Implementation Challenges of AI in Judicial Decision-Making in Jordan

Abstract This study examines the practical challenges of using artificial intelligence (AI) in judicial decision-making in Jordan. While many publications focus on the ethical risks of AI in courts, this paper concentrates on the realities of implementation. Using doctrinal sources and materials that reflect conditions in Jordan, it identifies four connected areas of challenge: (1) technology and funding: uneven digital infrastructure, inconsistent data quality, and the full life-cycle costs of procurement, licensing, operation, and maintenance; (2) organization and culture: judicial scepticism and resistance to change; (3) training needs for judges, court staff, and technical teams; and (4) legal and regulatory uncertainty about data governance, accountability, and evidentiary rules.The findings show that these challenges are varied, mutually reinforcing, and arise both before and during the rollout of AI tools. Although the analysis centers on Jordan, the obstacles it identifies and the remedies proposed are broadly representative of conditions in many developing countries facing similar institutional and resource constraints. The paper offers targeted recommendations for each area and argues that real progress requires clear long-term plans, strong coordination across state institutions, and careful adaptation of international experience to Jordan’s legal context. It concludes that the most credible path is a gradual, use-case-driven adoption in which AI serves as a decision-support tool for judges, rather than a replacement, thereby improving efficiency and consistency while preserving judicial independence.