PT-2026-40125 · Ludwig · Ludwig

Published 2026-05-12 · Updated 2026-05-12 · CVE-2026-31238

Positive Technologies (AI Summary)

CVSS v3.1

9.8

Critical

Base vector

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

Name of the Vulnerable Software and Affected Versions Ludwig framework versions prior to 0.10.5

Description The model serving component is subject to insecure deserialization. When initiating a model server via the ludwig serve command, the framework utilizes the torch.load() function to load model weight files without the weights only=True parameter enabled. This allows the deserialization of arbitrary Python objects through the pickle module, which is a process of converting a byte stream back into an object. An attacker can provide a maliciously crafted PyTorch model file to achieve arbitrary code execution on the host system.

Recommendations Update to version 0.10.5 or later.

Fix

Deserialization of Untrusted Data

Weakness Enumeration

CWE-502

Related Identifiers

CVE-2026-31238

GHSA-XP5Q-5Q7G-Q26R

Affected Products

Ludwig

References · 4

• https://osv.dev/vulnerability/GHSA-xp5q-5q7g-q26r · Vendor Advisory
• https://nvd.nist.gov/vuln/detail/CVE-2026-31238 · Security Note
• https://github.com/ludwig-ai/ludwig · Note
• https://notion.so/CVE-2026-31238-35d1e1393188819ea77ee98ca85a2878 · Note