Research on the application and challenges of generative Artificial Intelligence in cybersecurity threat detection

Анотація

The fast-paced advancement of cyber threats has created an urgent need for more sophisticated detection strategies, with Generative Artificial Intelligence (GenAI) proving to be a groundbreaking asset in cybersecurity. This research explores the utilization of GenAI models, such as Generative Adversarial Networks (GANs) and Transformer-based architectures like GPT-4, to improve threat detection effectiveness. The research employs a combination of mathematical formulations, experimental evaluations, and comparative analyses to assess the effectiveness of GenAI in identifying and mitigating cyber threats. Experimental results demonstrate that the GAN-based detector achieves an F1-Score of 0.90, while the GPT-4 detector achieves an F1-Score of 0.92, outperforming traditional machine learning and rule-based systems. Additionally, the models exhibit strong adversarial robustness, with scores of 0.85 and 0.88 for GAN and GPT-4, respectively. These results highlight the superior performance of GenAI models in detecting and mitigating cyber threats, even in the face of sophisticated attacks. The study also highlights the dual role of GenAI in cybersecurity, emphasizing its potential for both defensive and offensive applications. While GenAI can be used to generate synthetic attack patterns, analyze log data, and predict potential threats, it can also be exploited by malicious actors to craft convincing phishing emails, automate vulnerability discovery, or create adversarial attacks. This duality underscores the importance of addressing key challenges, such as adversarial vulnerabilities, ethical concerns, and data privacy issues. To mitigate these challenges, the study proposes strategies such as adversarial training, explainability tools, and robust risk management frameworks. These strategies aim to ensure the responsible use of GenAI in cybersecurity, balancing its transformative potential with the need for ethical and secure deployment. By combining theoretical insights with practical applications, this research contributes to the growing body of knowledge on AI-driven cybersecurity solutions. The findings underscore the potential of GenAI to revolutionize threat detection while highlighting the need for further research to address its limitations. Future work should focus on developing more robust adversarial defenses, improving model interpretability, and establishing ethical guidelines for GenAI deployment. This study provides a foundation for future advancements in the field, paving the way for more secure and resilient digital ecosystems.

Ключові слова: Generative Artificial Intelligence; Cybersecurity; Threat Detection; Generative Adversarial Networks (GANs); Transformer Models; Adversarial Robustness; Ethical AI

Як цитувати

@article{internauka10817,
  author   = {Чжен, Бангсю},
  title    = {Research on the application and challenges of generative Artificial Intelligence in cybersecurity threat detection},
  journal  = {Міжнародний науковий журнал «Інтернаука»},
  year     = {2025},
  number   = {3},
  doi      = {10.25313/2520-2057-2025-3-10817},
  url      = {https://www.inter-nauka.com/issues/2025/3/10817}
}

TY  - JOUR
AU  - Чжен, Бангсю
TI  - Research on the application and challenges of generative Artificial Intelligence in cybersecurity threat detection
JO  - Міжнародний науковий журнал «Інтернаука»
PY  - 2025
IS  - 3
SN  - 2520-2057
DO  - 10.25313/2520-2057-2025-3-10817
UR  - https://www.inter-nauka.com/issues/2025/3/10817
LA  - en
ER  -