A Study on AI-Powered Threat Intelligence Systems for Proactive Cyber Defence
A Manikandan, Kiranmai Vanaparthi · Global Journal of Engineering Innovations and Interdisciplinary Research · 2025
This study evaluates the comparative performance of traditional versus AI-powered threat intelligencesystems in the content of proactive cyber defence. Traditional threat intelligence systems, characterizedby manual processes and reliance on signature-based detection, exhibit limitations in terms of detectionrate, response time, and overall accuracy. In contrast, AI-powered systems leverage advancedtechnologies such as machine learning and deep learning to significantly enhance threat detectionand response capabilities. Our experimental results reveal that AI-powered systems achieve a higherdetection rate (92.3%) compared to traditional systems (78.5%), coupled with a lower false positiverate (8.7% versus 15.2%) and faster average response time (15.2 seconds versus 45.0 seconds). TheAI systems also demonstrate superior accuracy (94.5%) and are capable of detecting a greater volumeof threats (320 per day) while automating a higher percentage of responses (75.0%). These findingsunderscore the advantages of integrating AI into threat intelligence systems to improve the efficiencyand effectiveness of cybersecurity measures.