InfoSecBulletin Cybersecurity for mankind
The NVIDIA NeMo Framework has three vulnerabilities that could enable attackers to execute remote code, risking AI system compromise and data tampering.
The security flaws CVE-2025-23249, CVE-2025-23250, and CVE-2025-23251 have a CVSS base score of 7.6, indicating a significant risk to users of the popular generative AI framework.
Citrix NetScaler SAML 0-Day Flaw Under Attack
Major Danish university breached, 200,000 users at risk
Microsoft’s X account hijacked to promote Clippy crypto scam
Critical cPanel, GitLab AI Gateway and Dell CSM Flaws Enable RCE And Admin Hijacking
Nearly 100,000 email addresses exposed in first AI-related data breach in Singapore
Hackers Exploit Zimbra Mail Servers: TeamViewer patched 5 critical flaws
Google Warns of Hackers Actively Exploiting Citrix 0-Day Flaws
CISA Warns Critical MikroTik RouterOS Flaw While Cisco SD-WAN Zero-Day Exploited in Attacks
Apple Zero-Day Exploited: Pentagon Data Breach Reportedly Exposes Sensitive Data of 3 Million People
JadePuffer Agentic AI targets and destroys Azure’s cloud resources
On April 22, 2025, NVIDIA released security patches and advised users to update immediately to prevent potential exploitation on Windows, Linux, and macOS.
High-Severity Vulnerabilities in NVIDIA NeMo Framework:
CVE-2025-23249 is a vulnerability that permits unsafe deserialization of untrusted data, enabling attackers to execute arbitrary code remotely.
CWE-502 allows attackers to manipulate serialized objects and inject malicious code during data processing.
“NVIDIA NeMo Framework contains a vulnerability where a user could cause a deserialization of untrusted data by remote code execution. A successful exploit of this vulnerability might lead to code execution and data tampering,” states the official security bulletin.
The second vulnerability (CVE-2025-23250) involves improper path validation (CWE-22), allowing attackers to exploit path traversal techniques to write arbitrary files.
Security researchers warn that this vulnerability may let attackers overwrite sensitive files or alter configurations, which could disrupt training pipelines or contaminate datasets in AI processes.
The third vulnerability (CVE-2025-23251) involves improper code generation control (CWE-94) that could allow for remote code execution.
This is particularly worrying for a framework meant for generative AI, as it affects the line between trusted and untrusted code execution.
NVIDIA acknowledged security researcher Peng Zhou from Shanghai University for reporting three vulnerabilities in April.
All three vulnerabilities share the same attack vector specifications (AV:N/AC:L/PR:N/UI:R/S:U/C:L/I:H/A:L), indicating they can be exploited remotely with low attack complexity and no privileges required, though user interaction is needed.
The NeMo Framework is a scalable, cloud-native AI platform used by researchers and developers for large language models, multimodal models, and AI applications like speech recognition and computer vision.
The company released version 25.02 to fix these issues and urges all affected systems to update immediately.
