Am I affected by CVE-2026-17557?

Check whether you run Red Hat OpenShift AI at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

CVE-2026-17557: Red Hat OpenShift AI lm-evaluation-harness SSRF with credential disclosure

CVE-2026-17557 · Severity: medium · CVSS 6.5 · Published 2026-10-21

Red Hat OpenShift AI is a machine learning platform that runs evaluation jobs to test AI models. An authenticated user can submit a malicious evaluation job that tricks the system into revealing the authentication secret used to access the tenant's models, or makes the job pod scan internal networks that would normally be unreachable. This allows attackers to steal credentials or map and attack internal infrastructure.

Technologies: Red Hat OpenShift AI. Vendors: Red Hat.

Quick answers

Is CVE-2026-17557 exploited in the wild?: No. Junglewise Threat Intelligence has no report of CVE-2026-17557 being exploited in the wild.
How severe is CVE-2026-17557?: Medium severity, with a CVSS score of 6.5, published 21 October 2026.
Which products and versions are affected?: Red Hat OpenShift AI 3.4.
Is there a fix for CVE-2026-17557?: Yes: fixed in OpenShift AI 3.5. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.
How is CVE-2026-17557 attacked?: It is attacked over the network.
Am I affected by CVE-2026-17557?: Check whether you run Red Hat OpenShift AI at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

How to fix

Fixed in: OpenShift AI 3.5

Technical details

The lm-evaluation-harness component in OpenShift AI 3.4 is vulnerable to Server-Side Request Forgery (SSRF) via the model.url parameter in JobSpec. An authenticated user can provide a malicious URL that causes the evaluation job pod to make requests to attacker-controlled hosts or internal endpoints, disclosing the tenant's model authentication secret through bearer token forwarding or enabling blind SSRF attacks. OpenShift AI 3.5 addresses this through an architectural change.

Affected products

Red Hat OpenShift AI 3.4

Timeline

2026-10-21: disclosed

References

This page as Markdown. Free to reuse under CC BY 4.0 with a link back.