AI Won’t Solve Population Health On Its Own

MBC BioLabs recently hosted a conversation between Lee Sanders, MD, MPH, Professor of Pediatrics and Health Policy at Stanford University, and Leah Makley, PhD, COO at MBC BioLabs, to explore how the future of healthcare depends on keeping people, communities, and purpose at the center of innovation. The discussion brought together scientists, entrepreneurs, clinicians, and technology leaders, shaping a compelling vision that AI’s greatest impact will come from combining scientific breakthroughs with trust, equitable access, community engagement, and interdisciplinary collaboration to lead to better health outcomes for everyone.

Healthcare’s biggest opportunities aren’t always scientific ones – they’re implementation problems. Dr. Sanders illustrated this with hypertension, one of the leading preventable risk factors for premature death worldwide. We know how to diagnose it. We have inexpensive, highly effective medications.

Yet millions of people remain undiagnosed or untreated because of barriers to screening, access, adherence, and follow-up.

This is where AI can make a meaningful difference. AI can identify high-risk patients, personalize interventions, and help health systems deploy resources more effectively. But technology alone won’t eliminate health disparities. Better outcomes depend on combining digital tools with trusted community partnerships, accessible care, and healthcare systems designed to reach the people who need them most.

The discussion emphasized that the goal isn’t simply to build smarter technology. It’s to help more people live longer, healthier lives.

Community isn’t an afterthought – it’s part of the product

A recurring theme was that healthcare innovation should be built with communities rather than for them. Whether developing AI diagnostics, digital therapeutics, or population health platforms, founders have an obligation to engage intended users early. Community partnerships reveal barriers, behaviors, and priorities that improve product design, increase adoption, and ultimately lead to better health outcomes.

Several attendees described using digital twin models to better understand patient populations. But the conversation went beyond computational modeling to ask a more fundamental question: Whose biology are we modeling, and who has a voice in building these tools? These are examples of questions that need to be addressed early in the design process.

Health equity is a business decision

Perhaps the most provocative discussion centered on whether population health is embedded in a company’s business model, or treated as an initiative alongside it.

Companies that incorporate equitable trial design, representative datasets, and broad access into their strategy from the outset aren’t just advancing an important social goal. They’re building products that serve larger markets, generate stronger evidence, and create more durable businesses.

This is especially important for startups. Many founders eventually hand their technologies to larger organizations through acquisition or partnership. If commitments to community engagement and equitable access aren’t built into the company’s foundation, they can easily disappear during commercialization.

Throughout the conversation, participants returned to a simple idea: everyone involved in healthcare innovation will one day rely on the healthcare system themselves. We are all patients ourselves. Remembering that shared experience helps keep scientific ambition connected to human purpose.

As AI continues to transform healthcare, the organizations that create the greatest impact won’t necessarily build the most sophisticated algorithms. They’ll be the ones that successfully combine technological innovation with trust, community partnership, and a commitment to improving health for entire populations.