CVE-2026-41486
Published: 08 May 2026
Summary
CVE-2026-41486 is a high-severity Code Injection (CWE-94) vulnerability in Anyscale Ray. Its CVSS base score is 8.9 (High).
Operationally, exploitation aligns with the MITRE ATT&CK technique Python (T1059.006); ranked at the 37.5th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog.
This vulnerability is AI-related — categorised as Data Processing Libraries; in the Data-Related Vulnerabilities risk domain.
OWASP Top 10 for Web (2025)
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2026-28828
Vulnerability details
Ray is an AI compute engine. From version 2.54.0 to before version 2.55.0, Ray Data registers custom Arrow extension types (ray.data.arrow_tensor, ray.data.arrow_tensor_v2, ray.data.arrow_variable_shaped_tensor) globally in PyArrow. When PyArrow reads a Parquet file containing one of these extension types, it calls…
more
__arrow_ext_deserialize__ on the field's metadata bytes. Ray's implementation passes these bytes directly to cloudpickle.loads(), achieving arbitrary code execution during schema parsing, before any row data is read. This issue has been patched in version 2.55.0.
- CWE(s)
AI Security AnalysisAI
- AI Category
- Data Processing Libraries
- Risk Domain
- Data-Related Vulnerabilities
- OWASP Top 10 for LLMs 2025
- None mapped
- Classification Reason
- Matched keywords: ai
Related Threats
MITRE ATT&CK Enterprise TechniquesAI
Why these techniques?
Unsafe cloudpickle deserialization of Parquet metadata bytes directly enables Python code execution (T1059.006) via malicious file (T1204.002) or client-side exploitation (T1203).
CVEs Like This One
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.
Untrusted serialized data can be deserialized and observed inside the chamber, blocking gadget-chain exploitation outside the sandbox.
Validates inputs used in dynamic code generation to block injected directives.
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.
Makes persistent code injection into loaded programs impossible when the executable image itself resides on hardware-protected read-only media.
Directly prevents execution of attacker-supplied code written into data memory regions.
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.