MAGNET AI RESEARCH LAB
The Magnet AI Research Lab studies how AI performs in digital forensics and publishes the findings, giving digital investigative teams confidence in AI-assisted investigative outcomes.
As AI’s role in digital investigations grows, so do important questions about accuracy and reliability.
The Magnet AI Research Lab, led by Brandon Epstein, conducts empirical research into how AI performs in digital forensics. Working with practitioners and academic partners, the Lab examines real-world questions, measures outcomes, and publishes findings through peer-reviewed research.
Examine how AI performs against defined digital forensic tasks and benchmarks.
Evaluate accuracy, reliability, and practical applications using empirical research.
Submit results for peer review so the digital forensics community can learn from them.
Explore peer-reviewed research, expert perspectives, and conversations about the responsible use of AI in digital forensics.
FEATURED RESEARCH
The Research Lab’s first peer-reviewed study, published in the Journal of Forensic Sciences, examines how an AI expert system performed against trained examiners when interpreting complex media authentication results.
Access the research (opens in a new tab)
Principal Digital Forensic Specialist, Magnet ForensicsLead, Magnet AI Research Lab
Brandon Epstein is Principal Digital Forensic Specialist at Magnet Forensics, where he leads the Magnet AI Research Lab. A former police detective and co-founder of Medex Forensics, which Magnet Forensics acquired in 2024, he specializes in AI and media authentication. Brandon chairs the Scientific Working Group on Digital Evidence (SWGDE) and contributes to several digital forensic and forensic science organizations. He holds a Masters of Science in Recording Arts with an emphasis in media forensics and is a Fellow of the American Academy of Forensic Sciences.
The Magnet AI Research Lab works with practitioners, agencies, and academic researchers to explore meaningful questions about AI in digital forensics. Have a research question, dataset, or potential study in mind? Connect with the Research Lab to start a conversation.
