Year: 2024
Type: conference-paper
Abstract: It has recently emerged that generative Artificial Intelligence (AI) has been significantly developed and can create realistic speech, images, and video. Although such innovations provide many advantages, they pose major threats to the network's security. The research question of the study will, therefore, be the role and effects of generative artificial intelligence on network security, especially deepfake, artificial intelligence-driven phishing, and multimedia content manipulation. This research uses a case study analysis of real-life experiences of security threats in networked systems to show how generative AI can be misused. The findings reveal that: 1. Generative AI can create highly developed cyberattacks such as deepfake social engineering and AI-supplemented phishing, which use visually and audibly persuasive appeals to manipulate users. 2. The problem with this kind of security threat highlighted by AI is that most traditional security solutions lack awareness of such threats since they are incapable of effective real-time user authentication, over-dependence on visual and biometric verification, and lack of AI-awareness training. 3. There are significant gaps in cybersecurity policies and training. Many organizations lack formal protocols for addressing AI-related security risks, and often, organizations are applying generative AI technologies. In addressing the aforementioned vulnerabilities, the study recommends an improved security model that incorporates deepfake detection, more secure authentication, and even advanced anomaly detection systems. In furtherance, there should be improvements in policy that aim at commanding the AI media, ensuring comprehensive training of the employees on cybersecurity matters, and drives that aim at educating the public on how to enhance the strength for discouraging AI cybercrimes.
Author: Baiqiang Liu
Institution: University of Shanghai for Science and Technology
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W4411088886