CVE-2026-117574: Ai-gateway-payload-processing: ai-gateway-payload-processing: tenant/credential selection from client-controlled url path with no in-process authorization

Severity: critical (CVSS 9.1 v3.1)
Exploited in the wild: not listed in CISA KEV
EPSS: 0.2% probability of exploitation in the next 30 days
Public exploit or detection template: none known
Published: 2026-10-28 · updated 2026-10-28
Weakness: CWE-639

Description

A flaw was found in ai-gateway-payload-processing. An authenticated remote attacker can exploit this vulnerability by submitting a request specifying another tenant's model identifier in the request body or header. Because the service does not verify authorization before retrieving credentials from the shared store, it injects the target tenant's application programming interface (API) keys into the request. This flaw allows an attacker to perform unauthorized requests and consume another tenant's large language model (LLM) resources and quotas.

Summary

CVE-2026-117574 is a critical (CVSS 9.1) authorization flaw in ai-gateway-payload-processing, part of Red Hat OpenShift AI (RHOAI). An authenticated remote attacker can specify another tenant's model identifier in the request body or header; because the service does not verify authorization before pulling credentials from the shared store, it injects the target tenant's API keys into the request. This lets the attacker make unauthorized requests and consume another tenant's LLM resources and quotas. The flaw is not listed in CISA KEV, so no known exploitation is reported; apply the RHOAI updates that address this CVE, and until patched, restrict gateway access to trusted tenants and monitor for cross-tenant model identifiers.

CVE-2026-117574 ثغرة تفويض حرجة (CVSS 9.1) في ai-gateway-payload-processing ضمن Red Hat OpenShift AI (RHOAI). يستطيع مهاجم بعيد موثَّق تحديد معرّف نموذج تابع لمستأجر آخر في جسم الطلب أو ترويسته؛ ولأن الخدمة لا تتحقق من التفويض قبل جلب بيانات الاعتماد من المخزن المشترك، فإنها تحقن مفاتيح API الخاصة بالمستأجر المستهدف في الطلب. ويتيح ذلك للمهاجم تنفيذ طلبات غير مصرَّح بها واستهلاك موارد وحصص نماذج LLM الخاصة بمستأجر آخر. الثغرة غير مدرجة في CISA KEV ولا يُبلَّغ عن استغلال معروف؛ طبّق تحديثات RHOAI التي تعالج هذه الثغرة، وإلى حين الترقيع قيّد الوصول إلى البوابة على المستأجرين الموثوقين وراقب معرّفات النماذج العابرة للمستأجرين.

Affected products

Red Hat OpenShift AI (RHOAI): all versions

References

Get alerted when CVE-2026-117574 or the next vulnerability affects your stack