<p>Intrusion Detection System (IDS) plays a pivotal role in safeguarding network security by identifying and mitigating malicious attacks. Deep Learning (DL) and Bio-Inspired (BI) algorithms has revolutionized IDS by significantly enhancing their ability to classify and predict network intrusions. Novel hybrid framework, SSA-BOA-CNN introduced in this study, leverages the strengths of Deep Convolutional Neural Networks (Deep CNNs) and bio-inspired metaheuristic algorithms to achieve superior performance in intrusion detection. The proposed framework employs the Salp Swarm Algorithm (SSA) for effective feature selection, ensuring the elimination of redundant and irrelevant data, thereby reducing computational complexity while preserving critical information. Concurrently, the Butterfly Optimization Algorithm (BOA) is utilized to optimize the hyperparameters of the Deep CNN, leading to improved model efficiency and accuracy. CICIDS2019 and CICIDS2018 datasets used for training and testing our model. Extensive experimental evaluations demonstrate that the SSA-BOA-CNN model achieves state-of-the-art performance metrics, including an accuracy rate of 99.16% and a high detection rate, significantly outperforming contemporary IDS methodologies. The hybrid approach effectively addresses key challenges in intrusion detection, such as feature dimensionality reduction, parameter optimization, and accurate classification of diverse attack patterns. This research novelty lies in the synergistic integration of SSA and BOA algorithms with Deep CNN, presenting a robust and efficient solution for enhancing IDS performance. This innovative methodology offers a practical and scalable framework for real-world network security applications, making a substantial contribution to the field of Cybersecurity.</p>