Track vulnerabilities like this one.
Sign up free to get alerted when software you use is affected.
8.8
CVE-2024-37053: Malicious MLflow models can run code on user systems in versions 1.1.0 and later
CVE-2024-37053 · published 2 years ago
Summary
Versions 1.1.0 and newer of the MLflow platform have a security weakness that allows a maliciously uploaded machine learning model to run arbitrary code on a user's system when interacted with. This could allow an attacker to take control of a user's device or steal sensitive information. Users should update to the latest version of MLflow to fix this issue.
What to do
The CVE record does not list a fixed version. Check the vendor's site or the advisory links below - a fix may already be released.
Affected software
| Ecosystem | Vendor | Product | Affected versions |
|---|---|---|---|
| pip | databricks | mlflow | >= 1.1.0, <= 2.14.1 |
| – | lfprojects | mlflow |
>= 1.1.0 cpe:2.3:a:lfprojects:mlflow:*:*:*:*:*:*:*:* |
Original advisory text
Deserialization of untrusted data can occur in versions of the MLflow platform running version 1.1.0 or newer, enabling a maliciously uploaded scikit-learn model to run arbitrary code on an end use...
Deserialization of untrusted data can occur in versions of the MLflow platform running version 1.1.0 or newer, enabling a maliciously uploaded scikit-learn model to run arbitrary code on an end user’s system when interacted with.
Internet-facing
14 days
Internal
At next upgrade
- Not known to be exploited
- Needs hands-on effort to exploit
- Gives an attacker full control
Type
CWE-502Deserialization of Untrusted Data
Timeline
Published4 Jun 2024
Updated4 Nov 2026
First seen6 Mar 2026
Track software like this
Free during beta