CVE-2026-24141
Published: 24 March 2026
Summary
CVE-2026-24141 is a high-severity Deserialization of Untrusted Data (CWE-502) vulnerability in Custhelp (inferred from references). Its CVSS base score is 7.8 (High).
Operationally, exploitation aligns with the MITRE ATT&CK technique Exploitation for Client Execution (T1203); ranked at the 24.3th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog.
This vulnerability is AI-related — categorised as Machine Learning Libraries; in the Supply Chain and Deployment risk domain.
The strongest mitigations our analysis identified are NIST 800-53 SI-10 (Information Input Validation) and SI-2 (Flaw Remediation).
Deeper analysis
CVE-2026-24141 is a vulnerability in the ONNX quantization feature of NVIDIA Model Optimizer for Windows and Linux. The issue stems from unsafe deserialization triggered by processing a specially crafted input file, classified under CWE-502 (Deserialization of Untrusted Data). It carries a CVSS v3.1 base score of 7.8 (AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H), indicating high impact on confidentiality, integrity, and availability.
A local attacker requires no privileges but needs user interaction, such as convincing a user to process a malicious input file through the affected feature. Successful exploitation could result in arbitrary code execution, privilege escalation, data tampering, and information disclosure on the targeted system.
Mitigation guidance is available in the NVIDIA security advisory at https://nvidia.custhelp.com/app/answers/detail/a_id/5798, along with details from the NVD entry at https://nvd.nist.gov/vuln/detail/CVE-2026-24141 and CVE record at https://www.cve.org/CVERecord?id=CVE-2026-24141.
This vulnerability is relevant to AI/ML workflows, as it affects model optimization tools handling ONNX formats. No real-world exploitation has been reported in the provided details.
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2026-15003
Vulnerability details
NVIDIA Model Optimizer for Windows and Linux contains a vulnerability in the ONNX quantization feature, where a user could cause unsafe deserialization by providing a specially crafted input file. A successful exploit of this vulnerability might lead to code execution,…
more
escalation of privileges, data tampering, and information disclosure.
- CWE(s)
AI Security AnalysisAI
- AI Category
- Machine Learning Libraries
- Risk Domain
- Supply Chain and Deployment
- OWASP Top 10 for LLMs 2025
- None mapped
- Classification Reason
- Matched keywords: onnx
Related Threats
MITRE ATT&CK Enterprise TechniquesAI
Why these techniques?
Unsafe deserialization (CWE-502) of a crafted ONNX file directly enables client-side RCE when the victim opens/processes the file, mapping to Exploitation for Client Execution and User Execution: Malicious File.
CVEs Like This One
Affected Assets
Mitigating Controls
Mitigating Controls (NIST 800-53 r5) AI
Directly mitigates the unsafe deserialization vulnerability by requiring timely identification, reporting, and patching of the affected NVIDIA Model Optimizer.
Prevents exploitation of the ONNX quantization feature by validating the syntax, structure, and content of input files before deserialization processing.
Mitigates arbitrary code execution resulting from successful deserialization by implementing memory protection mechanisms like ASLR and DEP.