Lopes, JoséDias Lousã, Mário JorgeDias Lousã, Mário JorgePereira de Morais, José CarlosPereira de Morais, José CarlosMorais, José CarlosLousã, Mário2026-01-292026-01-292026-01-01Lopes, J., Lousã, M., & Morais, J. (2026).0874-8799http://hdl.handle.net/10400.26/61316The widespread adoption of distributed systems, driven by the growth of the Internet of Things (IoT), edge computing, and cloud infrastructure, has substantially expanded the attack surface of modern digital ecosystems. These environments, characterized by high heterogeneity, large data volumes, and stringent latency requirements, make real-time threat detection a complex task. Traditional, pre-dominantly centralized security mechanisms reveal clear limitations in scalability and response time in the face of increasingly dynamic attack patterns. In this context, Artificial Intelligence (AI) and Machine Learning have emerged as essential enablers for more effective intrusion detection. At the same time, the concept of “super nodes” is gaining prominence: strategically positioned network elements with enhanced computational capabilities that act as intelligent intermediaries between edge devices and the central cloud. This study presents a bibliometric analysis of the use of AI-based super nodes for real-time threat detection. The analysis focuses on a sample of 300 publications indexed in the Lens.org database (2015–2025), selected according to the PRISMA 2020 guidelines. Through descriptive indicators and network analysis (such as keyword co-occurrence), research trends, the-matic structures, and emerging directions in this field are identified.engArtificial IntelligenceMachine LearningIoT SecurityEdge IntelligenceBib-liometric Analysis.texthttps://doi.org/10.58086/yv38-0307