Cyberactivism and Open Source Intelligence (OSINT): Normative Standards for the Admissibility of Citizen-Gathered Digital Evidence in International Criminal Justice

Abstract

This article examines how the International Criminal Court evaluates citizen-gathered open source intelligence (OSINT). Through doctrinal legal research, it analyses the Rome Statute (Articles 21, 51–52, 64 and 66–69), the Rules of Procedure and Evidence (Rules 63–64, 69 bis, 79 and 81), ICC case law up to the 2025 Yekatom and Ngaïssona judgment and the 2026 Duterte decision on the e-Court Protocol, comparative regimes, and the epistemology of machine testimony. The central problem, it argues, lies not in admissibility but in an “authentication deficit”: the absence of binding standards for deciding whether digitally derived evidence, once admitted, is trustworthy enough to bear the weight placed on it. The article proposes a six-element framework: cryptographic hashing at capture, anchored with an entity independent of the collector; standardised, disclosable chain-of-custody documentation open to challenge by the defence; a permissive inference of authenticity, displaceable by any specific challenge and placing no persuasive onus on the accused, as Article 67(1)(i) requires; source-protection safeguards tested against Articles 67(1)(e), 67(2) and 68(1); a duty to give reasons when departing from the Berkeley Protocol and the Leiden Guidelines; and a best-evidence requirement that operates as a safety valve and never excludes evidence. The framework is distributed across the e-Court Protocol, Regulations of the Court under Article 52 and provisional Rules under Article 51(3). It would narrow the authentication deficit without displacing free proof or burdening the defence, although it cannot establish what an item depicts, where or when it was recorded, or who made it.