CVE-2026-5241
Published: 03 June 2026
Summary
CVE-2026-5241 is a critical-severity Inclusion of Functionality from Untrusted Control Sphere (CWE-829) vulnerability in Huggingface Transformers. Its CVSS base score is 9.6 (Critical).
Operationally, exploitation aligns with the MITRE ATT&CK technique Compromise Software Supply Chain (T1195.002); ranked at the 38.5th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog; a public proof-of-concept is referenced.
This vulnerability is AI-related — categorised as NLP and Transformers; in the Supply Chain and Deployment risk domain.
OWASP Top 10 for Web (2025)
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2026-34084
Vulnerability details
A vulnerability in the LightGlue model loading path of huggingface/transformers version 5.2.0 allows an attacker-controlled model repository to execute arbitrary code during model initialization. The issue arises because the `trust_remote_code` parameter, intended to prevent remote code execution, is overridden by…
more
untrusted serialized configuration data in a nested code path. Specifically, when loading a LightGlue model using `AutoModel.from_pretrained()` with `trust_remote_code=False`, the `LightGlueConfig` reads the `trust_remote_code` value from the untrusted `config.json` file and propagates it into nested `AutoConfig.from_pretrained()` calls. This results in the execution of attacker-provided Python modules, even when the victim explicitly disables remote code execution. The vulnerability poses a high risk for environments such as API inference servers, research notebooks, CI/CD pipelines, and model evaluation workers, potentially leading to credential theft, lateral movement, or persistence/backdoor deployment.
- 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: huggingface, transformers
Related Threats
MITRE ATT&CK Enterprise TechniquesAI
Why these techniques?
Vulnerability enables arbitrary code execution via attacker-controlled model repositories during loading, directly mapping to compromise of the software supply chain.
CVEs Like This One
Affected Assets
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.
Limiting P2P file sharing technology reduces inclusion of functionality or resources from untrusted external control spheres.
Enforcing installation policies prevents users from including functionality obtained from untrusted control spheres.
The inventory process requires identifying and recording the origin of all components, making inclusion of functionality from untrusted control spheres easier to detect during reviews.
Requiring approval and monitoring of maintenance tools prevents inclusion and execution of functionality obtained from untrusted sources.
Unowned portable devices represent untrusted control spheres; the prohibition prevents inclusion of functionality or data from such sources.
Strategy mandates assessment of third-party components and suppliers, directly reducing inclusion of functionality from untrusted control spheres.
Procedures can mandate supply-chain vetting and restrictions on functionality obtained from untrusted third-party or external control spheres.
Requires use of trusted sources and provenance tracking, tangibly limiting inclusion of functionality from untrusted control spheres.