Nvidia’s Pricey AI Chips Cometh – $NVDA $AMZN $DELL $MSFT

Nvidia’s (NVDA) upcoming Blackwell GPU-based AI server cabinets are set to come with eye-watering price tags of $2 million to $3 million each, according to a Morgan Stanley report. This pricing strategy has sparked both amusement and concern in the tech industry, as companies prepare to invest heavily in AI infrastructure for the coming years.

Nvidia’s Pricey AI Cabinets

The tech world is abuzz with Nvidia’s latest offering: AI server cabinets that come with a price tag hefty enough to make even Silicon Valley billionaires do a double-take. These high-powered machines, based on the upcoming Blackwell architecture, are set to redefine both computing power and wallet-emptying capabilities. The NVL36, sporting 36 B200 GPUs, is expected to start at a mere $2 million, while its beefier sibling, the NVL72 with 72 B200 GPUs, will set you back a cool $3 million. It’s like choosing between a luxury yacht and a private island – except these islands crunch numbers at mind-boggling speeds.

Demand and Revenue Projections

Demand for Nvidia’s Blackwell-based AI server cabinets is expected to skyrocket, with projections indicating shipments of 60,000 to 70,000 units in the coming year. This surge in demand could translate to an estimated annual revenue of at least $210 billion for Nvidia, setting new records in the AI hardware market. The staggering figures have prompted Nvidia to reportedly increase its orders with chip manufacturer TSMC by approximately 25%, highlighting the company’s confidence in the market’s appetite for these high-powered AI solutions.

Tech Giants Adopting Blackwell GPUs

Major tech companies like Amazon Web Services, Dell, Google, Meta, and Microsoft are all set to adopt Nvidia’s Blackwell GPUs for their AI servers, according to industry sources cited in the Morgan Stanley report. This widespread adoption is expected to drive demand beyond initial projections, solidifying Nvidia’s position as the de facto standard for AI training and inference workloads. The influx of new players in the market, including Asus joining traditional partners like Foxconn, Quanta, and Wistron, may help ease supply constraints for Blackwell-based machines.

Impact on AI Industry

The astronomical costs associated with Nvidia’s AI server cabinets are expected to have far-reaching implications for the AI industry. According to calculations by Sequoia Capital partner David Cahn, the AI sector will need to generate a staggering $600 billion in revenue just to cover the expenses of these machines and their associated data centers. This financial burden could potentially reshape the landscape of AI development, favoring well-funded tech giants while posing challenges for smaller companies and startups. Despite the hefty price tags, the demand for these high-powered AI servers shows no signs of slowing down, indicating a bullish outlook for both AI server manufacturers and GPU developers in the coming years.