Smart Cyber Threat Intelligence System for Autonomous Threat Aggregation and Reporting
Year: 2026
Type: conference-paper
Abstract: Development of an autonomous cyber threat intelligence (CTI) systems that can identify and analyze hostile activity in real time has become necessary due to the exponential increase in cyberattacks and zero-day vulnerabilities. Organizations are vulnerable to new attack vectors because traditional CTI solutions are limited by static signature-based datasets and delayed threat reports. This study suggests an AI-powered cyber threat intelligence process that combines automated Open-Source Intelligence (OSINT) collection pipelines with a Large Language Model (LLM) to overcome these constraints. The EXA intelligence API gathers real-time threat indicators, malware campaigns, and Common Vulnerabilities and Exposures (CVEs), which are processed using multi-agent reasoning framework. Threat aggregation, vulnerability correlation, severity evaluation, and mitigation strategy creation are all done by the AI on its own. The suggested architecture greatly enhances early threat detection, according to experimental results. By converting unstructured cybersecurity inputs into useful intelligence reports, it increases visibility and lessens the workload associated with human analysis. This study offers an automated, scalable CTI solution appropriate for autonomous defense platforms, CERT (Computer Emergency Response Team) and SOC (Security Operations Center).
Institutions: RV University, B.M.S. College of Engineering
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7165656265