Canada's Drug Agency ran a peer-reviewed, 7-project comparison of Lens.org, SpiderCite and Microsoft Copilot against manual systematic-review search for drug-reimbursement evidence, finding …
Ottawa, Canada
Good practice★ 80
Evidence: Observational / pre–post
The US Treasury's Office of Payment Integrity uses ML to screen federal disbursements. In FY2024, ML-driven processes prevented or recovered $4.059 billion …
Taiwan's MoE integrates AI into its free national Adaptive Learning Platform (TALP) for K-12. A quasi-experiment (n=4,442 sixth-graders) showed significant maths gains; …
National adaptive mathematics platform (PAM) for all Uruguayan primary (grades 3-6) and lower secondary students since 2013. An independent CINVE quasi-experimental study …
Montevideo, Uruguay
Good practice★ 80
Evidence: Descriptive / self-reported
Singapore's national online platform serves ~500,000 students and teachers with AI features including an adaptive maths/geography engine, AI lesson-planning copilot, short-answer auto-marking, …
Singapore, Singapore
Good practice★ 80
Evidence: Descriptive / self-reported
The EU AI Act classifies AI for admissions, grading, steering learning and exam monitoring as 'high-risk' — mandating risk management, human oversight …
Brussels, Belgium
Good practice★ 80
Evidence: Descriptive / self-reported