CVE-2024-35198
Pytorch Torchserve 0.4.2 – 0.11.0
Raw vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:HSummary
CVE-2024-35198 is a critical-severity Use of Incorrectly-Resolved Name or Reference (CWE-706) vulnerability in Pytorch Torchserve. Its CVSS base score is 9.8 (Critical).
Operationally, exploitation aligns with the MITRE ATT&CK technique Path Interception (T1034); ranked in the top 47% of CVEs by exploit likelihood; it is not currently listed in the CISA KEV catalog.
This vulnerability is AI-related — categorised as Other AI Platforms; in the Supply Chain and Deployment risk domain.
The strongest mitigations our analysis identified map to AC-3 (Access Enforcement) and AC-4 (Information Flow Enforcement) — see the control section below for these in your framework.
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2024-2442
Vulnerability Data
TorchServe is a flexible and easy-to-use tool for serving and scaling PyTorch models in production. TorchServe 's check on allowed_urls configuration can be by-passed if the URL contains characters such as ".." but it does not prevent the model from…
more
being downloaded into the model store. Once a file is downloaded, it can be referenced without providing a URL the second time, which effectively bypasses the allowed_urls security check. Customers using PyTorch inference Deep Learning Containers (DLC) through Amazon SageMaker and EKS are not affected. This issue in TorchServe has been fixed by validating the URL without characters such as ".." before downloading see PR #3082. TorchServe release 0.11.0 includes the fix to address this vulnerability. Users are advised to upgrade. There are no known workarounds for this vulnerability.
- CWE(s)
AI Security AnalysisAI
- AI Category
- Other AI Platforms
- Risk Domain
- Supply Chain and Deployment
- OWASP Top 10 for LLMs 2025
- None mapped
- Classification Reason
- TorchServe is a production serving tool specifically designed for PyTorch models, which is a deep learning framework. The vulnerability affects model downloading and serving in the PyTorch ecosystem.
Related Threats
MITRE ATT&CK Enterprise Techniques
CVEs Like This One
Affected Assets
Mitigating Controls
Mitigating Controls (NIST 800-53 r5) AI
Access enforcement applies authorization checks to the resolved resource, blocking access outside the intended sphere.
Information flow enforcement can constrain flows that result from an incorrectly resolved name or reference.
Mitigating Controls (NIST CSF 2.0) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.
Secure-development practices directly prevent incorrect name/reference resolution bugs during coding.
Enforced authorization boundaries limit damage from an incorrectly resolved reference.
Logical segmentation and access controls reduce the chance an out-of-sphere resolution succeeds.
Hardened configuration baselines can constrain allowable name-to-resource mappings.
Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.
Security testing can detect incorrect name or reference resolution through fuzzing and negative test cases.
Network segmentation and routing policies reduce the chance that a mis-resolved name leads to an unintended external resource.
Segregated networks limit the blast radius when a name or reference resolves outside the intended control sphere.
Secure SDLC practices include design reviews that can catch incorrect name or reference handling before deployment.
Application security requirements can mandate validation of all external references and names used at runtime.
Secure architecture principles discourage reliance on ambient or globally-resolvable names without explicit scoping.