On Tuesday, Jensen Huang, the CEO of Nvidia, indicated that he would be very interested in taking part in India’s ambitious ambition to build a 10,000-GPU processing cluster for domestic artificial intelligence computing needs.
According to him, this program not only gives Nvidia a chance to enter India’s quickly expanding industry, but it also demonstrates the government’s dedication to enhancing the IT workforce in the nation.
Huang also underlined the government’s stated goal of making India more than just “the back room for global firms” by establishing it as a hub for value creation.
“AI is utilized in the front office in engineering, marketing, sales, finance, business operations, and strategy development, among other areas. And, India wants to be included in this front office,” Huang remarked in a press conference one day after Nvidia debuted Blackwell, their newest GPU, which is intended to satisfy the growing need for AI processing on a worldwide scale.
The Indian government on March 7 approved the allocation of ₹10,372-crore ($1.2 billion) for several key initiatives, including the development of domain-specific AI models and the creation and upkeep of an open-source dataset covering all 22 official Indic languages, in an effort to combat GPU supply constraints and mitigate the steep acquisition costs.
In addition, the Center intends to launch a public-private partnership (PPP) for AI computing, which would provide 10,000 GPUs for domestic research, academic institutions, businesses, and startups.
According to Huang, India’s ascent to prominence in the global technology industry might be accelerated by properly developing its own AI technology. “Prime Minister Narendra Modi informed me that it made perfect sense for India to import bread instead of exporting flour. Why import the value-added product after exporting the raw material? Why export data from India and import AI instead?
GPUs are an essential part of the infrastructure for training AI models, which depend heavily on data—many refer to it as the currency of the modern world. Furthermore, India’s enormous amounts of data are what are propelling its rise to prominence as a worldwide market.
India has more than 900 million broadband subscribers, according to the Telecom Regulatory Authority of India (Trai)’s most recent subscriber report, released on February 22. This highlights the country’s enormous data reservoir, which may help meet the world’s demand for AI.
According to a Mint article from April 2023, there were several obstacles facing applied research projects at prestigious engineering schools that depended on locally created AI models because of insufficient AI computing infrastructure and a lack of funding for the purchase of extra GPUs.
On February 23, Nvidia’s worth surpassed $2 trillion, having increased by almost 300% in just a single year. Due to its GPUs’ inability to keep up with the rapidly increasing demand worldwide, it is currently the third most valuable company in the world.
Nvidia’s AI-specific GPUs are already being used by a number of Indian enterprises in partnership with them. The company and Mumbai-based Tata Communications are working together to create a cloud platform for AI services.
Yotta Infrastructure, a Hiranandani Group-backed data center provider, has ordered roughly $1 billion worth of GPU chips. On Monday, CEO Sunil Gupta told Mint that the new Blackwell chip will be included in the order by October.
However, analysts noted that over time, the effects of Nvidia’s most recent GPUs might appear gradually.
The most recent GPU announcement from Nvidia affects providers of large or extra-large data centers. Large investments have not yet been made, despite the fact that a number of these companies are quickly growing their services in India. We witnessed a statement earlier this year about purchasing GPUs from Nvidia,” stated Anushree Verma, Gartner’s director of emerging technologies analysis.
It will take some time to determine the demand for these sophisticated GPUs and to amass the resources necessary for an infrastructure investment of this magnitude. I don’t think this will have any immediate effects on the Indian market.”
Therefore, Nvidia may be able to profit from a ready market for GPUs as a result of the Center’s AI goal, even while private enterprises assess returns on investments in AI infrastructure.
According to a March 9 Mint article, Nvidia and other multinational technology companies may gain a strategic advantage by investing in India’s AI GPU infrastructure, as it would grant them priority access to the country’s enormous AI application, end-user, and corporate marketplaces. This will be critical for Nvidia as it works to maintain demand, protect its growth, and maintain its market value.