What is the CVSS score of CVE-2025-33249?

CVE-2025-33249 has a CVSS 3.1 base score of 7.8 (High). CVSS vector: CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H. EPSS exploitation probability: 0.0%.

Which versions of AI / ML are affected by CVE-2025-33249?

NVIDIA NeMo Framework contains a code injection vulnerability affecting voice-preprocessing functionality across all platforms, with no patch currently available.

How to fix or mitigate CVE-2025-33249?

Within 24 hours: Inventory all systems running NVIDIA NeMo Framework and identify those processing untrusted voice input; isolate affected systems from production networks if possible. Within 7 days: Implement input validation and sanitization for all voice preprocessing inputs; restrict NeMo Framework access to trusted users and applications only. Within 30 days: Monitor NVIDIA security advisories for patch release; conduct code review of voice preprocessing implementations; consider alternative frameworks if patch timeline is unacceptable.

AI / ML CVE-2025-33249

HIGH

Command Injection (CWE-77)

2026-02-18 psirt@nvidia.com

Privilege Escalation Information Disclosure Code Injection AI / ML Nemo

7.8

CVSS 3.1

CVSS VectorNVD

CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H

Attack Vector

Local

Attack Complexity

Low

Privileges Required

Low

User Interaction

None

Scope

Unchanged

Confidentiality

High

Integrity

High

Availability

High

Lifecycle Timeline

Analysis Generated

Mar 12, 2026 - 21:55 vuln.today

CVE Published

Feb 18, 2026 - 14:16 nvd

HIGH 7.8

DescriptionNVD

NVIDIA NeMo Framework for all platforms contains a vulnerability in a voice-preprocessing script, where malicious input created by an attacker could cause a code injection. A successful exploit of this vulnerability might lead to code execution, escalation of privileges, information disclosure, and data tampering.

AnalysisAI

NVIDIA NeMo Framework for all platforms contains a vulnerability in a voice-preprocessing script, where malicious input created by an attacker could cause a code injection. [CVSS 7.8 HIGH]

Technical ContextAI

Classified as CWE-77 (Command Injection). Affects Nemo. NVIDIA NeMo Framework for all platforms contains a vulnerability in a voice-preprocessing script, where malicious input created by an attacker could cause a code injection. A successful exploit of this vulnerability might lead to code execution, escalation of privileges, information disclosure, and data tampering.

RemediationAI

Monitor vendor advisories for a patch.

CVE-2025-33249 vulnerability details – vuln.today

Back

AI / ML CVE-2025-33249

CVSS VectorNVD

Lifecycle Timeline

DescriptionNVD

AnalysisAI

Technical ContextAI

RemediationAI

Share

External POC / Exploit Code