Research report launch
Global experiences with Justice AI are mixed: it can reduce costs and improve efficiency, but its impact on public trust remains unproven, while posing risks to transparency, rule of law, fairness, and privacy.
In fragile and conflict-affected settings, Justice AI is already deployed, from informal shadow applications to official systems for transcription, triage, and case management. Effective integration demands a rights-first, incremental approach with strong governance, international cooperation, and principles like explainability to balance risks with the potential for greater access.
This research paper offers a wide-angle view on lessons and experiences, but also frames the major coming questions concerning privacy, sustainability and equitability.