TreasureX Researchcs.AIAI Turns Car Vulnerability Reports into Structured Threat Intel
Md Erfan, Ahmed Ryan, Md Kamal Hossain Chowdhury, Md Rayhanur Rahman · July 17, 2026
Connected and autonomous vehicles (CAVs) are packed with software and sensors, making them vulnerable to cyberattacks. Currently, vulnerabilities are reported as plain text in the CVE database, which is hard for security tools to process. This paper tests open-weight large language models (LLMs) to automatically convert those text descriptions into STIX, a structured format for threat intelligence. The researchers built a dataset called CAV-STIXGen and evaluated 11 LLMs (4B to 120B parameters). Single models achieved high accuracy for identifying affected assets (F1=0.94) and weakness types (F1=0.99), but struggled with mapping attack techniques. A multi-agent setup using Gemma-4-31B and Codestral-22B improved results. The work shows that AI can automate threat intelligence for CAVs, helping security teams prioritize defenses faster. This could reduce manual analysis costs and speed up patch deployment, making autonomous driving safer.
Read more: https://arxiv.org/abs/2607.16175
Educational content only, generated with AI assistance from public research abstracts. Not investment advice. Thank you to arXiv for use of its open access interoperability.