Attack Risks and Defense Research of Generative Artificial Intelligence in Power Grid Security
Abstract：The article focuses on the application of generative artificial intelligence in the field of power grid security, deeply analyzes its potential network attack risks, and proposes two countermeasures (based on adversarial sample detection and a deep neural network-based intrusion detection method). The results show that adversarial sample detection technology can accurately identify generated adversarial samples, while the deep neural network intrusion detection method performs well in terms of accuracy and robustness, providing technical support for power grid security protection.
Keywords：Generative Artificial IntelligencePower Grid SecurityAdversarial Sample Detection
Publication Date：2025-06-25
Pages：3( 51-53 )