EUVD-2026-17518Severity by source
AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
Primary rating from NVD · only source for this CVE.
CVSS VectorNVD
Lifecycle Timeline
5DescriptionNVD
NVIDIA BioNeMo contains a vulnerability where a user could cause a deserialization of untrusted data. A successful exploit of this vulnerability might lead to code execution, denial of service, information disclosure, and data tampering.
AnalysisAI
Deserialization of untrusted data in NVIDIA BioNeMo Framework enables local attackers to execute arbitrary code, cause denial of service, disclose sensitive information, or tamper with data when users open malicious files. CVSS 7.8 (High) reflects local attack vector requiring user interaction. EPSS data not available; no public exploit identified at time of analysis. Affects NVIDIA BioNeMo Framework, a platform for AI-driven drug discovery and biomolecular research.
Technical ContextAI
This vulnerability stems from CWE-502 (Deserialization of Untrusted Data), a class of flaws where applications deserialize externally-controlled data without proper validation. The NVIDIA BioNeMo Framework (CPE: cpe:2.3:a:nvidia:bionemo_framework) likely accepts serialized objects from user-provided files or inputs during machine learning model loading, configuration parsing, or data processing workflows. When Python pickle, YAML, or similar serialization formats are used without sanitization, attackers can embed malicious payloads that execute during deserialization. In scientific computing frameworks like BioNeMo, which process complex biomolecular datasets and pretrained models, deserialization vulnerabilities are particularly concerning as researchers routinely share model files and configuration data across institutions.
RemediationAI
Organizations should immediately consult NVIDIA's official security bulletin at https://nvidia.custhelp.com/app/answers/detail/a_id/5808 for patched versions and upgrade guidance. Patch available per vendor advisory, though specific remediated version numbers are not confirmed in available data at time of analysis. As an interim mitigation, restrict BioNeMo framework access to trusted users only, implement strict input validation for all model files and configuration data before deserialization, and employ sandboxing or containerization to limit blast radius of potential exploitation. Consider disabling or restricting deserialization features for external data sources until patching is complete. Organizations should also audit existing model repositories for potentially malicious serialized objects and establish policies requiring cryptographic signing of trusted model files.
More in Bionemo Framework
View allInsecure deserialization in NVIDIA BioNeMo Framework enables remote code execution when attackers can induce users to pr
Path traversal in NVIDIA BioNeMo Core for Linux allows remote attackers to escape intended directory boundaries when a u
Arbitrary code execution in NVIDIA BioNemo Framework on Linux allows a local attacker to abuse unsafe deserialization of
Same weakness CWE-502 – Deserialization of Untrusted Data
View allShare
External POC / Exploit Code
Leaving vuln.today