Nvidia projects 70% sales growth as AI demand accelerates

Nvidia projected sales growth of approximately 70 percent for the next fiscal year, signaling that the artificial intelligence boom continues to accelerate without abating. The company’s data center division, its largest revenue driver, grew 117 percent in the most recent quarter as clients from cloud providers to enterprise software firms race to secure the computing power required for large-scale AI model training and inference. The projection, delivered during the chipmaker’s quarterly earnings call, exceeded analyst expectations and sent shares higher in after-hours trading.

The 70 percent projection marks a slight deceleration from the triple-digit growth rates of the past three quarters, but it still represents an extraordinary expansion for a company that reported annual revenue of roughly $61 billion last year. Nvidia’s dominance in the AI chip market stems from its CUDA software platform, which has become the de facto standard for developers building and deploying neural networks. Competitors such as AMD and Intel have struggled to break its grip, while custom chips from cloud giants like Google and Amazon remain limited to internal use. The result is a bottleneck: Nvidia’s Hopper and upcoming Blackwell architecture chips are effectively the only high-volume option for training frontier models from OpenAI, Meta, Google, and dozens of startups.

The demand is being fueled by two parallel trends. First, hyperscale cloud providers are spending aggressively on data center infrastructure, with capital expenditure budgets forecast to rise by more than 30 percent this year across the big three US cloud companies. Second, enterprise adoption of generative AI tools is broadening beyond the technology sector. Banks, healthcare firms, and manufacturers are all incorporating AI into workflows, creating persistent demand for inference compute as well as training. Nvidia’s data center revenue of roughly $19 billion in the latest quarter dwarfs its gaming segment, which grew just 19 percent.

The implications for markets and macroeconomic policy are significant. Nvidia’s sales trajectory suggests that the AI investment cycle will continue to absorb large amounts of capital and energy, potentially crowding out other forms of technology spending. It also raises questions about the concentration of market power: Nvidia now accounts for a larger share of the S&P 500’s valuation than any single company has held since the dot-com era. Policymakers are paying close attention, with export controls on advanced chips to China adding geopolitical risk to Nvidia’s growth narrative. The company has warned that its projection assumes no further tightening of those restrictions.

The main uncertainty is not whether AI demand will persist but whether supply can keep pace. Nvidia’s manufacturing partner TSMC is expanding its advanced packaging capacity, but output remains constrained. If production bottlenecks ease, Nvidia could outperform its forecast; if not, clients may face waiting times measured in quarters for the latest chips. Either way, the AI boom is still in its early innings, and the chipmaker that powers it is not expecting to slow down.