Nvidia's quarterly revenue more than doubled, and the company forecast third-quarter revenue exceeding Wall Street estimates, signaling sustained demand for its artificial intelligence chips. The company also projected revenue growth of approximately 70% in fiscal 2028 and announced an expansion of its collaboration with Amazon Web Services (AWS).
Under the expanded partnership, an additional 2 million Nvidia Graphics Processing Units (GPUs) will be deployed across Amazon's global infrastructure in 2027 and 2028, according to Nvidia's finance chief, Colette Kress. Following the announcement, Nvidia shares climbed more than 4% in after-hours trading.
Nvidia's performance is closely watched as a bellwether for the AI market, as its chips are integral to major data centers and advanced AI models worldwide. This report emerges as major technology firms, including Microsoft and Meta, have reinforced expectations that Big Tech will invest over $730 billion in AI infrastructure this year. This figure represents a significant increase from the $400 billion spent in the previous year.
However, a growing portion of this anticipated spending is being directed towards in-house chip development by tech companies aiming to reduce their dependence on Nvidia's processors, which are costly and subject to supply constraints. Concurrently, Nvidia faces increased scrutiny regarding its role in the AI financing ecosystem, particularly after agreeing to guarantee certain deals through a new alliance with six financial institutions, intended to raise over $500 billion for AI infrastructure.
Analysts are closely monitoring Nvidia's margins, anticipating pressure from the production ramp-up of its Rubin chips and rising memory prices impacting the silicon supply chain. For the third quarter, Nvidia expects an adjusted gross margin of 74%, with a potential variation of 50 basis points. Analysts' consensus estimate for this figure was 74.77%.
The company forecast third-quarter revenue between $105.9 billion and $110.1 billion, a range centered around $108 billion, with a plus or minus 2% variance. This forecast surpasses the average analyst estimate of $104.19 billion, according to data compiled by LSEG.
Nvidia's outlook does not include any revenue from data center chip sales in China. The company's business in China has faced considerable uncertainty. While Washington previously authorized approximately 10 Chinese firms, including Alibaba, Tencent, and ByteDance, to purchase Nvidia's H200 AI chip, deliveries have been delayed for months. Nvidia began marketing its new Vera CPU to Chinese clients in June, indicating potential availability by August, as China considered permitting limited purchases of the H200 by top AI firms. A U.S. Commerce Department official recently stated that shipments had commenced but remained minimal.
As AI adoption grows for task automation and query responses, Nvidia's GPUs face increasing competition from central processing units and custom chips better suited for inference tasks. This trend has led major technology companies to invest in their own silicon solutions. Meta, for instance, plans to start manufacturing its in-house "Iris" AI chip in September, part of a multi-generation custom silicon project aimed at reducing computing costs. Alphabet has reportedly ordered over 3 million chips from Intel for 2028, with Nvidia also exploring Intel's manufacturing technology for a processor that combines four GPUs.
Rivals such as Intel and AMD are also targeting the inference market, and several Chinese firms, including Baidu, have developed their own chips for these tasks.