Karnataka sets up responsible AI committee to shape state's AI governance framework

The Karnataka government has formed a Committee on Responsible Artificial Intelligence (AI). The first meeting of the committee was held in Bengaluru today.

The committee will develop a comprehensive policy framework to guide the ethical, safe, and transparent deployment of AI across government systems and public services.

Chaired by Kris Gopalakrishnan, Co-founder of Infosys, and co-chaired by Dr N Manjula, Secretary of the Department of Electronics, IT, Biotechnology and Science & Technology, the committee brings together senior leaders from industry, academia, law, and policy.

Members also include representatives from tech companies such as IBM, Accenture, Wipro, Kyndryl, IIIT Bangalore, NASSCOM, and Sarvam AI.

Karnataka IT Minister Priyank Kharge said the initiative reflects Karnataka's commitment to leading responsibly as the state enters what he called its "Deeptech Decade". He emphasised that innovation must be matched with accountability and citizen trust.

The committee's agenda spans several critical areas, including establishing core responsible AI principles, building a risk classification framework for government AI applications, and identifying prohibited practices such as social scoring, unlawful surveillance, and discriminatory profiling. It will also recommend safeguards for high-risk sectors including healthcare, education, policing, and welfare delivery.

Other focus areas are data governance aligned with India's Digital Personal Data Protection Act, transparency mechanisms requiring disclosure when citizens interact with AI systems, independent audits, and procurement guidelines for government departments adopting AI tools.

Gopalakrishnan called the initiative an opportunity for Karnataka to become the first state in India to develop a truly comprehensive responsible AI framework, one that creates 21st century jobs and strengthens the innovation ecosystem.

The committee is expected to submit an interim report within 60 days and final recommendations within 90 days, covering a policy framework, a risk classification system, and an implementation roadmap.