•  
  •  
 

Bulletin of Chinese Academy of Sciences (Chinese Version)

Keywords

cybersecurity, large models, intelligent cyber offense and defense systems, artificial intelligence

Abstract

Cybersecurity serves as a critical pillar for national security and social stability. Large models in cybersecurity are emerging as key enablers for the intelligent transformation of cyber offense and defense systems. As one of the most advanced core technologies in artificial intelligence, large models are introducing new research directions and application paradigms in the cybersecurity domain. This study systematically reviews the current landscape of cybersecurity-oriented large model applications and products, and explores their deployment scenarios in practice. It further analyzes the development trends in model capabilities, industry ecosystems, and trustworthiness, while identifying major practical challenges such as data privacy protection, model security, continuous updates, and result reliability. The study argues that coordinated multi-stakeholder collaboration is essential to advancing intelligence sharing, fostering new talent development frameworks, cultivating specialized expertise, building a model ecosystem, reducing development costs, improving policy and standards, and strengthening regulatory oversight. These efforts are crucial to accelerating the intelligent and modern evolution of cybersecurity technologies and practices, and to constructing secure and intelligent cyber offense and defense systems.

First page

1671

Last Page

1682

Language

Chinese

Publisher

Bulletin of Chinese Academy of Sciences

References

[1] 中华人民共和国国务院新闻办公室. 携手构建网络空间命运共同体. (2022-11-07)[2025-02-18]. https://www.gov.cn/zhengce/2022-11/07/content_5725117.htm. The State Council Information Office of the People's Republic of China. Building a community of shared future in cyberspace. (2022-11-07)[2025-02-18]. https://www.gov.cn/zhengce/2022-11/07/content_5725117.htm. (in Chinese)

[2] 中国信通院. 大模型安全研究报告（2024年）. 北京: 中国信通院, 2024. CAICT. Research Report on Large Model Security (2024). Beijing: CAICT, 2024. (in Chinese)

[3] Perrina F, Marchiori F, Conti M, et al. AGIR: Automating cyber threat intelligence reporting with natural language generation// 2023 IEEE International Conference on Big Data (BigData). Sorrento: IEEE, 2023: 3053-3062.

[4] Hu Y L, Zou F T, Han J J, et al. LLM-TIKG: Threat intelligence knowledge graph construction utilizing large language model. Computers & Security, 2024, 145: 103999.

[5] Singla T, Anandayuvaraj D, Kalu K G, et al. An empirical study on using large language models to analyze software supply chain security failures// Proceedings of the 2023 Workshop on Software Supply Chain Offensive Research and Ecosystem Defenses. Copenhagen: ACM, 2023: 5-15.

[6] Deng Y L, Xia C S, Peng H R, et al. Large language models are zero-shot fuzzers: Fuzzing deep-learning libraries via large language models// Proceedings of the 32nd ACM SIGSOFT International Symposium on Software Testing and Analysis. Seattle: ACM, 2023: 423-435.

[7] Xia C S, Paltenghi M, Tian J L, et al. Fuzz4ALL: Universal fuzzing with large language models// 2024 IEEE/ACM 46th International Conference on Software Engineering (ICSE). Lisbon: IEEE, 2024: 1547-1559.

[8] Lemieux C, Inala J P, Lahiri S K, et al. CodaMosa: Escaping coverage plateaus in test generation with pre-trained large language models// 2023 IEEE/ACM 45th International Conference on Software Engineering (ICSE). Melbourne: IEEE, 2023: 919-931.

[9] Meng R J, Mirchev M, Böhme M, et al. Large language model guided protocol fuzzing// 2024 Network and Distributed System Security Symposium. San Diego: Internet Society, 2024.

[10] Jensen R I T, Tawosi V, Alamir S. Software vulnerability and functionality assessment using llms// 2024 IEEE/ACM International Workshop on Natural Language-Based Software Engineering (NLBSE). Lisbon: IEEE, 2024: 25-28.

[11] Hu S H, Huang T S, İlhan F, et al. Large language model-powered smart contract vulnerability detection: New perspectives// 2023 5th IEEE International Conference on Trust, Privacy and Security in Intelligent Systems and Applications (TPS-ISA). Atlanta: IEEE, 2023: 297-306.

[12] Mao Z, Li J L, Jin D M, et al. Multi-role consensus through LLMs discussions for vulnerability detection// 2024 IEEE 24th International Conference on Software Quality, Reliability, and Security Companion (QRS-C). Cambridge: IEEE, 2024: 1318-1319.

[13] Sun Y Q, Wu D Y, Xue Y, et al. GPTScan: Detecting logic vulnerabilities in smart contracts by combining GPT with program analysis// 2024 IEEE/ACM 46th International Conference on Software Engineering (ICSE). Lisbon: IEEE, 2024: 2048-2060.

[14] Das Purba M, Ghosh A, Radford B J, et al. Software vulnerability detection using large language models// 2023 IEEE 34th International Symposium on Software Reliability Engineering Workshops (ISSREW). Florence: IEEE, 2023: 112-119.

[15] Zhang C Y, Liu H, Zeng J T, et al. Prompt-enhanced software vulnerability detection using ChatGPT// 2024 IEEE/ACM 46th International Conference on Software Engineering: Companion Proceedings (ICSE-Companion). Lisbon: IEEE, 2024: 276-277.

[16] Qi J X, Huang S H, Luan Z Z, et al. LogGPT: Exploring ChatGPT for log-based anomaly detection// 2023 IEEE International Conference on High Performance Computing & Communications, Data Science & Systems, Smart City & Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys). Melbourne: IEEE, 2023: 273-280.

[17] Han X, Yuan S H, Trabelsi M. LogGPT: Log anomaly detection via GPT// 2023 IEEE International Conference on Big Data (BigData). Sorrento: IEEE, 2023: 1117-1122.

[18] Liu Y L, Tao S M, Meng W B, et al. Interpretable online log analysis using large language models with prompt strategies// 2024 IEEE/ACM 32nd International Conference on Program Comprehension (ICPC). Lisbon: IEEE, 2024: 35-46.

[19] Guastalla M, Li Y Y, Hekmati A, et al. Application of large language models to DDoS attack detection// Security and Privacy in Cyber-Physical Systems and Smart Vehicles. Cham: Springer, 2024: 83-99.

[20] Ferrag M A, Ndhlovu M, Tihanyi N, et al. Revolutionizing cyber threat detection with large language models: A privacy-preserving BERT-based lightweight model for IoT/IIoT devices. IEEE Access, 2024, 12: 23733-23750.

[21] Vörös T, Bergeron S P, Berlin K. Web content filtering through knowledge distillation of large language models// 2023 IEEE/WIC International Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT). Venice: IEEE, 2023: 357-361.

[22] Jamal S, Wimmer H, Sarker I H. An improved transformer-based model for detecting phishing, spam and ham emails: A large language model approach. Security and Privacy, 2024, 7(5): e402.

[23] Deng G L, Liu Y. PentestGPT: Evaluating and harnessing large language models for automated penetration testing// 33rd USENIX Security Symposium. Philadelphia: USENIX Association, 2024.

[24] Happe A, Cito J. Getting pwn'd by AI: Penetration testing with large language models// Proceedings of the 31st ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering. San Francisco: ACM, 2023: 2082-2086.

[25] Bianou S G, Batogna R G. PENTEST-AI, an LLM-powered multi-agents framework for penetration testing automation leveraging MITRE ATTACK// 2024 IEEE International Conference on Cyber Security and Resilience (CSR). London: IEEE, 2024: 763-770.

[26] Wang P, D'Cruze H. AI-assisted pentesting using ChatGPT-4// ITNG 2024: 21st International Conference on Information Technology-New Generations. Cham: Springer, 2024: 63-71.

[27] Hilario E, Azam S, Sundaram J, et al. Generative AI for pentesting: The good, the bad, the ugly. International Journal of Information Security, 2024, 23(3): 2075-2097.

[28] 向夏雨, 顾钊铨, 曾丽仪. 网络威胁情报共享与融合技术综述. 网络空间安全科学学报, 2024, (2): 2-17. Xiang X Y, Gu Z Q, Zeng L Y. A survey of cyber threat intelligence sharing and fusion technologies. Journal of Cybersecurity, 2024, (2): 2-17. (in Chinese)

Download
Request a Copy

Share

COinS