Cyber Resilience

CVE-2025-6920

Medium

Published: 01 July 2025

Published
01 July 2025
Modified
18 August 2025
KEV Added
Patch
CVSS Score v3.1 5.3 CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N
EPSS Score 0.0018 39.7th percentile
Risk Priority 11 60% EPSS · 20% KEV · 20% CVSS

Summary

CVE-2025-6920 is a medium-severity Missing Authentication for Critical Function (CWE-306) vulnerability in Redhat Ai Inference Server. Its CVSS base score is 5.3 (Medium).

Operationally, exploitation aligns with the MITRE ATT&CK technique Exploit Public-Facing Application (T1190); ranked at the 39.7th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog.

This vulnerability is AI-related — categorised as NLP and Transformers; in the Supply Chain and Deployment risk domain.

EU & UK References

Vulnerability details

A flaw was found in the authentication enforcement mechanism of a model inference API in ai-inference-server. All /v1/* endpoints are expected to enforce API key validation. However, the POST /invocations endpoint failed to do so, resulting in an authentication bypass.…

more

This vulnerability allows unauthorized users to access the same inference features available on protected endpoints, potentially exposing sensitive functionality or allowing unintended access to backend resources.

CWE(s)

AI Security AnalysisAI

AI Category
NLP and Transformers
Risk Domain
Supply Chain and Deployment
OWASP Top 10 for LLMs 2025
None mapped
Classification Reason
Matched keywords: ai

Related Threats

MITRE ATT&CK Enterprise TechniquesAI

T1190 Exploit Public-Facing Application Initial Access
Adversaries may attempt to exploit a weakness in an Internet-facing host or system to initially access a network.
Why these techniques?

The authentication bypass in the public-facing model inference API (/invocations endpoint) allows unauthorized access to protected features, enabling exploitation of a public-facing application.

Affected Assets

redhat
ai inference server
all versions

Mitigating Controls

Likely Mitigating Controls AI

Per-CVE control mapping for this CVE has not run yet; the list below is derived from the weakness types (CWEs) cited in the NVD entry.

addresses: CWE-306

Requires established identification and authentication to unlock, mitigating missing authentication for continued system access.

addresses: CWE-306

Requiring identification and rationale for actions allowed without authentication ensures critical functions are not left unprotected by forcing review of authentication requirements.

addresses: CWE-306

Authorizing mobile device connections to organizational systems ensures authentication is performed for this critical access function.

addresses: CWE-306

Guarantees critical functions are protected by mandatory invocation of the access control mechanism.

addresses: CWE-306

Auditing sessions makes it possible to detect access to critical functions without required authentication.

addresses: CWE-306

The assessment process confirms authentication is present and effective for critical functions, preventing exploitation from missing authentication.

addresses: CWE-306

Certification assesses that critical functions have required authentication controls in place.

addresses: CWE-306

Disabling non-essential functions and services eliminates the need to secure them, reducing exposure from missing authentication on unnecessary components.

References