“AI REGULATION IN HEALTHCARE: RECOMMENDATIONS FOR KAZAKHSTAN”

Abstract

This paper explores the regulatory challenges and opportunities with the integration of artificial intelligence (AI) into Kazakhstan’s healthcare system in regulatory aspects. While generative AI technologies promise significant improvements in diagnosis, rehabilitation, and administrative processes, they also raise critical concerns about transparency, data privacy, algorithmic bias, and accountability. The study employs a systematized scoping review guided by PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) to ensure the rigour and transparency in identifying and analysing relevant literature. Through this structured review process and comparative policy analysis, the research examines global AI governance frameworks—including those of the EU, USA, China, and Russia—and assesses their applicability within Kazakhstan’s context. Drawing from over 50 academic and policy sources, the study identifies five key pillars of responsible AI regulation: risk-based classification, transparency and explainability, legal accountability, ethical governance, and fairness in algorithmic outcomes. It reveals that Kazakhstan lacks formal legal provisions in these areas and proposes a phased, contextsensitive regulatory framework grounded in international best practices. The findings underscore the urgent need for clear legal definitions, institutional oversight, ethical safeguards, and inclusive datasets to ensure the safe and equitable deployment of AI technologies in healthcare. Ultimately, this study contributes a practical roadmap for policy development, institutional reform, and stakeholder engagement, enabling Kazakhstan to govern healthcare AI in a manner that is ethical, effective, and future-oriented.