AI in Healthcare Poses Regulatory Challenges and Efficiencies, Say Industry Leaders at IIMA Summit

The integration of artificial intelligence into the healthcare and pharmaceutical sectors presents a dual narrative of operational efficiency and regulatory uncertainty. Speaking at the third edition of the IIMA Healthcare Summit in Ahmedabad, Drugs Controller General of India (DCGI) Rajeev Raghuvanshi highlighted the complexities of overseeing adaptive technologies.

During a walkthrough of the summit's exhibition, Raghuvanshi noted that an exhibit advertising an “AI Doctor” underscored mounting regulatory risks. He pointed out that traditional medical devices and drugs rely on a clear chain of accountability centered on a doctor's prescription. When mobile applications or software begin directly dictating courses of action to consumers, establishing accountability becomes difficult.

Furthermore, Raghuvanshi explained the fundamental mismatch between traditional regulatory frameworks and AI. Regulatory authorities are typically trained to approve products based on fixed characteristics of quality, safety, and performance, ensuring consistency over a product's lifecycle. Because artificial intelligence updates and changes continuously, pulling a product back to an originally approved baseline defeats the core purpose of the technology. He acknowledged that managing a moving target remains a global challenge for regulators.

Despite these regulatory hurdles, AI adoption is accelerating within specific segments of the pharmaceutical industry. According to Raghuvanshi, the adoption remains concentrated among the top 10 percent of pharma companies, particularly those engaged in discovery research. It is also utilized in the generic sector to develop data-analysis algorithms, significantly cutting timelines and improving success rates in clinical trials.

Pankaj Patel, chairman of Zydus Lifesciences and chairman of the IIMA Board of Governors, noted that AI is altering the traditional economics of drug development, which typically requires over a decade and substantial capital due to high failure rates. While machine-designed medicine is not yet fully independent, AI tools help researchers predict molecular behavior, propose candidate molecules, identify trial patients, and recognize failures earlier.

Beyond research and development, Patel highlighted the administrative and diagnostic applications of AI in clinical settings. Tools such as AI scribes can draft clinical notes and summarize complex patient histories or discharge summaries in seconds. Additionally, AI screening tools can analyze chest X-rays or retinal scans in district hospitals where specialists may not be immediately available. The primary value of these tools lies in returning time to overburdened clinicians.

Echoing these sentiments, Abbott India Executive Vice President Vivek Mohan emphasized that trust and deployment models will dictate the success of healthcare AI. He stated that AI's primary role should be to support clinician decision-making rather than replace them, noting that predictive tools, biosensors, and continuous glucose monitoring generate vast amounts of data that improve patient-doctor conversations.

"The intersection of artificial intelligence and healthcare presents a unique challenge for both startups and regulatory bodies. While AI significantly reduces the timelines for drug discovery and helps clinicians manage administrative burdens, the lack of a clear regulatory framework for adaptive software creates compliance hurdles. Healthtech startups building AI solutions must proactively factor in accountability, data consistency, and clinical validation to align with regulatory expectations while scaling their innovations." — Dr. Shishir Gupta, Founder & CEO, StartupLanes