AI regulation in healthcare around the world: what is the status quo?

The rapid adoption of artificial intelligence (AI) raises challenges related to ethics, safety, equity, and governance that require robust regulatory frameworks. In most jurisdictions, AI-driven medical devices are already covered by existing medical device frameworks, although new AI-specific legislation may be required to address the challenges posed by recent advancements. This expert review focuses on frameworks and legislation explicitly tailored to AI, synthesizing research literature, government and intergovernmental framework programs, and online media coverage to provide an up-to-date assessment of global AI-specific regulation or strategies in healthcare as of December 2024. Our findings show that only 15.2% (n=30/197) of countries or territories have enacted legally binding AI-specific legislation, including the 27 member states of the European Union (EU) following the adoption of the EU AI Act. A further 9.1% (n=18/197) have drafted legislation, and 28.4% (n=56/197) have issued non-binding guidelines. Notably, 47.2% (n=93/197) of countries or territories do not have an AI-specific framework or legislation in place. Furthermore, our results highlight disparities between the Global North and South, with 60.3% (n=82/136) of Global South countries or territories lacking frameworks or legislation, compared to 18% (n=11/61) in the Global North. In conclusion, our work provides an overview of the status quo of AI regulation around the world, highlights disparities in the adoption of frameworks and legislation, and calls for the need for intergovernmental and regional cooperation.