Nvidia reported a blockbuster quarter, with its data center revenue almost doubling to $75.2 billion and overall revenue jumping 85% to $81.62 billion, significantly beating analyst estimates. Despite robust performance, strong guidance, an $80 billion share buyback, and a raised dividend, the chipmaker’s stock surprisingly dipped, extending its post-earnings losing streak as investor expectations reached ‘unattainable highs’ amidst accelerating AI factory buildouts and a changing competitive landscape.
CEO Jensen Huang emphasized that ‘agentic AI has arrived,’ driving ‘parabolic demand’ for Nvidia’s platforms, including its next-gen Vera Rubin system. The company is also making strategic shifts, redefining its reporting segments and aiming to become the world’s leading CPU supplier, even as it navigates competitive pressures from customers developing their own AI chips and uncertainties surrounding sales to China.
Nvidia, the undisputed leader in AI chips, delivered a powerful first-quarter performance that saw its data center revenue nearly double, propelling overall sales well past Wall Street estimates. Despite the impressive financial results and an optimistic outlook, the company's stock experienced a post-earnings dip, extending a recent trend attributed to 'unattainable highs' set by analysts.
The chipmaker announced adjusted earnings per share of $1.87, topping the $1.76 estimate, on a remarkable $81.62 billion in revenue against an estimated $78.86 billion. This robust growth underscores the accelerating global buildout of AI factories, a phenomenon CEO Jensen Huang described as 'extraordinary.'
Reflecting on the quarter, Huang emphasized that 'agentic AI has arrived,' driving parabolic demand for Nvidia's platforms across various sectors. He highlighted Nvidia's unique position as the only platform supporting every frontier AI model, from Anthropic and OpenAI to Meta and Google's Gemini, and powering hyperscalers with core data processing and machine learning workloads.
In a strong show of confidence, Nvidia's board authorized an additional $80 billion for share repurchases and significantly increased its quarterly cash dividend to 25 cents per share, up from 1 cent. The company's free cash flow also surged to $48.6 billion during the period.
Looking ahead, Nvidia provided optimistic guidance for the current fiscal second quarter, forecasting $91 billion in revenue, comfortably above analysts' average estimate of $86.84 billion. Notably, this outlook does not factor in any data center compute revenue from China, highlighting ongoing uncertainties in that market.
Strategic Shifts and Competitive Dynamics
Nvidia revealed a significant restructuring of its quarterly earnings reporting, now categorizing revenue into two main segments: Data Center and Edge Computing. The Data Center unit, encompassing hyperscale revenue and the ACIE division (AI clouds, industrial, enterprise), accounted for a staggering $75.2 billion in Q1, representing 92% of total sales and a 92% year-over-year jump. Hyperscalers alone contributed over half of this, reaching $38 billion.
The company's new Edge Computing segment covers data processing for agentic and physical AI in devices like PCs, game consoles, robotics, and automotive, posting $6.4 billion in sales with a 29% year-over-year increase. This shift reflects Nvidia's evolving business model, where gaming, once its primary focus, now comprises less than 8% of revenue.
The competitive landscape is rapidly changing. Nvidia acknowledged in a 10-Q filing that some of its key customers, including tech giants like Google, Amazon, Meta, and Microsoft, are developing their own application-specific integrated circuits (ASICs) and other custom chips. This trend, coupled with the potential for these customers to offer competing AI cloud services, poses a challenge to Nvidia's market share and ability to secure foundry capacity.
Despite the market's hunger for alternatives to Nvidia's GPUs, CEO Huang tempered expectations for their custom Groq Language Processing Unit (LPU), calling it a 'niche product for some time' due to its low throughput. This comes after Cerebras Systems' recent blockbuster IPO, signaling a strong demand for specialized AI chips. Nvidia itself entered the custom ASIC space through its $20 billion acquisition of Groq's technology.
Nvidia is also aggressively pursuing the CPU market, traditionally dominated by Intel and Advanced Micro Devices. CFO Colette Kress stated the company aims to become the 'world's leading CPU supplier,' anticipating $20 billion in CPU revenue this year and highlighting the Vera CPU as a new $200 billion opportunity.
Innovation and Global Dynamics
The company’s next-generation rack-scale AI system, Vera Rubin, is expected to be 'even more successful than Grace Blackwell,' according to Huang, with demand anticipated to outstrip supply throughout its lifecycle. Comprising 1.3 million components, including 72 Rubin GPUs and 36 Vera CPUs, Vera Rubin promises 10 times more performance per watt than its predecessor.
Demand for AI infrastructure remains robust, with the price of renting an H100 GPU up 20% year-to-date and A100 cloud pricing up nearly 15%. Nvidia is facilitating the rapid setup of AI compute capacity in over 80 data centers globally.
While Nvidia has invested $18.6 billion in private companies and infrastructure funds, some of which are AI model makers that indirectly utilize its products, global affairs present potential headwinds. The company noted in its 10-Q filing that while the conflict in the Middle East has not significantly impacted its supply chain or Israel operations (5,900 employees), escalation or extension of the conflict 'could affect future product development, supply chain, and revenue, and create business uncertainty.'
Uncertainty also looms over chip sales to China. Despite CEO Huang's recent trip to Beijing, clarity on H200 (older Hopper GPU) sales remains elusive. While some Chinese firms like Alibaba and Tencent have received U.S. Commerce Department approval to purchase H200s, chip export controls were reportedly not discussed in recent high-level U.S.-China talks. China historically represented a significant portion of Nvidia's data center revenue before export restrictions.
Financial Health and Market Position
Nvidia's gross margin stood at a healthy 75%, meeting analyst expectations and demonstrating its sustained pricing power, even amidst a worldwide shortage and rising costs of high-bandwidth memory (HBM). HBM, crucial for high-power GPUs, is in high demand, leading to price increases across the memory market.
The company's rapid growth has propelled its market capitalization to $5.5 trillion, having surpassed $5 trillion in October. While its stock underperformed some semiconductor peers this year, its valuation remains the largest in the world, albeit with recent post-earnings volatility. Industry experts like Gene Munster of Deepwater Asset Management lauded Nvidia's 'remarkable' revenue acceleration, while acknowledging 'noise' around China sales.
As tech giants ramp up capital expenditures on AI infrastructure, projected to exceed $1 trillion by 2027, Nvidia stands to benefit significantly. However, concerns remain about broadening its customer base, as a small number of key clients currently account for roughly half of its business.
