India strengthens AI sovereignty with major Nvidia partnerships

India is set to deepen its artificial intelligence capabilities following a series of partnerships announced by Nvidia aimed at strengthening domestic computing infrastructure, supporting sovereign AI development and expanding industry adoption.

The collaborations bring together Indian cloud providers, engineering firms, software companies and academic institutions, signalling a broad-based push to build what Nvidia describes as local “AI factories”, large-scale computing facilities designed to train, deploy and scale advanced AI models within the country.

USD 2 Billion AI Supercluster with Yotta Data Services

A central element of the announcement is a partnership between Nvidia and Yotta Data Services, which plans to invest USD 2 billion in deploying an AI supercluster in India. The facility is expected to feature 20,736 liquid-cooled Nvidia Blackwell Ultra GPUs by August 2026, positioning it among the largest AI compute installations in the region.

The supercluster is intended to provide high-performance computing capacity for enterprises, start-ups and research institutions, reducing dependence on overseas infrastructure. Liquid cooling technology will be used to manage the significant power and heat requirements associated with advanced AI workloads.

In addition to the supercluster investment, Nvidia and Yotta have entered into a separate USD 1 billion commercial agreement to establish a cloud-based AI cluster within Yotta’s existing infrastructure. This cluster will expand domestic access to advanced computing resources for model training, inference and large-scale data processing.

Sovereign AI Factories and Data Centre Expansion

Engineering and infrastructure major Larsen & Toubro will develop a gigawatt-scale sovereign AI factory in India as part of the wider initiative. The project includes expanding data centre capacity in Chennai to 30 megawatts and building a new 40-megawatt facility in Mumbai.

The term “sovereign AI” refers to the development and deployment of AI systems that are hosted, governed and controlled within national boundaries. By building domestic compute capacity at scale, the initiative seeks to ensure that critical AI infrastructure and data remain under Indian jurisdiction.

Cloud service provider E2E Networks will contribute to this effort by building an advanced Blackwell GPU cluster on its TIR platform. The aim is to broaden access to high-performance AI computing for Indian enterprises and developers, particularly those requiring large-scale model training capabilities.

Tools for Indian Developers and Industry Adoption

To support the development of locally relevant AI models, Nvidia has released the Nemotron suite for Indian developers. This includes the Nemotron-Personas-India dataset, which contains 21 million synthetic personas designed to reflect India’s demographic and linguistic diversity. The dataset is intended to help developers create population-scale, localised AI systems tailored to Indian use cases.

Several major Indian IT services firms, Infosys, Wipro, Tech Mahindra and Persistent Systems, will integrate Nvidia AI Enterprise software into their service offerings. The companies plan to deploy AI agents across sectors including financial services and healthcare, where automation, data analysis and decision-support systems are expected to play an increasing role.

Software-Defined Manufacturing and Digital Twins

The initiative also extends to advanced manufacturing. Synopsys, Siemens and Cadence Design Systems are using Nvidia’s CUDA-X and Omniverse libraries to develop software-defined factory solutions for Indian manufacturers including Tata Motors and Havells.

These tools enable the creation of digital twins and simulation environments, allowing manufacturers to design, test and optimise production processes virtually before implementing them on the factory floor. The approach is expected to improve efficiency, reduce costs and shorten product development cycles.

Building Domestic Capacity at Scale

To strengthen the research ecosystem, Nvidia will collaborate with the Anusandhan National Research Foundation to provide academic institutions with free access to AI software and mentorship. The partnership aims to equip universities and research centres with advanced tools and guidance, fostering domestic expertise in AI development.

Taken together, the agreements reflect a coordinated effort to expand India’s AI infrastructure across cloud services, manufacturing, enterprise software and academia. By investing in large-scale GPU clusters, expanding data centre capacity and developing locally relevant AI tools, the initiative seeks to reduce reliance on foreign compute resources while positioning India as a significant hub for AI development and deployment.