← Back to Library

LLM-Empowered Functional Safety and Security by Design in Automotive Systems

Authors: Nenad Petrovic, Vahid Zolfaghari, Fengjunjie Pan, Alois Knoll

Published: 2026-01-05

arXiv ID: 2601.02215v1

Added to Library: 2026-01-07 10:13 UTC

Safety

📄 Abstract

This paper presents LLM-empowered workflow to support Software Defined Vehicle (SDV) software development, covering the aspects of security-aware system topology design, as well as event-driven decision-making code analysis. For code analysis we adopt event chains model which provides formal foundations to systematic validation of functional safety, taking into account the semantic validity of messages exchanged between key components, including both CAN and Vehicle Signal Specification (VSS). Analysis of security aspects for topology relies on synergy with Model-Driven Engineering (MDE) approach and Object Constraint Language (OCL) rules. Both locally deployable and proprietary solution are taken into account for evaluation within Advanced Driver-Assistance Systems (ADAS)-related scenarios.

🔍 Key Points

  • Proposes an LLM-powered workflow for Software Defined Vehicle (SDV) development that integrates security-aware system topology design and event-driven decision-making code analysis.
  • Utilizes an event chains model as a formal foundation to validate functional safety, addressing the semantic validity of messages within automotive systems like CAN and VSS.
  • Incorporates Model-Driven Engineering (MDE) and Object Constraint Language (OCL) for systematic evaluation of system topology security aspects.
  • Demonstrates practical application through a case study on Advanced Driver-Assistance Systems (ADAS), showcasing the efficacy of the proposed framework for functional safety analysis in various emergency scenarios.
  • Compares the performance of different LLMs in functional safety tasks, highlighting the potential of locally deployable models in sensitive automotive applications that require stringent security measures.

💡 Why This Paper Matters

This paper makes significant contributions to the intersection of AI and automotive systems by enhancing safety and security measures through the innovative use of LLMs. The proposed framework not only streamlines the SDV development process but also ensures compliance with safety standards, which is critical for the growing complexity of automotive software. As vehicles become increasingly software-driven, this work will serve as a crucial reference for integrating AI technologies in safe automotive system designs.

🎯 Why It's Interesting for AI Security Researchers

This paper is particularly relevant to AI security researchers as it addresses the challenges of integrating AI-driven solutions into safety-critical systems like automotive software. By focusing on functional safety and security from the early design phases using LLMs, it highlights a novel approach to mitigate risks associated with the deployment of AI technologies in autonomous vehicles. The findings emphasize the need for robust security measures in AI methods, aligning with ongoing research in AI safety and security domains.

📚 Read the Full Paper

📄 View on arXiv 📎 Download PDF