The Ludwig framework through version 0.10.4 is affected by an insecure deserialization vulnerability (CWE-502) in its predict() method. When a user provides a dataset file path, the framework detects the file format and if it is a pickle file, it loads it using pandas.read_pickle() without any validation or security restrictions. Because pickle deserialization can execute arbitrary Python code, an attacker can craft a malicious pickle file that, when loaded, leads to arbitrary code execution on the system running Ludwig predictions. This vulnerability is remotely exploitable by supplying a malicious pickle file to the predict() method. The CVSS v3.1 base score is 9.8 (critical), reflecting network attack vector, no privileges or user interaction required, and full impact on confidentiality, integrity, and availability. There is no vendor advisory or patch information available at this time, and the vulnerability is not related to a cloud service.

Successful exploitation allows remote attackers to execute arbitrary code on the system running the Ludwig framework's predict() method by supplying a malicious pickle file. This compromises confidentiality, integrity, and availability of the affected system. The vulnerability is critical with a CVSS score of 9.8, indicating a severe security risk.

Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Until an official fix is available, users should avoid passing untrusted or unauthenticated pickle files to the predict() method. Consider disabling or restricting the use of pickle files in predictions or implementing input validation and sandboxing as temporary mitigations.