AI-Driven Autonomous Threat Detection and Response in Cybersecurity: A MITRE ATT&CK Framework-Aligned Approach

This systematic review synthesizes findings from over 50 recent studies on AI-driven autonomous threat detection and response systems, focusing on their alignment with the MITRE ATT&CK framework. A comprehensive literature search was conducted using major databases and conference proceedings from 2020–2025. Studies were selected based on relevance to AI paradigms (machine learning, deep learning, reinforcement learning, and agentic AI) and explicit mapping to MITRE ATT&CK tactics and techniques. Our synthesis reveals that while AI integration enhances detection accuracy and response speed, significant challenges remain regarding scalability, false positives, and adversarial vulnerabilities. We identify research gaps in evaluation frameworks and propose future directions for robust, MITRE ATT&CK-aligned cybersecurity solutions.