A new field study comparing clinicians in the United States and India reveals a striking gap in how doctors use artificial intelligence to make diagnostic decisions. Researchers found that AI-generated diagnostic cues become a primary decision-making tool in the mature US healthcare market, but in India's emerging market, clinicians continue to rely on traditional diagnostic methods even when AI recommendations are available. The findings suggest that simply deploying AI tools globally without accounting for local context, workflows, and institutional trust may limit their real-world impact .

How Do Clinicians Actually Use AI Recommendations in Practice?

The study involved clinicians in both countries assessing 15 patient scenarios, with some clinicians receiving AI-generated diagnostic cues and others working without them. The results showed a fundamental difference in adoption patterns. In the United States, clinicians treated the AI output as a "take-the-best" heuristic, meaning they relied on the AI recommendation as the dominant cue to guide their judgment. In India, however, clinicians continued to integrate multiple traditional diagnostic indicators rather than defaulting to the AI suggestion .

This difference matters because it reveals that AI adoption is not simply a matter of access or technology capability. Instead, cultural, technological, and institutional factors shape whether clinicians trust and use AI tools in their daily work. The research suggests that healthcare systems hoping to scale AI globally need to understand these local dynamics rather than assuming a one-size-fits-all approach will work.

What Barriers Prevent AI Adoption in Emerging Healthcare Markets?

The gap between AI availability and actual clinical use reflects deeper structural challenges. In mature healthcare markets like the US, clinicians face significant time pressure, information overload, and uncertainty in decision-making. These conditions make AI-generated summaries of complex patient data particularly valuable. In emerging markets, different institutional constraints and workflows may mean that clinicians have different needs or that the AI tools were not designed with their specific context in mind .

Dr. Umair Shah, Chief Medical Officer of Jaan Health, an AI-powered care management company, has emphasized that technology adoption in healthcare depends on alignment with real-world workflows. "Technology can help close that gap, but only if it is designed around the real workflows and real barriers that patients and communities face, not around an idealized version of how care should work," Shah explained in a recent keynote address. He described this as a "high-tech, high-touch" approach, where technology connects people doing the work with those being served, but the human connection remains essential .

Umair Shah, Chief Medical Officer of Jaan Health, an AI-powered care management company

Steps to Improve AI Adoption in Global Healthcare Settings

• Localize AI Design: Develop AI tools that account for the specific workflows, data availability, and institutional structures of each healthcare market rather than deploying identical systems globally.
• Present AI as Complementary Support: Frame AI recommendations as tools that support clinical judgment rather than replace it, allowing clinicians to validate AI assessments using familiar diagnostic approaches.
• Build Trust Through Transparency: Ensure clinicians understand how AI reaches its conclusions and can integrate those insights into their existing decision-making processes.
• Invest in Change Management: Recognize that technology adoption requires more than access; it requires training, cultural alignment, and organizational support tailored to local contexts.

The research suggests that AI-generated output should be presented as a complement to clinical decision-making rather than a replacement. This approach allows clinicians in different markets to use AI in ways that align with their existing expertise and institutional practices. In the US context, where time pressure and information overload are acute, AI can serve as a dominant decision-making shortcut. In India's context, where clinicians may have different workflows or institutional constraints, AI might function best as one input among many that clinicians integrate into their judgment .

Beyond diagnostic AI, the healthcare technology landscape is expanding rapidly. The American Telemedicine Association recently announced nine finalists for its 2026 Innovators Challenge, showcasing emerging virtual care solutions that address diverse clinical needs. These innovations include AI-driven voice analysis to detect early signs of heart failure, AI-enhanced cognitive behavioral therapy platforms, and AI cardiac monitoring tools for early detection of heart attacks and sepsis. The diversity of these solutions reflects growing recognition that technology adoption requires matching tools to specific clinical problems and patient populations .

The broader lesson from the US-India comparison is that healthcare leaders cannot assume that deploying advanced AI tools will automatically improve clinical outcomes or efficiency. Instead, successful implementation requires understanding local context, building trust with clinicians, and designing systems that enhance rather than disrupt existing workflows. As healthcare systems worldwide invest in AI, this research suggests that the most effective implementations will be those that respect clinical expertise while strategically introducing technology where it genuinely reduces burden and improves decision-making.