#125 Nvidia’s Most Sought After H100 AI Chips Lands In India.
Fresh & Hot curated AI happenings in one snack. Never miss a byte 🍔
This snack byte will take approx 4 minutes to consume.
AI BYTE # 📢: Nvidia’s Most Sought After H100 AI Chips Lands In India
While tech giants like OpenAI and Google have invested billions in AI chips manufactured by Nvidia, Yotta Data Services is spearheading India’s most significant foray into AI.
Sunil Gupta, the CEO and co-founder of Yotta, has managed to outpace India’s tech conglomerates, thanks in part to his relationship with Jensen Huang, Nvidia’s renowned CEO. Yotta’s pioneering efforts in AI are expected to be showcased at Nvidia’s upcoming developer conference in California.
Gupta, 52, is a man on a mission. “I’m ambitious, I’m hungry,” he says. “I’m willing to take a bet on the future of AI.”
Yotta’s strategy is to provide high-performance computing capabilities from data centers in India. This will enable Indian corporations, startups, and researchers to develop their own AI services. Nvidia’s chips, the most advanced on the market, are crucial for training large language models and building applications like OpenAI’s ChatGPT and Microsoft Corp.'s coding assistant, GitHub Copilot.
Gupta believes he has an advantage over foreign cloud computing services due to latency issues. He promises to offer the world’s most affordable access to Nvidia AI chips. He’s even considering accepting equity from cash-strapped Indian startups.
The demand for AI is skyrocketing. The global AI market is projected to grow from $168.5 billion in 2022 to over $2 trillion by 2032, according to Spherical Insights & Consulting. “This is a gold rush,” says Stacy Rasgon, an analyst at Sanford C. Bernstein. “It’s still the early days of AI, and companies just can’t buy enough of this stuff.”
However, the dawn of this new era in India was not without its challenges. Indian customs officials were baffled by the high value of the Nvidia chips Yotta had purchased, leading to additional paperwork and bureaucratic hurdles. Gupta spent a day pacing the marble floors of his data center lobby, working tirelessly to get his chips released.
When the delivery truck finally arrived, it unloaded the first of over 4,000 H100 chips that Yotta had ordered from Nvidia. These powerful GPUs, priced between $30,000 to $40,000 each, are affectionately known as ‘Hoppers’ in honor of computer science pioneer Grace Hopper.
The delivery was a moment of celebration for Gupta. A priest blessed the boxes with red vermilion marks and strings of yellow chrysanthemum flowers, while ancient Sanskrit hymns filled the air. Gupta symbolically smashed a coconut on the floor near the truck, marking what he called a “dream moment.”
By June, Yotta expects to have about 20,000 Nvidia chips. While this may not seem significant compared to the tens of thousands purchased by tech giants like Microsoft Corp., Nvidia’s supply is limited. CEO Huang must carefully allocate resources as demand from corporate titans and heads of state continues to grow.
India, however, is receiving special attention. In a meeting with Prime Minister Narendra Modi last September, Huang promised to prioritize any orders from India.
Gupta has serious bona fides in the field. He’s been working for decades on data center businesses and co-founded Yotta in 2019 with the backing of real estate billionaire Niranjan Hiranandani. As a cloud computing operator, Yotta offers companies like Wells Fargo & Co. access to data storage and computing power they can scale up or down as needed, without buying and installing their own hardware.
Tata Group and Reliance Industries Ltd., two of the country’s largest conglomerates, plan to develop AI infrastructure too, but have yet to order Nvidia’s most advanced chips.
An Nvidia spokesperson declined to comment on the specifics of Yotta’s order, pointing out that more will be revealed this week.
One reason for the attention is a global imbalance in AI. If the technology has the potential to transform virtually every industry, as Huang and Microsoft CEO Satya Nadella argue, then countries like India, Indonesia or Turkey are at risk without access.
In India, that could stymie scientific research, startup development and, more broadly, Modi’s ambitions to create a technology superpower. “GPU disparity” is an increasingly popular term for the dilemma. “Countries who don’t have their own AI infrastructure and models will woefully lose the AI race,” said Umakant Soni, cofounder of a nonprofit AI and robotics research park called ARTPARK.
Gupta sees a clear need to develop India-built AI models, trained with local languages and cultural diversity. “India needs sovereign AI, India needs sovereign models,” he said.'