Evaluating Digital Open Source Imagery: A Guide for Judges and Fact-finders

Key Takeaways

• The guide, developed by multiple institutions and experts, serves as an instructional tool for judges and fact-finders to evaluate digital open source imagery, without providing methods for conducting investigations. It focuses on assessing the authenticity, reliability, and probative value of such evidence.
• Digital open source information is increasingly used in courts and human rights bodies for evaluating evidence in international criminal and human rights cases. It includes publicly accessible data such as social media content, videos, and government-published data.
• Key issues for evaluating digital open source information include analyzing content, metadata, source, location, and time information. Special attention is needed for verification, which involves techniques like geolocation and metadata analysis to authenticate and assess evidence.
• Digital open source imagery presents unique challenges such as potential inauthenticity due to misattribution, staging, AI manipulations (deepfakes), and altered metadata. Despite these challenges, even manipulated content can have evidentiary value under careful scrutiny.
• Various methods are employed to determine the location and time of digital imagery, such as geolocation techniques involving satellite imagery and terrain mapping, reverse searches, shadow analysis, and examination of metadata. These methods each have limitations and should ideally be used in conjunction to support evidence verification.