One of the most important requirements for the modern, rapidly expanding network systems is the open network intrusion detection system (IDS) mechanism. Methods to data extraction and machine learning are frequently used to detect network anomalies during the recent years. Meta-heuristic algorithms can be effective for intrusion detection in encrypted packets because they excel at optimizing complex problems with large solution spaces, which is often the case in intrusion detection systems (IDS). Encrypted packets pose a challenge for traditional IDS since they cannot inspect the payload directly. However, meta-heuristic algorithms can be applied to features extracted from encrypted packets or to other aspects of network traffic to detect anomalies indicative of intrusions. Following the provision of the new generation’s absolute foundations meta-heuristics, current trends in research, hybrid meta-heuristics, the absence of theoretical underpinnings, unresolved issues, and concurrent advancements with parallel meta-heuristics and new research opportunities are identified.