During the India AI Impact Summit 2026, former G20 Sherpa Amitabh Kant said that big tech companies are using data from Global South countries to train their Large Language Models (LLM). He stressed that India and other developing countries should develop models based on their own data to get similar benefits from AI.
During a panel discussion at the India AI Impact Summit 2026, Kant said that data from the Global South plays an important role in LLM training and India alone provides about 33 percent more data than the US. He warned that global tech companies can create their own commercial products based on this data and later sell them in the market at a higher price.
He also said that India’s demographic strength and technological ambition give it a unique position in the global AI race. He said that the country’s young population and increasing interest towards AI shows the transformative potential of this technology.