Nanotechnology Perceptions ISSN 1660-6795 www.nano-ntp.com Nanotechnology Perceptions 20 No. S10 (2024) 573-585 AI-Driven Intrusion Detection Systems: Leveraging Deep Learning for Network Security Manish Joshi 1 , Sunderlal Birla 2 , Hemant Pal 1 , Kavita Khatri 3 , Mohit Kadwal 4 , Dinesh Salitra 5 1 Assistant Professor, Department of Computer Science, Medi-Caps University, India 2 Assistant Professor, Department of Computer Applications, Medi-Caps University, India 3 Assistant Professor, Department of Computer Applications, Medi-Caps University, India 4 Assistant Professor, Department of AI and Data Science, Prestige Institute of Engineering Management and Research, India 5 Assistant Professor, CSE Department, Mandsaur University, Mandsaur, India Email: manish_riya@yahoo.co.in In order to improve network security, this study investigates the integration of deep learning and artificial intelligence (AI) in the development of advanced intrusion detection systems (IDS). The inadequacy of traditional security methods has been demonstrated by the exponential rise in cyber threats that target complex network systems. Deep learning techniques are used by AI-driven IDS to evaluate large datasets, allowing for the real-time identification and categorisation of normal and deviant behaviour. This paper examines many deep learning approaches, including Convolutional Neural Networks (CNNs), Deep Neural Networks (DNNs), and Recurrent Neural Networks (RNNs), emphasising how well these methods detect sophisticated attacks, such as advanced persistent threats and zero-day exploits. Furthermore, these systems' performance is assessed using important metrics including recall, accuracy, and precision. The results highlight how deep learning has the ability to transform intrusion detection and hence greatly increase the overall resilience of network security frameworks against changing cyber threats. Keywords: Accuracy, Adversarial, AI-driven, CNN, Deep Learning, Detection, Intrusion, LSTM, Network Security, Zero-Day. 1. Introduction Modern businesses are rapidly becoming digital, which has drastically increased the complexity of network systems and exposed them to an increasing range of cyber threats. For a considerable amount of time, conventional intrusion detection systems (IDS) have been the primary line of defence for protecting vital network infrastructure. But as cyberattacks get