India’s AI Edge Has a Missing Middle, Says Report

India has become the world’s second-largest market for ChatGPT but ranks 76th out of 118 countries on per-capita AI use, according to a new report that estimates the country could add $1 trillion to its gross domestic product (GDP) by 2035 if it closes the gap.

The report, ‘The India AI Edge,’ released this week by venture firm Z47, OpenAI, and consulting firm Zinnov, draws on OpenAI usage data covering more than 100 million weekly ChatGPT users, a survey of over 100 Indian chief executives, and an analysis of the country’s AI funding landscape.

Fewer than 10% of Indians use AI tools at all, the report found. Activity is heavily concentrated in cities: the top ten urban centers, home to less than one-tenth of India’s population, generate roughly half of all AI use.

Bengaluru and Hyderabad lead the country on developer activity, while Delhi-NCR has the highest penetration.

“What stands out is the depth of use,” Thomas Jeng, head of startups at OpenAI, said in a statement accompanying the report. “India’s young, ambitious user base is already shaping some of the most advanced AI use cases we see globally.”

The $1 trillion GDP projection by 2035 is the figure India needs to reach its $8.3 trillion “Viksit Bharat” target. The report ties that potential to the country’s existing digital public infrastructure—Aadhaar, the Unified Payments Interface, and the Account Aggregator framework—which it says provides a base for AI deployment that most economies will take years to replicate.

AI funding into Indian startups nearly doubled in a year, rising from more than $600 million in 2024 to about $1.3 billion in 2025. Vertical AI startups in healthcare, financial services, and legal technology accounted for a growing share of that capital, the report said.

India also ranks first globally on AI skill penetration. Twenty of the world’s top 100 AI companies have at least one Indian co-founder, though only one is headquartered in India.

“There is a version of this decade where India builds the talent, the applications, and the infrastructure to matter globally,” Vikram Vaidyanathan, Managing Director at Z47, said in a statement accompanying the report. “There is also a version where we remain a large consumer market, and the value accrues elsewhere. Which one we get is being decided right now.”

A survey of more than 100 chief executives found that 95% had embedded AI in their workflows, but maturity varied widely.

The report classified enterprises into four groups based on adoption patterns: tinkerers, which experiment in pockets and allocate about 9% of software spending to AI; democratizers, which have scaled grassroots adoption across teams at 25% spending; enforcers, which issue top-down mandates without execution capacity and report that 19% cannot quantify return on investment; and transformers, which combine both approaches at 17% spending and have deployed AI across four or more business functions.

Outcome gaps between the groups are widening, the report found. About 90% of mature adopters have reduced spending on business-process outsourcing, and more than a third have cut outsourced work by over 25%. Mature customer-support operations are resolving 80% of queries autonomously. Engineering, customer support, and marketing have emerged as the functions where enterprise AI has stuck.

“This is not a story of displacing labor today,” Ashwin Kannan, a Vice-President at Z47 and a co-author of the report, told The Economic Times. “It is about re-architecting the organization.”

Pari Natarajan, chief executive of Zinnov, said in a statement that India’s role in the AI economy could differ from previous technology cycles.

“Every major technology wave has followed a familiar pattern. The world invents, and India operationalizes at scale,” he said. “But AI could be the first wave where India does more than adopt. We now have an opportunity to shape how AI is deployed, governed, and made useful for everyday life.”

The report concluded that India’s challenge over the next five years is shifting AI use beyond major cities, young early adopters, and enterprise pilots into broader workflows across the economy.