Cyber & AI intelligence
Wasteland.
Briefs indexed3316
Issues34
Published Mondays07:30 CT
⚡ Active KEV CVE-2026-117574 2026-10-28

CVE-2026-117574: Cross-Tenant API Key Injection in Red Hat OpenShift AI Gateway

"A critical authorization flaw (CVSS 9.1) in the ai-gateway-payload-processing component of Red Hat OpenShift AI lets an authenticated attacker use another tenant's LLM API keys and consume that tenant's resources and…"

A critical authorization flaw (CVSS 9.1) in the ai-gateway-payload-processing component of Red Hat OpenShift AI lets an authenticated attacker use another tenant's LLM API keys and consume that tenant's resources and quotas.

What Is It

CVE-2026-117574 is an authorization bypass in ai-gateway-payload-processing, classified as CWE-639 (Authorization Bypass Through User-Controlled Key). An authenticated remote attacker can put another tenant's model identifier in the request body or header. The service doesn't check authorization before it pulls credentials from the shared store, so it injects the target tenant's API keys into the attacker's request.

Red Hat ([email protected]) published the record on 2026-10-28. NVD lists it as "Awaiting Analysis."

Why It Matters

Red Hat rates the flaw 9.1 (CRITICAL) with vector CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:L/A:L:

On multi-tenant AI platforms, this breaks the separation between tenants and can push LLM costs and quota use onto the victim.

Exploitation status: The supplied CISA KEV entry is empty. KEV does not confirm active exploitation of this CVE at this time.

What's Vulnerable

The supplied record doesn't list specific affected or fixed version ranges.

Patch Status

The supplied NVD data doesn't name a fixed version or patch release. With no KEV listing, there is no CISA-mandated required action or due date. Organizations running Red Hat OpenShift AI should:

Sources