India’s sovereign AI ambitions are entering a deeper infrastructure phase, with more than 1,700 AI-native companies collectively raising about $5.5 billion in equity funding, according to a latest report released by Tracxn today.

The report, titled India and the Sovereign AI Shift, outlines how India’s AI ecosystem is transitioning from an application-led growth model to a more infrastructure-backed and compute-intensive framework supported by policy coordination and capital formation.

As per the report, AI funding in India hit a high of $1.1 billion in 2022, before moderating in 2023 and 2024 amid a broader global venture capital slowdown. Funding rebounded to $856 million in 2025 and has already reached $626 million year-to-date in 2026, indicating renewed investor appetite.

Globally, AI funding has surpassed $473 billion, with OpenAI, Anthropic, and xAI together accounting for an estimated $170 billion, roughly 36% of the total fund, underscoring the capital intensity and concentration defining frontier AI development globally. 

IndiaAI mission emerges as key driver

A key driver of the structural shift is the ₹10,372 crore IndiaAI mission, which aims to strengthen domestic capabilities across foundational model development and compute infrastructure. Under the mission, 12 companies building foundational and specialised models have received structured GPU allocations to reduce capital barriers associated with large-scale model training.

The report highlighted that domestic model development currently spans parameter sizes ranging from 2.9 billion to 105 billion, including mixture-of-experts architectures designed for inference efficiency. Among the beneficiaries, Sarvam AI has been allocated 4,096 NVIDIA H100 GPUs, supported by approximately ₹99 crore in compute subsidies, reflecting the scale of public–private coordination underway.

Infra capital gains prominence

The broader ecosystem spans enterprise AI, vertical-specific applications, consumer platforms and infrastructure layers. Unlike earlier phases, when Indian startups primarily built applications on globally available models and foreign-controlled cloud infrastructure, the current phase reflects deeper participation across upstream layers of the AI stack, including compute provisioning and semiconductor-linked incentives.

Infrastructure-focused capital is also gaining prominence. In one of the largest recent rounds, Neysa raised $600 million at an enterprise valuation of around $1.4 billion, highlighting growing institutional confidence in AI cloud and GPU-backed infrastructure as an investable asset class.

In contrast, India’s funding base remains smaller in absolute terms but is more widely distributed across companies rather than concentrated within a handful of hyperscale labs.

Deployment depth anchors sovereign strategy

India’s sovereign AI positioning is also anchored in deployment depth. The country operates large-scale digital public infrastructure platforms such as Aadhaar, UPI, DigiLocker, Open Network for Digital Commerce and Bhashini. These systems generate population-scale digital interactions across identity, payments, commerce and language services, creating structured real-world environments for AI deployment.

Global hyperscalers are responding with localized infrastructure investments. Partnerships such as Google–Adani’s $15 billion data centre initiative and AWS’s $8.4 billion infrastructure commitment reflect growing compute localization driven by India’s digital user scale.

According to the report, India’s AI trajectory reflects a hybrid model- neither isolationist nor purely consumption-driven - built on incremental infrastructure deepening, coordinated public–private capacity formation and expanding domestic training cycles. While foundational model ecosystems remain in active build phase, India’s participation in the global AI economy is shifting from peripheral adoption toward deeper internalization of core intelligence layers.