In todays world, ensuring the security and integrity of networks is of utmost importance. With the evolving digital landscape, malicious actors employ increasingly sophisticated tactics to gain unauthorized access to sensitive information. Intrusion Detection Systems (IDSs) are pivotal in safeguarding networks by identifying abnormal activities or intrusions. Traditional rule-based IDSs have limitations in detecting evolving threats, leading to the emergence of machine learning-based approaches. This paper explores the integration of Ant Colony Optimization (ACO) and K-means clustering (ACO-KM) to enhance intrusion detection on the NSL-KDD dataset, addressing the need for adaptive IDSs capable of identifying emerging threats. The paper presents a comprehensive literature review, details the ACO-KM algorithm, and evaluates intrusion detection performance. The approach is implemented using NETBEANS IDE and provides flexibility in data selection and classification. Results indicate superior accuracy in detecting Denial of Service (DoS) attacks, emphasizing the efficacy of the proposed ACO-KM algorithm in bolstering network security.