Raw vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:HSummary
CVE-2026-24747 is a high-severity Code Injection (CWE-94) vulnerability in Linuxfoundation Pytorch. Its CVSS base score is 8.8 (High).
Operationally, exploitation aligns with the MITRE ATT&CK technique Command and Scripting Interpreter (T1059); ranked at the 50th 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 Deep Learning Frameworks; in the Supply Chain and Deployment risk domain.
The strongest mitigations our analysis identified map to SA-11 (Developer Testing and Evaluation) and SI-10 (Information Input Validation) — see the control section below for these in your framework.
Deeper analysis AI-assisted summary
Synthesised by an AI model from the NVD description and linked references — a reading aid, not an authoritative source.
CVE-2026-24747 affects PyTorch, a Python package for tensor computation, in versions prior to 2.10.0. The vulnerability lies in the `weights_only` unpickler, enabling an attacker to craft a malicious checkpoint file (.pth) that corrupts memory when loaded via `torch.load(..., weights_only=True)`. This flaw maps to CWE-94 (Code Injection) and CWE-502 (Deserialization of Untrusted Data), earning a CVSS v3.1 base score of 8.8 (AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H).
Exploitation requires an attacker to deliver a specially crafted .pth file to a victim, who must then load it using the vulnerable `torch.load` call with `weights_only=True`, involving user interaction. No privileges are needed (PR:N), and attacks can originate over the network (AV:N) with low complexity (AC:L). Successful exploitation corrupts memory and can lead to arbitrary code execution, with high impacts on confidentiality, integrity, and availability (C:H/I:H/A:H).
PyTorch version 2.10.0 resolves the issue. Mitigation involves updating to this version or later. Official resources include the security advisory (GHSA-63cw-57p8-fm3p), the fixing commit (954dc5183ee9205cbe79876ad05dd2d9ae752139), the issue discussion (issues/163105), and release notes for v2.10.0.
PyTorch's role in machine learning makes this vulnerability pertinent to AI/ML workflows, where checkpoint files are commonly shared and loaded with `weights_only=True` for safer deserialization.
OWASP Top 10 for Web (2025)
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2026-4738
Vulnerability Data
PyTorch is a Python package that provides tensor computation. Prior to version 2.10.0, a vulnerability in PyTorch's `weights_only` unpickler allows an attacker to craft a malicious checkpoint file (`.pth`) that, when loaded with `torch.load(..., weights_only=True)`, can corrupt memory and potentially…
more
lead to arbitrary code execution. Version 2.10.0 fixes the issue.
- CWE(s)
AI Security AnalysisAI
- AI Category
- Deep Learning Frameworks
- Risk Domain
- Supply Chain and Deployment
- OWASP Top 10 for LLMs 2025
- None mapped
- Classification Reason
- Matched keywords: pytorch
Related Threats
MITRE ATT&CK Enterprise Techniques
CVEs Like This One
Affected Assets
Mitigating Controls
Control response
—
—
—
V1.3.1
Mitigating Controls (NIST 800-53 r5) AI
Developer testing and evaluation finds code paths that accept and execute externally influenced strings.
Input validation directly stops untrusted data from being used to construct executable code without neutralization.
Least privilege limits the damage an injected code fragment can perform once executed.
Requiring documented secure development standards and tools enforces use of safe code-generation APIs and escaping.
Engineering principles such as safe deserialization and input sanitization structurally prevent the weakness from being introduced.
Integrity verification tools can detect malformed or tampered serialized data after the fact.
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.
PR.PS-06's SDLC practices directly target injection flaws via secure coding and testing (mostly), yet as a single broad outcome it leaves many code-generation specifics unaddressed (partial).
PR.DS-10 protects runtime data confidentiality/integrity but has no bearing on neutralizing externally influenced input during code generation, so neither direction shows any preventive effect.
PR.PS-02 addresses only post-deployment updates/patching and cannot prevent introduction of unsafe deserialization code, yet it can remediate some instances when the flaw exists in outdated libraries or components.
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 includes validation of deserialization routines and the use of untrusted data, reducing the likelihood that unsafe object reconstruction will be deployed.
Banning unapproved code samples and unauthenticated web services, combined with secure-coding standards and SAST, prevents the dynamic generation or inclusion of attacker-supplied code.
Regular scanning of third-party libraries and timely patching reduce the likelihood that unsafe deserialization vulnerabilities remain active.
Controls that restrict unauthorized or malicious code from being introduced via external networks or removable media limit opportunities for an attacker to inject and execute arbitrary code.