VU#369093: MLflow dspy and statsmodels flavors bypass pickle deserialization control

Summary

Overview A vulnerability in MLflow’s dspy and statsmodels model flavors allows unauthorized pickle deserialization executions despite a safety control. Specifically, the dspy flavor conditionally applies the control based on the model path’s file extension, and the statsmodels flavor does not apply the control. Description MLflow is an open-source platform for managing machine learning lifecycles, including model packaging, versioning, and deployment. "Flavors" refer to the specialized frameworks through which supported models are stored and loaded. In response to previous vulnerability concerns, MLflow implemented the MLFLOW_ALLOW_PICKLE_DESERIALIZATION safety control to block and disable executing any pickle deserialization and subsequent loads per the user’s choice. When loading models through mlflow.pyfunc.load_model(model) , users...

Why it matters

CERT/CC notes often cover coordinated disclosures, multi-vendor exposure, infrastructure risk, or cases where remediation guidance is still developing.

Operator check

Review the CERT/CC Vulnerability Note and compare the affected vendor and product list against your inventory. Apply vendor fixes or compensating controls for loading models where available.

Sources