AI-Driven Threat Intelligence System (AIDTIS): Leveraging large language models for automated threat research and detection development

Cyber threats are evolving at an unprecedented pace, challenging organizations to stay ahead of sophisticated adversaries. Traditional threat research methods often require extensive manual effort, leading to delays in identifying and mitigating threats. This paper proposes an AI-Driven Threat Intelligence System (AIDTIS), a theoretical approach that leverages large language models (LLMs) to automate and enhance threat research and detection development. Our simulations and theoretical models suggest that such a system could significantly reduce threat research time, improve detection accuracy, and streamline security operations. The proposed solution demonstrates potential for efficiency improvement, potentially cutting analysis time from 8 hours to just 1 hour per report while maintaining high-quality threat intelligence and detection outputs. Moreover, this research highlights the urgency of adopting AI-driven threat research across the broader cybersecurity landscape, particularly in the United States. With rising cyber threats targeting critical infrastructure, financial systems, and government networks, the proposed AIDTIS provides a scalable model for national security initiatives, demonstrating how AI-driven intelligence could revolutionize threat detection and mitigation.