Ahead of the Union Budget, EY India’s Rajnish Gupta said that the government should release more public data to support India’s push for sovereign artificial intelligence. He added that it is key to building AI models that reflect the country’s cultural and linguistic diversity.Gupta, Partner with EY India’s Tax and Economic Policy Group, said most global large language models are trained largely on data from the US and Europe, resulting in a western bias that often overlooks India’s unique social and cultural context. He argued that making government-held data widely available would enable domestic developers to build AI systems rooted in Indian realities.Speaking to PTI, Gupta said that greater public access to data would help counter this imbalance. “One of the big things which we could do a little more in India is make a lot of data available publicly. (Most of) the answers that you get are very western. They’re very much from a US or a European standpoint,” he said.Emphasising India’s diversity, he added, “We have our own culture. We have our own languages, our own nuances. A lot of this data can be generated by the government and made available to the people who are developing LLMs.”Drawing parallels with national digital successes such as Aadhaar and UPI, Gupta said India could democratise AI by treating data and computing infrastructure as a public good. While public investment should not control innovation, he said it is crucial in laying a strong foundation for the ecosystem.At the same time, Gupta cautioned against heavy regulation, urging the government to maintain a light-touch approach. He praised India for not adopting the European Union’s AI regulatory framework, saying minimal interference would allow local innovators to experiment and build culturally relevant models.“Don’t tinker with this. If you don’t license, don’t create frictions, don’t introduce any new taxes and just let them operate, then I think somebody sitting over here will create something,” Gupta said.
