Article
Investigation of large language models, GenAI, and proprietary AI systems: Digital forensic evidence, readiness and regulation
Mark Scanlon. Forensic Science International: Digital Investigation, Volume 57, Article 302135, 2026.
Contribution Summary
The increasing use of large language models and proprietary AI systems raises concerns about digital forensic evidence and regulation. This paper explores the challenges of investigating these systems, including the need for transparency, explainability, and accountability. It highlights the importance of AI forensic readiness, including the preservation of digital evidence, and the need for regulators to ensure that AI systems are examinable and compliant with digital forensic principles. The paper also discusses the role of the digital forensic community in defining what makes an AI-system record evidentially useful and the need for research into forensic artefacts from AI clients, local LLM environments, and provider disclosure workflows.
Keywords: Digital Forensics; Artificial Intelligence; Large Language Models; AI Forensic Readiness; Regulation; Examinability; Transparency; Accountability
BibTeX
@article{SCANLON2026302135,
title = {Investigation of large language models, {GenAI}, and proprietary {AI} systems: Digital forensic evidence, readiness and regulation},
journal = {Forensic Science International: Digital Investigation},
volume = {57},
pages = {302135},
year = {2026},
issn = {2666-2817},
doi = {10.1016/j.fsidi.2026.302135},
url = {https://www.sciencedirect.com/science/article/pii/S2666281726000922},
author = {Scanlon, Mark}
}