Experts call for ethical AI and stronger research in higher education at Kanpur summit

Over 56 lakh students are registered on UP's Samarth digital platform, with 22.5 lakh first-semester admissions already processed online. Speakers at the Kanpur summit pushed for locally trained AI models and ethical safeguards as the tech moves into classrooms.

Categorized in: AI News Education
Published on: Sep 13, 2026
Experts call for ethical AI and stronger research in higher education at Kanpur summit

The second edition of AI Manthan, held at Chhatrapati Sahu Ji Maharaj University (CSJMU) in Kanpur on Saturday, drew sharp calls for ethical frameworks, stronger research, and locally trained AI models in higher education. The summit, which expanded its focus this year to include healthcare and agriculture, comes as over 56 lakh students across Uttar Pradesh are actively registered on the Samarth digital platform, with nearly 22.5 lakh first-semester admissions processed online for the 2025-26 academic session.

OSD to Governor Sudhir M Bobde shared the figures during a presentation, noting that 49 state universities and more than 9,600 colleges in UP have been brought onto the Samarth portal over the past two years. The platform, introduced by the Centre, automates administrative and academic operations for higher education institutions. More than 21 lakh results have been processed through it so far. Bobde also recognized CSJMU for adopting all 38 available modules - the highest coverage among participating institutions.

A decades-old concept, not a sudden breakthrough

IIT Kanpur director Prof Manindra Agrawal traced AI's evolution from the late 1950s to today's generative models. "There is immense excitement around AI today, but the core concept has been intertwined with computing since day one," he said. "AI is not a sudden breakthrough; its underlying mathematical and structural concepts have evolved alongside computing since the late 1950s."

Prof Agrawal pushed for institutional safeguards and ethical governance to keep AI beneficial. He said higher education institutions must place greater emphasis on critical thinking, problem-solving, and original research - capacities that become more essential as AI tools spread through classrooms and labs.

Smaller language models and the Indian context

Former IIT Kharagpur director Prof Partha Pratim Chakrabarti warned against relying on AI models trained primarily on Western data. "AI models trained primarily on Western data may not adequately understand the languages, cultures and social conditions of Indian communities," he said. He argued that India needs to invest more in smaller language models built on local datasets.

Prof Chakrabarti also addressed a core limitation: AI does not have an inherent understanding of human ethics. Human feedback is required to keep these systems aligned with human expectations and values. For educators designing curricula around AI for Education, that feedback loop is not a technical footnote - it is a design requirement.

Practical applications in healthcare

Dinesh MS from Philips India outlined how AI is already changing clinical workflows. "AI can help doctors by reducing repetitive work and providing additional information during diagnosis," he said. In radiology, AI can serve as a secondary reader, flagging subtle abnormalities that fatigue or interpretation differences might cause a clinician to miss.

CSJMU vice-chancellor Prof Vinay Pathak said the summit's scope had broadened since its first edition, which focused on higher and technical education. "Expanding the scope of AI discussions, we have incorporated healthcare, agriculture and other sectors prominently into the second edition of the summit," he said.

Why this matters for education professionals

The Samarth portal's scale - 56 lakh students, 7,200-plus colleges, 21 lakh results - shows that digital infrastructure in Indian higher education is no longer experimental. For faculty and administrators, that means AI tools are arriving on campuses where the plumbing for large-scale digital adoption already exists. The speakers at AI Manthan were not debating whether AI belongs in education. They were arguing about which models to use, whose data trains them, and what ethical guardrails to build. For educators, the immediate takeaway is practical: demand for AI for Teachers training will grow as institutions shift from adopting platforms to integrating AI into teaching, assessment, and research design. The question is no longer if you will use these tools, but whether you will have a say in how they are built and governed.


Get Daily AI News

Your membership also unlocks:

700+ AI Courses
700+ Certifications
Personalized AI Learning Plan
6500+ AI Tools (no Ads)
Daily AI News by job industry (no Ads)