Advancing health equity through inclusive medical AI: findings from an interdisciplinary workshop

Abstract Background Artificial Intelligence (AI) is increasingly shaping healthcare but remains prone to biases, errors, and inequities. Medical inclusivity aims to improve the ability, dignity, and opportunity to access health services, receive high-quality care and achieve equitable outcomes. To realise this, we organised a five-day workshop with 25 experts from 16 countries, including researchers, policymakers, and patient advocates, to co-create solutions for inclusive medical AI. This structured setting facilitated knowledge exchange and interdisciplinary collaboration, offering a first step towards advancing health equity through inclusive medical AI. Objectives The workshop aimed to identify and address challenges in inclusive medical AI. Using the consensus-building tool POLIS, participants prioritised three sub-challenges: 1. Integrating lived experiences and stakeholder voices across the AI lifecycle; 2. Designing data collection practices that promote fairness and prevent inequalities; 3. Fostering regulatory frameworks that uphold human rights and promote inclusivity. A combination of design and systems thinking methods was used to address them, combining scientific insights and stakeholder perspectives to develop actionable solutions. Results Preliminary findings suggest that structured co-creation among diverse stakeholders is critical for responsible AI in health. The proposed solutions can be grouped into active stakeholder engagement, improved AI literacy, robust ethical/regulatory safeguards, and governance frameworks with accountability mechanisms. Conclusions This initiative highlights how participatory approaches can bridge the gap between AI innovation and equitable healthcare. A systemic shift is needed to embed inclusivity as a core AI principle rather than a reactive outcome. Insights from this workshop and its framework can support policymakers and researchers in embedding inclusivity at all stages of the AI lifecycle to promote health equity. Key messages • Health equity can be improved by including participatory approaches, co-creation and interdisciplinary collaboration across the AI life cycle to minimise bias and errors and increase inclusivity. • To achieve health equity, inclusivity must be a core AI principle throughout the entire AI lifecycle, not a reactive fix. Design and systems thinking methods can foster this systemic shift.