As artificial intelligence becomes more deeply embedded in health systems — from diagnostics to health system optimization — experts are warning that regulation is not keeping pace. Without globally coherent standards, AI risks reinforcing the very inequities it promises to resolve.
“One of the biggest challenges we face is the tremendous inequities we already see,” said Dr. Ricardo Baptista Leite, CEO of HealthAI - The Global Agency for Responsible AI in Health. “On one hand, it’s a reflection of the digital divide. But on the other hand, it’s about different levels of maturity in terms of the regulatory capabilities of each country.”
On the sidelines of the 78th World Health Assembly, Devex hosted a series of discussions around the theme of “Global Health’s Next Era.” During one such session called “Is health system transformation possible without resolving the AI governance challenge?” Devex Executive Vice President Kate Warren spoke with Leite to understand how countries can work together to shape equitable and context-specific approaches to AI in health.
“AI has the potential to be the greatest equalizer of our time,” Leite said. “But if we do not design the ecosystem properly, it can actually become the greatest divider of our time.”
Watch the session on YouTube.
Avoiding a digital divide starts with equity and strong governance
Historically, new technologies have not brought equity in their wake. Leite referenced the book “Power and Progress,” a history of technological revolutions that charts how power and wealth were often consolidated rather than distributed.
“[With] every technological revolution over the last 1,000 years … equity was almost always not the outcome,” he said. “The outcome was a tremendous concentration of wealth and power … not necessarily helping people on the ground.” Without deliberate design, he argued, AI could follow the same path.
To prevent history from repeating itself, Leite said HealthAI is helping countries build their capacity to evaluate and scale AI solutions that reflect their own needs and contexts — from the ground up.
Rather than viewing regulation as a barrier to innovation, Leite stressed that robust governance is what will enable innovation to scale. “Fixing the governance at the governmental and regulatory level is a prerequisite,” he said. “Only then can countries fully benefit from the power of AI to improve health outcomes.”
Reimagining care delivery from the bottom up
One challenge, Leite noted, is that many low-income countries hesitate to invest in AI, assuming it’s a “future” solution that should start with the wealthier countries. But he warned against this mindset.
“If you are a bit behind, this is the moment to get ahead,” he said. He pointed to the opportunity for countries to redesign health care delivery entirely. “An AI-based health system … can help countries completely rethink the way health care is being delivered,” Leite said.
To bridge this gap, HealthAI, a Geneva-based nonprofit, created a Community of Practice — with 25 governments and more than 280 institutions already joining to share best practices, surface ethical risks, and make responsible AI implementation feasible, especially in contexts where resources are limited but need is high.
Setting shared rules for a shared future
One of the major risks discussed was that AI is currently being layered onto outdated systems, reinforcing existing inefficiencies rather than solving them.
“In many contexts where AI is being used, it’s being retrofitted into models of care that are broken, and so AI is really accelerating the inefficiencies of today’s health systems,” Leite said.
In response to this fragmentation, HealthAI set up the Global Regulatory Network — a new multilateral platform designed to support countries in creating shared standards, building regulatory capacity, and shaping AI ecosystems. The initiative already includes 10 signatory countries from diverse income levels and geographies and aims to co-develop internationally aligned standards while allowing each country to assess and scale AI tools suited to their unique contexts.
Rather than importing models, Leite stressed that countries must be active co-designers of the AI systems they use. “This is the moment to use AI so your health system can leapfrog and not repeat the mistakes of the richer world.”
Join the Global Regulatory Network. If your government, regulatory agency, or institution plays a role in national AI governance, this is your opportunity to lead in shaping responsible AI in health on a global stage. For further information, visit this site.
About the author
Devex Partnerships
Thanks for reading and for your interest in Devex. In collaboration with our partners, Devex’s partnerships editorial team produces content to promote a partner’s work or perspectives on a particular issue. It gives actors across the global development sector — including nongovernmental organizations, private sector stakeholders, aid agencies and government institutions — the opportunity to go beyond traditional advertising and tell their stories in an impactful way. If you’d like to learn more about how you can shine a spotlight on a particular issue with Devex, please email partnerships@devex.com. We look forward to hearing from you.
