At New Delhi summit, India pushes for a "Global AI Commons"

India is the second-largest market for both ChatGPT and Claude. Now the country wants to shape AI policy at the summit in New Delhi.

India has 100 million weekly active ChatGPT users, making it ChatGPT's second-largest user base after the United States. CEO Sam Altman shared the figures in an article published in the Times of India ahead of the India AI Impact Summit in New Delhi. India is also the second-largest market for Claude.ai, according to an Anthropic blog post.

Starting Monday, more than 20 heads of state and government, including France's Emmanuel Macron and Brazil's Luiz Inácio Lula da Silva, will convene in New Delhi, along with AI business leaders such as Altman, Dario Amodei, and Sundar Pichai. According to the Financial Times, India will seek international agreement on a "global AI commons."

What the Indian government means by this, however, is more modest than the term might suggest. Abhishek Singh, chief executive of India's AI mission, said: "The global AI commons means creating a repository of use cases for AI in key sectors, which can then be shared." India wanted to make AI applications "interoperable and available to the global community to ensure they are diffused widely and adopted at scale," with priority given to social ends such as education, health, and agriculture.

AI researcher and governance expert Jibu Elias, who has worked with the Indian government, takes the idea further. "Frontier AI is being built, trained and controlled by a handful of firms and states, mostly in the US and China," he told the Financial Times. For India and the global south, Elias said, the "commons" is "a way to argue that foundational AI capabilities, data sets, standards and safety norms should not quietly become private infrastructure controlled by a few companies."

Widespread adoption, few homegrown models

Despite being the home of leading IT companies such as Infosys, Wipro, and Tata Consultancy Services, India has not been a leader in developing AI large language models. The Modi government has criticized the country's $300 billion IT industry for not investing in product innovation.

Recently, India is positioning itself more as a site for data centers. Google, Microsoft, and Amazon last year announced plans to invest more than $50 billion for AI investment in the country. Monetization remains a challenge: OpenAI rolled out a sub-$5 ChatGPT Go tier that was later made free for a year for Indian users.

Anthropic bets on Indian languages and partnerships

Alongside the summit, Anthropic officially opened its Bengaluru office — its second in Asia after Tokyo. Six months ago, the company launched a company-wide effort to improve model quality by curating higher-quality, more representative training data in 10 of the most widely spoken languages throughout India, including Hindi, Tamil, and Urdu.

In the public sector, Anthropic says it is supporting Adalat AI to improve access to judicial services with a national WhatsApp helpline addressing India's 50 million pending court cases. The Indian Ministry of Statistics and Programme Implementation, with the support of nonprofit Bharat Digital, recently launched the first official Indian government MCP server. Companies including Air India, CRED, and Cognizant are already deploying Claude in production, according to Anthropic.