HighVulnerability
CVE-2026-31232: CosyVoice insecure deserialization in model loading via torch.load
- Source
- NVD(opens in a new tab)
- Identifier
- CVE-2026-31232
- Published
- Record updated
- Known exploitation
- Not listed in the CISA Known Exploited Vulnerabilities catalog at the last check.
- EPSS
- 0.5%
Summary
CVE-2026-31232 affects the CosyVoice project through commit 6e01309e01bc93bbeb83bdd996b1182a81aaf11e (2025-30-21) and is classified as CWE-502. The model loading process calls torch.load() without weights_only=True when reading .pt files from a directory passed via --model_dir, so arbitrary Python objects can be deserialized through Pickle. An attacker who supplies a crafted model directory can trigger remote code execution on a victim who loads it through the CosyVoice web interface.
Mitigation
The source does not state a fix yet. Check the original advisory for updates.
Topics
Related items
- MediumAI broke the cybersecurity world. What’s next?Similar attack · The Verge (AI)
- InfoCVE-2026-123357: GROBID command execution through crafted PDF filename in Linux conversionSimilar attack · NVD/CVE Database
- MediumShai-Hulud Descendant Surfaces in Malicious Ruby Gem That Steals Credentials and Poisons AI ToolsSimilar attack · Socket Blog
- MediumCPython flaw exposes AI inference servers to sandbox escape attacksSimilar attack · CSO Online
- HighCVE-2026-88981: Kubeflow Pipelines enables users to build and deploy portable, scalable machine learning workflows. Prior to 2.18.0…Similar attack · NVD/CVE Database