AI Must Move Beyond Hype To Deliver Real-World Impact : IT Secretary S. Krishnan

New Delhi, Feb 18 Artificial intelligence must prove its value beyond the hype cycle and deliver solutions that meaningfully improve people’s lives, Union Electronics and Information Technology Secretary S. Krishnan said at the India AI Impact Summit 2026. Stressing real-world impact over experimentation, he said India’s AI push is focused on applications that enhance public services, boost productivity and strengthen governance. From healthcare and education to agriculture and manufacturing, AI must translate into measurable outcomes, he noted, adding that responsible scaling, privacy safeguards and institutional adoption will determine whether AI succeeds in delivering citizen-centric growth.

Focus on Real-World Applications

Krishnan said the India AI Mission has been structured to address diverse real-world challenges by providing access to compute infrastructure, AI models and datasets.

“We are providing compute, models, and data for one reason only — to build applications with real impact,” he said, pointing to startups working across healthcare, agriculture, education and manufacturing as key drivers of AI-led transformation.

He added that while governments may face resource constraints in areas such as teachers, doctors and judges, AI-driven productivity enhancements can significantly improve service delivery and quality.

“The challenge is to choose what works, scale it responsibly, protect privacy, and ensure public money creates measurable outcomes,” he said.

People-Centric and Sovereign AI

The high-level session explored the dual priorities of people-centric AI deployment and sovereign technological capability, focusing on translating algorithms into deployable public service solutions.

According to the official statement, discussions centred on ensuring that investments in compute, models and data lead to scalable applications that enhance governance, productivity and citizen welfare.

Iqbal Singh Dhaliwal, Global Executive Director of Abdul Latif Jameel Poverty Action Lab (J-PAL), stressed the importance of rigorous evaluation in AI deployment, cautioning against over-reliance on untested “silver bullets”.

Michael Kremer, University Professor in Economics at the University of Chicago and affiliated with J-PAL, cited early evidence of AI impact in traffic enforcement, automated driver’s licence testing, healthcare and personalised adaptive learning. He noted that limited AI-assisted interventions have already demonstrated significant improvements in student learning outcomes.

The session underscored that AI’s long-term credibility will hinge on measurable outcomes, responsible scaling and citizen-centric governance, rather than technological hype.