CVE-2024-5452
Lightningai Pytorch Lightning ≤ 2.3.3
Raw vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:HSummary
CVE-2024-5452 is a critical-severity Improperly Controlled Modification of Dynamically-Determined Object Attributes (CWE-915) vulnerability in Lightningai Pytorch Lightning. Its CVSS base score is 9.8 (Critical).
Operationally, exploitation aligns with the MITRE ATT&CK technique Reflective Code Loading (T1620); ranked in the top 2% of CVEs by exploit likelihood; 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 Other ATLAS/OWASP Terms risk domain.
The strongest mitigations our analysis identified map to AC-3 (Access Enforcement) and AC-6 (Least Privilege) — 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.
A remote code execution vulnerability affects the lightning-ai/pytorch-lightning library in version 2.2.1. It stems from unsafe deserialization of user-supplied input passed to deepdiff.Delta objects, which the application uses to apply frontend-driven state changes. The deepdiff library's handling of dunder attributes can be bypassed, allowing a crafted serialized delta to evade the intended whitelist and grant arbitrary attribute writes that lead to full code execution.
Unauthenticated remote attackers can exploit the flaw by sending a malicious delta payload to the delta endpoint, which is enabled by default in self-hosted deployments. Successful exploitation grants complete control over the application process, including access to other modules and instances, without requiring authentication or user interaction.
A fix was merged in commit 330af381de88cff17515418a341cbc1f9f127f9a. Public details of the issue and proof-of-concept are available in the referenced huntr.dev bounty report.
The affected component is widely used in machine-learning workflows. EPSS scores climbed from lower values to a peak of 0.6262, indicating that exploitation interest emerged after disclosure.
OWASP Top 10 for Web (2025)
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2024-2023
Vulnerability Data
A remote code execution (RCE) vulnerability exists in the lightning-ai/pytorch-lightning library version 2.2.1 due to improper handling of deserialized user input and mismanagement of dunder attributes by the `deepdiff` library. The library uses `deepdiff.Delta` objects to modify application state based…
more
on frontend actions. However, it is possible to bypass the intended restrictions on modifying dunder attributes, allowing an attacker to construct a serialized delta that passes the deserializer whitelist and contains dunder attributes. When processed, this can be exploited to access other modules, classes, and instances, leading to arbitrary attribute write and total RCE on any self-hosted pytorch-lightning application in its default configuration, as the delta endpoint is enabled by default.
- CWE(s)
AI Security AnalysisAI
- AI Category
- Deep Learning Frameworks
- Risk Domain
- Other ATLAS/OWASP Terms
- OWASP Top 10 for LLMs 2025
- None mapped
- Classification Reason
- PyTorch Lightning (lightning-ai/pytorch-lightning) is a high-level deep learning framework built on PyTorch, simplifying training and deployment of deep learning models.
Related Threats
MITRE ATT&CK Enterprise Techniques
CVEs Like This One
Affected Assets
Mitigating Controls
Control response
—
—
- 2 hardening rules · 2 OS baselines
—
Mitigating Controls (NIST 800-53 r5) AI
Enforces authorizations so that only permitted attributes may be modified on an object.
Limits the set of modifiable attributes a subject is authorized to touch.
Validates incoming attribute names and values so that only explicitly allowed fields are accepted for update.
Isolating security functions from non-security code prevents unintended manipulation of dynamically managed executable resources.
Process isolation keeps each process's dynamic code resources in separate domains, blocking cross-process tampering.
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 SDLC practices explicitly include controls that prevent improper handling of dynamic code resources.
Blocking unauthorized code execution directly limits the ability to abuse dynamically-managed resources.
Runtime-environment monitoring can detect exploitation of the weakness but does not prevent it.
Vulnerability identification can surface instances of CWE-913 but does not mitigate the root weakness.
Hardened configuration baselines can restrict dynamic code execution and variable access at runtime.
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.
Secure coding standards explicitly forbid unsafe dynamic attribute assignment and require property allow-lists.
Security testing can detect mass-assignment flaws but does not itself prevent them at runtime.
Secure development lifecycle requires input validation and object-property whitelisting that directly mitigates mass-assignment risks.
Application security requirements include explicit rules for allowable object attributes and safe deserialization.
Secure architecture principles mandate strict control over dynamic object modification and attribute binding.
Information access restriction limits who can modify objects but does not address which attributes may be changed.
Hardening callouts derived
Configuration rules from DISA STIG baselines that bear on weaknesses of the type cited by this CVE. Each rule is shown with the relationship its mapping actually records, against the CWE it was authored against. Derived via CVE→CWE over `controls_xwalks` (authoritative rows only; rows rated `none` are excluded).
Windows 10 (1 rule)
- V-220726 Data Execution Prevention (DEP) must be configured to at least OptOut. prevents CWE-913
Windows 11 (1 rule)
- V-253283 Data Execution Prevention (DEP) must be configured to at least OptOut. prevents CWE-913