The rapid escalation of cyber threats, including ransomware attacks, phishing campaigns, distributed denial-of-service activities, and advanced persistent intrusions, has exposed the limitations of conventional cybersecurity monitoring infrastructures. Traditional rule-based and signature-driven security systems often fail to identify evolving attack patterns in real time, resulting in delayed incident response, increased false positives, and inadequate situational awareness. These challenges have created a strong demand for intelligent and adaptive cyber defense mechanisms capable of performing automated threat analysis and rapid incident mitigation in dynamic digital environments. This research presents a hybrid deep learning and threat intelligence framework for an AI-enabled cyber incident response and safety web portal designed to improve real-time threat detection and automated security management. The proposed framework integrates Convolutional Neural Networks (CNNs) for spatial feature extraction, Long Short-Term Memory (LSTM) networks for sequential attack behavior analysis, and Transformer-based contextual learning models for advanced cyber incident interpretation. In addition, a dedicated Threat Intelligence Engine is incorporated to correlate Indicators of Compromise (IOCs), vulnerability signatures, and external threat feeds for enhanced cyber incident analysis and risk prioritization. The developed web portal provides intelligent intrusion detection, automated alert generation, incident classification, and centralized threat visualization through a scalable and user-friendly interface. Experimental evaluation was conducted using benchmark cybersecurity datasets and simulated real-time network traffic environments. The proposed framework achieved an overall detection accuracy of 98.1\%, precision of 97.4\%, recall of 97.9\%, and F1-score of 97.6\%, outperforming several conventional machine learning-based intrusion detection approaches. The obtained results demonstrate the effectiveness of deep learning-based security models combined with threat intelligence integration for building reliable and automated cyber response systems capable of supporting modern cybersecurity operations.

Keywords: Cybersecurity, Deep Learning, Threat Intelligence, Incident Response, AI-Based Security, Intrusion Detection, Web Portal Security, Real-Time Threat Monitoring