A Comprehensive Review on Artificial Intelligence for Cyber Threat Intelligence and Prediction
Waliu Adebayo Ayuba1*
Abstract
The increasing sophistication of cyber threats has intensified the need for intelligent, predictive defense mechanisms. Cyber Threat Intelligence (CTI) leverages data driven insights to understand and anticipate malicious activities, yet traditional CTI methods struggle with scalability, adaptability, and timeliness. This review explores how Artificial Intelligence (AI) is reshaping CTI through automation, anomaly detection, and predictive analytics. It presents a comprehensive analysis of AI techniques including machine learning, deep learning, and graph based models applied to threat detection, malware analysis, and attack prediction. The paper further discusses available datasets, tools, and frameworks such as MISP, STIX/TAXII, and MITRE ATT&CK, along with their integration into AI driven pipelines. Key challenges identified include data scarcity, lack of explainability, adversarial vulnerabilities, and limited interoperability. The review concludes that future research should prioritize multimodal learning, explainable AI, federated intelligence sharing, and human AI collaboration to develop transparent, adaptive, and ethically grounded CTI systems capable of predicting and mitigating evolving cyber threats.
Keywords:
Cyber Threat Intelligence (CTI); Artificial Intelligence (AI); Machine Learning; Deep Learning; Threat Prediction; Knowledge Graphs; Explainable AI; Federated Learning.
