Kenya’s Judiciary has published Draft 2.0 of an Artificial Intelligence Policy, the first formal framework governing the adoption, management, and oversight of AI across all court levels. The Policy is rooted in the Constitution of Kenya, 2010, and applies to everyone from the Chief Justice to third-party technology vendors. It establishes nine foundational principles, including judicial independence, human oversight, fairness, and a distinctly Kenyan commitment to locally led innovation. A multi-tiered governance structure, a mandatory Judicial AI Impact Assessment before any deployment, and a tiered risk classification system together form its core architecture. Significant questions remain, however, particularly around the absence of a self-hosting pathway for Judiciary-controlled AI, the exclusion of alternative dispute resolution from scope, and whether the uniform approval requirement is workable in practice. Our detailed brief unpacks every chapter of the Policy and sets out where it is strong, where it falls short, and what needs to change before it is finalised.
