CyberRota Analysis
AI-GeneratedMLflow's statsmodel flavor, specifically versions 2.1.0 to 3.14.0, is vulnerable due to the absence of the MLFLOW_ALLOW_PICKLE_DESERIALIZATION=False security control in the _load_model() function. This flaw enables remote attackers to execute arbitrary code by exploiting a crafted MLmodel artifact, posing a significant risk to systems utilizing these versions. Organizations using affected versions should prioritize remediation to mitigate potential exploitation risks.
Original NVD Description
MLflow's statsmodel flavor, versions 2.1.0 to 3.14.0, omits the MLFLOW_ALLOW_PICKLE_DESERIALIZATION=False security control entirely in _load_model(), which allows a remote attacker to execute arbitrary code via a crafted MLmodel artifact.