Artificial intelligence in healthcare is often depicted through gleaming hospital robots or precision oncology dashboards. The Indian School of Business (ISB), Hyderabad, chose a different lens for its three-day AI in Public Health Innovation Programme: district hospitals, surveillance officers, health secretaries, and the messy realities of state systems.
The programme brought together 78 doctors and government officials from Andhra Pradesh, Telangana, Goa, and Assam. Rather than segregate clinicians and administrators, ISB intentionally seated them side by side, on the premise that AI deployments fail as often on governance as they do on accuracy.
From the outset, faculty framed AI not as a silver bullet, but as a set of tools that must plug into India’s evolving digital health rails. Sessions unpacked the National Digital Health Mission (NDHM), the Ayushman Bharat Digital Mission (ABDM), and the Digital Personal Data Protection (DPDP) Act in accessible but unflinching detail. Participants pored over case studies where well-intentioned analytics pilots had stumbled due to misaligned data flows, inadequate consent, or unclear accountability.
At the heart of the curriculum was SAHI—the Smart AI for Health Initiative—positioned less as a discrete project and more as a mindset. SAHI thinking demands that every algorithmic intervention be backed by clear evidence of clinical benefit, continuous validation in local contexts, and explicit mechanisms to detect and correct bias. It insists that someone, somewhere, be formally responsible when AI-mediated decisions affect patients.
Hands-on labs exposed participants to concrete examples: triage models that help rank patients in emergency rooms, image analysis tools for tuberculosis and diabetic retinopathy screening, and predictive dashboards that flag districts at risk of vector-borne outbreaks. Doctors critiqued model outputs against their own heuristics; civil servants asked uncomfortable questions about procurement, maintenance, and vendor lock-in.
A recurring theme was the tension between scale and nuance. State health systems crave tools that can be deployed quickly and uniformly, yet health needs vary sharply between coastal Andhra, tribal Assam, and urban Telangana. Faculty encouraged participants to think in terms of modular building blocks—shared data standards and governance frameworks—that can support diverse local applications rather than one-size-fits-all apps.
On the data protection front, the DPDP Act was not treated as a hurdle to innovation, but as a guardrail. Simulated exercises forced participants to grapple with scenarios where a seemingly harmless analytics dashboard could reveal sensitive information about individuals or small communities. Groups debated anonymisation thresholds, role-based access, and redress mechanisms if data misuse were alleged.
Crucially, the programme did not shy away from political economy. Discussions surfaced questions about vendor incentives, the risk of “pilotitis” driven more by press releases than public health needs, and the challenge of sustaining projects beyond the tenure of individual champions. Some officers spoke candidly about previous attempts to roll out AI-powered tools that fizzled out once initial grant funding dried up.
By the final day, mixed teams of doctors and officials were presenting mini-roadmaps: not wishlists of futuristic AI, but sequences of concrete steps—data cleaning in a particular registry, capacity building for frontline workers, procurement reforms—that would make their systems AI-ready in the first place.
For ISB, which sits at the intersection of management, policy, and technology, the programme served as proof-of-concept that executive education can move beyond PowerPoint optimism. It showed that doctors and bureaucrats will willingly interrogate AI if given a structured, jargon-light space to do so; and that they can become informed demand-setters rather than passive recipients of vendor pitches.
Scaled up across more states and cadres, such training could quietly reshape the trajectory of AI in Indian public health. Instead of scattered pilots and opaque black boxes, we might see a generation of systems that are context-aware, governed by design, and judged by their impact on primary health centres as much as tertiary hospitals.
-Shaik shakeel Ahmed




