Newsletter
India just crossed a milestone that most people didn’t even notice. The country has become one of the world’s most advanced users of AI, especially in areas like coding, data analysis, and complex problem-solving.
In fact, India now ranks among the top five globally in “thinking usage” per person, which basically means people here are not just asking AI to write birthday messages, they are using it to solve actual, layered problems.
At the same time, AI usage in India is nearly three times more concentrated in a handful of cities compared to countries like the US or UK. The top ten cities, led by Delhi NCR, Bengaluru, Hyderabad, and Chennai, account for about 50% of all AI users, even though they represent less than 10% of the population.
So what’s really happening here is not a simple “India is adopting AI fast” story. It is a story of two Indias moving at very different speeds. In the top metros, AI is already becoming part of daily work, especially for developers, analysts, and knowledge workers.
The gap becomes even sharper when you look at advanced use cases. Data analysis usage is up to 30 times higher in leading cities compared to lagging ones, coding usage is 4 times higher, and AI developer tools show a 9 times gap. That is not a small difference. That is a structural divide.
But here’s where it gets interesting. Outside these metros, AI is not absent. It is just being used very differently. In states like Assam, Odisha, and Tripura, a large chunk of AI usage is focused on education.
In Assam alone, 22% of all interactions relate to learning, which is about 20% higher than the national average. Meanwhile, regions like Jammu and Kashmir, Punjab, and Kerala are using AI more for health-related queries, with nearly 1 in 10 conversations focused on health in some regions.
Last year, the Indian government has gone all in on building its own AI backbone. Under the ₹10,000+ crore IndiaAI Mission, the country is building a massive public compute infrastructure with tens of thousands of GPUs.
At the same time, companies like Sarvam AI are working on sovereign large language models, designed to run on Indian data, in Indian languages, and within Indian regulatory frameworks. The idea is simple. If AI is going to power everything from governance to healthcare, India does not want to depend entirely on external models.
Meanwhile, Indian companies are not waiting for perfect conditions. Enterprise adoption has already picked up pace. Close to 87% of Indian firms are now using AI in some form, across areas like product development, marketing, operations, and supply chains.
But there is a catch here too. While adoption is high, deep expertise is still limited. Many companies are experimenting and scaling at the same time, which means the learning curve is steep and ongoing.
Then there is the infrastructure race. India has crossed 100 million weekly ChatGPT users, and global players are now partnering with Indian firms to build large-scale AI-ready data centres.
See, India is not behind in AI.
In fact, in terms of usage and enthusiasm, it is ahead of many developed markets. But the real challenge is distribution. The benefits of AI are currently clustered in a few urban pockets, while the rest of the country is only beginning to tap into its potential.
Bridging this gap is not just about more apps or better models. It is about cheaper
access, better infrastructure, local language support, and actual skill-building at scale.
The next phase of India’s AI journey will not be defined by how powerful the models become. It will be defined by how widely they spread. Because if the current trends hold, India has the ingredients to become one of the most important AI markets in the world.
The only question is whether that growth will stay concentrated in a few cities, or whether it will actually reach the millions of people who could use it the most.
