Autonomous Forensics: Integrating AI and Machine Learning in Digital Evidence Standards

Abstract

Digital forensic standards are crucial for maintaining the integrity, admissibility, and reliability of digital evidence in cybersecurity investigations. Frameworks like NIST SP 800-86 and ISO/IEC 27037 are widely used for foundational guidance. However, their limited focus on artificial intelligence (AI), cloud-native structures, and automated evidence processing reduces their effectiveness against modern threats. This study performs a structured comparison  of NIST SP 800-86 and ISO/IEC 27037 to assess their readiness for AI-enabled digital forensics. Through a systematic literature review and standards-based gap analysis, the research identifies shortcomings in automated triage, explainable decision-making, resilience against adversaries, and forensic governance. Results show that AI-assisted forensic methods can cut evidence triage time by up to 35%, lower false positives by 42%, and enhance evidentiary correlation by 28% compared to traditional methods. Based on these results,   an AI-enabled forensic framework is proposed. This framework integrates machine learning, explainable AI, and governance processes to improve transparency and legal   defensibility. The study contributes to the evolution of digital forensic standards by providing practical guidance for developing autonomous, reliable, and future-ready forensic systems.