Towards a Hybrid Generative AI-Based Framework for Automotive Cybersecurity: A Feasibility Study
Resumo
Ensuring cybersecurity in Software-Defined Vehicles (SDVs) requires continuous operational observability; however, Vehicle Security Operations Centers (VSOCs) face two critical challenges: the scarcity and imbalance of real-world attack data for training Intrusion Detection Systems, and the cognitive overload on analysts interpreting raw vehicular logs. This paper proposes a Hybrid Generative AI-Based (GenAI) framework to address these issues by integrating a Conditional Tabular Generative Adversarial Network (GAN) for data augmentation and a lightweight Large Language Model (LLM) for alert interpretation. A first feasibility study using a GAN and LLM demonstrates the potential of GenAI but also reveals critical limitations in off-the-shelf models, including semantic integrity violations in generated packets and hallucination risks in alert interpretation, underscoring the need for robust architectural orchestration and tailored software engineering practices in critical automotive environments.
Palavras-chave:
Generative AI, Software-Defined Vehicle, GAN, IDS, Security, VSOC
Referências
Rafael Abreu, Emanuel Simão, Carlos Serôdio, Frederico Branco, and Antonio Valente. 2024. Enhancing IoT security in vehicles: A comprehensive review of AI-driven solutions for cyber-threat detection. AI (2024). DOI: 10.3390/ai5040112
R. Aishwarya, V. Vetriselvi, Srinivas Naveen, and Muthuraman A. Ashwin. 2025. An integrated IDS for the Internet of Vehicles using a Large Language Model framework. Internet of Things 33 (2025). DOI: 10.1016/j.iot.2025.101666
MITRE Corporation. 2025. CVE: Common Vulnerabilities and Exposures. [link]. Accessed: Aug. 09, 2026.
João Carmo de Almeida Neto, Leandro Santiago de Araújo, Leopoldo André Dutra Lusquino Filho, and Claudio Miceli de Farias. 2025. Time series forecasting for multidimensional telemetry data based on Generative Adversarial Network in a Digital Twin. Journal of C. S. (2025). DOI: 10.1016/j.jocs.2025.102589
Chloe Soriano de Leon and Cedric Angelo Festin. 2026. Privacy-Preserving Vehicle Intrusion Detection System Using Federated Learning and Homomorphic Encryption. In Proc. of theWorkshop on Computation: Theory and Practice. Atlantis Press. DOI: 10.2991/978-94-6239-638-8_31
Erica Espinosa and Alvaro Figueira. 2023. On the Quality of Synthetic Generated Tabular Data. Mathematics 11, 15 (2023). DOI: 10.3390/math11153278
Andre Gheventer, Patricia do Amaral Gurgel, Carlos Henrique Brito, Rafael Maiani de Mello, Sabrina Rocha, Rodrigo Feitosa, and Guilherme Horta Travassos. 2026. Generative AI solutions for software quality: Assessing industrial readiness. Software Qual J 34, 24 (2026). DOI: 10.1007/s11219-026-09754-7
Jenny Hofbauer, Kevin Klaus Gomez Buquerin, and Hans-Joachim Hof. 2023. From SOC to VSOC: Transferring key requirements for efficient vehicle security operations. In Proc. of the 21st escar Europe (2023). DOI: 10.13154/294-10389
ISO/SAE. 2021. ISO/SAE 21434:2021. Road vehicles — Cybersecurity engineering.
Aristeidis Karras, Leonidas Theodorakopoulos, Christos Karras, and Alexandra Theodoropoulou. 2026. Towards LLM-Driven Cybersecurity in Autonomous Vehicles: A Big Data-Empowered Framework with Emerging Technologies. Machine Learning and Knowledge Extraction 8, 2 (2026), 43. DOI: 10.3390/make8020043
Junhui Li, Nikolaos Ersotelos, Michail-Antisthenis Tsompanas, and Gregory Epiphaniou. 2026. Challenges and opportunities of synthetic data generation for machine learning-based intrusion detection systems in in-vehicle networks. Vehicular Communications 58 (2026), 100998. DOI: 10.1016/j.vehcom.2025.100998
Zachary Susskind, Aman Arora, Alan T. L. Bacellar, Diego L. C. Dutra, Igor D. S. Miranda, Mauricio Breternitz, Priscila M. V. Lima, Felipe M. G. França, and Lizy K. John. 2023. An FPGA-Based Weightless Neural Network for Edge Network Intrusion Detection. In Proc. of the 2023 ACM/SIGDA Int. Symp. on Field Programmable Gate Arrays (FPGA ’23). ACM. DOI: 10.1145/3543622.3573140
UNECE. 2021. UN Regulation No. 155 - Cyber security and cyber security management system. United Nations. E/ECE/TRANS/505/Rev.3/Add.154.
Xiang Wang, Jian Zhao, Pengbo Liu, Nianmin Yao, and Zheng Xu. 2026. CAN Bus Intrusion Detection Based on Deep Learning With Data Augmentation for Connected Autonomous Vehicles. IEEE Transactions on Vehicular Technology (2026). DOI: 10.1109/TVT.2025.3603056
R. Aishwarya, V. Vetriselvi, Srinivas Naveen, and Muthuraman A. Ashwin. 2025. An integrated IDS for the Internet of Vehicles using a Large Language Model framework. Internet of Things 33 (2025). DOI: 10.1016/j.iot.2025.101666
MITRE Corporation. 2025. CVE: Common Vulnerabilities and Exposures. [link]. Accessed: Aug. 09, 2026.
João Carmo de Almeida Neto, Leandro Santiago de Araújo, Leopoldo André Dutra Lusquino Filho, and Claudio Miceli de Farias. 2025. Time series forecasting for multidimensional telemetry data based on Generative Adversarial Network in a Digital Twin. Journal of C. S. (2025). DOI: 10.1016/j.jocs.2025.102589
Chloe Soriano de Leon and Cedric Angelo Festin. 2026. Privacy-Preserving Vehicle Intrusion Detection System Using Federated Learning and Homomorphic Encryption. In Proc. of theWorkshop on Computation: Theory and Practice. Atlantis Press. DOI: 10.2991/978-94-6239-638-8_31
Erica Espinosa and Alvaro Figueira. 2023. On the Quality of Synthetic Generated Tabular Data. Mathematics 11, 15 (2023). DOI: 10.3390/math11153278
Andre Gheventer, Patricia do Amaral Gurgel, Carlos Henrique Brito, Rafael Maiani de Mello, Sabrina Rocha, Rodrigo Feitosa, and Guilherme Horta Travassos. 2026. Generative AI solutions for software quality: Assessing industrial readiness. Software Qual J 34, 24 (2026). DOI: 10.1007/s11219-026-09754-7
Jenny Hofbauer, Kevin Klaus Gomez Buquerin, and Hans-Joachim Hof. 2023. From SOC to VSOC: Transferring key requirements for efficient vehicle security operations. In Proc. of the 21st escar Europe (2023). DOI: 10.13154/294-10389
ISO/SAE. 2021. ISO/SAE 21434:2021. Road vehicles — Cybersecurity engineering.
Aristeidis Karras, Leonidas Theodorakopoulos, Christos Karras, and Alexandra Theodoropoulou. 2026. Towards LLM-Driven Cybersecurity in Autonomous Vehicles: A Big Data-Empowered Framework with Emerging Technologies. Machine Learning and Knowledge Extraction 8, 2 (2026), 43. DOI: 10.3390/make8020043
Junhui Li, Nikolaos Ersotelos, Michail-Antisthenis Tsompanas, and Gregory Epiphaniou. 2026. Challenges and opportunities of synthetic data generation for machine learning-based intrusion detection systems in in-vehicle networks. Vehicular Communications 58 (2026), 100998. DOI: 10.1016/j.vehcom.2025.100998
Zachary Susskind, Aman Arora, Alan T. L. Bacellar, Diego L. C. Dutra, Igor D. S. Miranda, Mauricio Breternitz, Priscila M. V. Lima, Felipe M. G. França, and Lizy K. John. 2023. An FPGA-Based Weightless Neural Network for Edge Network Intrusion Detection. In Proc. of the 2023 ACM/SIGDA Int. Symp. on Field Programmable Gate Arrays (FPGA ’23). ACM. DOI: 10.1145/3543622.3573140
UNECE. 2021. UN Regulation No. 155 - Cyber security and cyber security management system. United Nations. E/ECE/TRANS/505/Rev.3/Add.154.
Xiang Wang, Jian Zhao, Pengbo Liu, Nianmin Yao, and Zheng Xu. 2026. CAN Bus Intrusion Detection Based on Deep Learning With Data Augmentation for Connected Autonomous Vehicles. IEEE Transactions on Vehicular Technology (2026). DOI: 10.1109/TVT.2025.3603056
Publicado
11/09/2026
Como Citar
GHEVENTER, André; FARIAS, Claudio Miceli de; MELLO, Rafael Maiani de.
Towards a Hybrid Generative AI-Based Framework for Automotive Cybersecurity: A Feasibility Study. In: WORKSHOP BRASILEIRO DE ENGENHARIA DE SOFTWARE INTELIGENTE (ISE), 5. , 2026, São Paulo/SP.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 137-140.
DOI: https://doi.org/10.5753/ise.2026.32982.
