CVE-2025-43847
Published: 05 May 2025
Summary
CVE-2025-43847 is a high-severity Deserialization of Untrusted Data (CWE-502) vulnerability in Rvc-Project Retrieval-Based-Voice-Conversion-Webui. Its CVSS base score is 8.9 (High).
Operationally, ranked in the top 12.9% of CVEs by exploit likelihood; it is not currently listed in the CISA KEV catalog.
Deeper analysis
Retrieval-based-Voice-Conversion-WebUI is a voice conversion framework based on VITS. Versions 2.2.231006 and earlier contain an unsafe deserialization flaw in the handling of the ckpt_path2 variable. User-supplied paths are passed directly to the extract_small_model function in process_ckpt.py, which invokes torch.load without safeguards, enabling arbitrary code execution via malicious serialized objects.
An unauthenticated remote attacker can supply a crafted model path through the web interface and trigger deserialization during model processing. Successful exploitation grants remote code execution with the privileges of the application process, allowing full compromise of the host system.
No patches were available at the time of disclosure. The referenced GitHub Security Lab advisory details the affected code paths in infer-web.py and process_ckpt.py but provides no remediation guidance beyond avoiding untrusted model inputs.
The vulnerability affects an AI/ML voice-conversion workload that relies on PyTorch model loading. EPSS rose from a low baseline to a peak of 0.0602 before receding, indicating measurable post-disclosure exploitation interest.
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2025-13509
Vulnerability details
Retrieval-based-Voice-Conversion-WebUI is a voice changing framework based on VITS. Versions 2.2.231006 and prior are vulnerable to unsafe deserialization. The ckpt_path2 variable takes user input (e.g. a path to a model) and passes it to the extract_small_model function in process_ckpt.py, which…
more
uses it to load the model on that path with torch.load, which can lead to unsafe deserialization and remote code execution. As of time of publication, no known patches exist.
- CWE(s)
Related Threats
No named actor attribution yet. ATT&CK technique mapping in progress for this CVE.
Affected Assets
Mitigating Controls
Likely Mitigating Controls AI
Per-CVE control mapping for this CVE has not run yet; the list below is derived from the weakness types (CWEs) cited in the NVD entry.
Penetration testing supplies malicious serialized objects, detecting unsafe deserialization and supporting corrective actions.
Evaluation of untrusted data handling (deserialization testing) reveals unsafe processing, which the required remediation process addresses.
Untrusted serialized data can be deserialized and observed inside the chamber, blocking gadget-chain exploitation outside the sandbox.
Validates or rejects untrusted serialized data before deserialization occurs.
Identifies and blocks malicious code introduced through deserialization of untrusted data at system boundaries.
Integrity verification of serialized information can detect tampering before deserialization occurs.
Provenance of associated data allows detection of untrusted sources before deserialization or processing occurs.