Skip to main content

CVE-2023-31024: CWE-121 in nvidia DGX A100

Critical
VulnerabilityCVE-2023-31024cvecve-2023-31024cwe-121
Published: Fri Jan 12 2024 (01/12/2024, 18:31:35 UTC)
Source: CVE Database V5
Vendor/Project: nvidia
Product: DGX A100

Description

NVIDIA DGX A100 BMC contains a vulnerability in the host KVM daemon, where an unauthenticated attacker may cause stack memory corruption by sending a specially crafted network packet. A successful exploit of this vulnerability may lead to arbitrary code execution, denial of service, information disclosure, and data tampering.

AI-Powered Analysis

AILast updated: 07/05/2025, 23:39:34 UTC

Technical Analysis

CVE-2023-31024 is a critical vulnerability identified in the Baseboard Management Controller (BMC) of NVIDIA DGX A100 systems, specifically affecting all BMC firmware versions prior to 00.22.05. The vulnerability is classified under CWE-121, which corresponds to a stack-based buffer overflow. The flaw resides in the host Kernel-based Virtual Machine (KVM) daemon component of the BMC, where an unauthenticated attacker can send a specially crafted network packet that triggers stack memory corruption. This corruption can lead to severe consequences including arbitrary code execution, denial of service (DoS), information disclosure, and data tampering. The vulnerability has a CVSS v3.1 base score of 9.0, indicating a critical severity level. The attack vector is network-based (AV:N), requires high attack complexity (AC:H), no privileges (PR:N), and no user interaction (UI:N). The scope is changed (S:C), meaning the vulnerability affects resources beyond the vulnerable component, and it impacts confidentiality, integrity, and availability at a high level (C:H/I:H/A:H). Although no known exploits are currently reported in the wild, the potential for exploitation is significant given the critical nature of the flaw and the high-value target represented by DGX A100 systems, which are used extensively for AI, machine learning, and high-performance computing workloads. The absence of available patches at the time of reporting increases the urgency for mitigation and monitoring.

Potential Impact

For European organizations, the impact of this vulnerability could be substantial, especially for entities relying on NVIDIA DGX A100 systems for AI research, data analytics, and computational workloads. Successful exploitation could allow attackers to execute arbitrary code on the BMC, potentially gaining control over the management interface of the hardware. This could lead to unauthorized access to sensitive data, disruption of critical AI and HPC operations, and manipulation or destruction of data. The compromise of BMCs can also serve as a foothold for lateral movement within enterprise networks, escalating the risk of broader infrastructure compromise. Given the critical role of DGX A100 in sectors such as research institutions, financial services, automotive, and telecommunications across Europe, the vulnerability poses a risk to confidentiality, integrity, and availability of sensitive and mission-critical systems. Additionally, disruption of AI workloads could impact innovation and operational capabilities in competitive industries. The high attack complexity somewhat limits exploitation to skilled attackers, but the lack of authentication requirement and no user interaction needed make remote exploitation feasible if the vulnerable BMC is network accessible.

Mitigation Recommendations

European organizations using NVIDIA DGX A100 systems should immediately assess their BMC firmware versions and prioritize upgrading to version 00.22.05 or later once available. Until patches are released, organizations should implement network-level mitigations such as isolating BMC management interfaces from untrusted networks, restricting access via firewalls and VLAN segmentation, and employing strict access control lists (ACLs) to limit exposure. Monitoring network traffic for anomalous packets targeting the KVM daemon on the BMC is recommended to detect potential exploitation attempts. Enabling logging and alerting on BMC access and unusual behavior can aid early detection. Organizations should also review and harden their supply chain and update management processes to ensure timely deployment of firmware updates. Given the criticality, coordination with NVIDIA support for early patch availability and guidance is advised. Finally, incorporating BMC security into broader endpoint and network security frameworks will help mitigate risks from this and future vulnerabilities.

Need more detailed analysis?Get Pro

Technical Details

Data Version
5.1
Assigner Short Name
nvidia
Date Reserved
2023-04-22T02:38:27.191Z
Cvss Version
3.1
State
PUBLISHED

Threat ID: 683ffd67182aa0cae2a3882b

Added to database: 6/4/2025, 8:01:43 AM

Last enriched: 7/5/2025, 11:39:34 PM

Last updated: 7/26/2025, 11:32:38 AM

Views: 10

Actions

PRO

Updates to AI analysis are available only with a Pro account. Contact root@offseq.com for access.

Please log in to the Console to use AI analysis features.

Need enhanced features?

Contact root@offseq.com for Pro access with improved analysis and higher rate limits.

Latest Threats