Economic Survey startup snapshot: GenAI explodes, but India still lacks data-curation startups

India’s generative AI ecosystem is scaling at a breathtaking pace, but the Economic Survey 2025–26 flags a quieter structural gap: the absence of startups focused on curating training data at scale. While India has rapidly built companies around AI models, applications and services, the foundational data layer—critical to building differentiated and reliable AI systems—remains underdeveloped.

The Survey notes that India’s technology startup ecosystem, now the world’s third largest, comprises an estimated 32,000–35,000 startups, with more than 2,000 added in calendar year 2025 alone. Over 900 of these were funded during the year, signalling sustained investor interest despite tighter capital cycles.

Within this expanding universe, generative AI has emerged as the fastest-growing segment, with the number of active GenAI startups jumping more than threefold—from about 240 in H1 CY2024 to over 890 by H1 CY2025.

Yet much of this growth is concentrated in model development, AI infrastructure and application-layer startups. Funding has largely flowed towards building and deploying models, followed by end-use applications and services, while companies focused on creating, cleaning, labelling and governing training data have failed to emerge at comparable scale.

India enters the AI era with several structural strengths. It ranks among the top global contributors to AI research output and has one of the world’s largest pools of AI and machine-learning talent. By 2024, India had the second-most AI-literate workforce globally, trailing only the United States. The country also possesses a vast reservoir of domestic data.

Beyond AI, the Survey highlights that innovation activity is becoming more broad-based, spanning biotech, digital services and sustainability, supported by improved access to risk capital, credit and incubator-led support. Since the launch of the Startup India initiative in 2016, DPIIT-recognised startups have grown from around 500 to over two lakh by 2025.

Together, these trends suggest that while India’s GenAI story is accelerating, the next phase of value creation may hinge on building the missing data-curation layer—turning India’s raw data abundance into scalable and defensible AI businesses.