India Lags in AI Training Data Startups at 2%, Far Behind US’s 40%: Economic Survey

India lags in AI training data startups; needs bottom-up, application-led strategy: Economic Survey

Startups focused on curating and organising training data for artificial intelligence (AI) models have yet to scale meaningfully in India, limiting the country’s ability to fully leverage its vast and diverse data resources, according to the Economic Survey 2025–26.

India accounts for just 2% of global startups working in AI training data, significantly behind the US (40%), the European Union (21%), and the UK (9%), the survey said, citing World Bank data. China accounts for around 5%.

“India holds considerable potential advantage in terms of domestic data sources, but this asset remains underutilised,” the report noted.

Two paths to AI development

The survey highlighted that global AI development has diverged into two approaches.

While Western technology leaders are investing heavily in building large frontier models backed by substantial capital and computing power, many other countries are adopting a decentralised, application-driven strategy.

India, the survey suggested, should prioritise the latter.

Focus on sector-specific AI

Instead of competing directly in building expensive foundational models, the government recommended a bottom-up approach, focusing on application-specific AI solutions.

Given limited access to advanced computing infrastructure and funding, India is better positioned to develop smaller, targeted AI models tailored to sectors such as:

• Healthcare

• Agriculture

• Finance

• Education

• Governance

“These models are more efficient, easier to fine-tune, and better suited to India’s existing infrastructure,” the survey said.

Lower barriers, wider innovation

Such an approach would lower entry barriers and enable startups, research institutions, public bodies, and domain-focused companies to play a larger role in innovation and deployment.

The report also called for greater participation from Indian firms to scale successful AI applications and absorb early-stage risks.

Developing sector-specific solutions could help India’s IT services industry move up the value chain — from being a traditional back-office service provider to becoming an “AI front office,” it added.

Roadmap ahead

Overall, the survey emphasised that a bottom-up, open and application-focused AI strategy offers India the most practical pathway to building a globally competitive AI ecosystem while maximising its domestic data advantage.