Is Nvidia Set to Surge Again? Data Center Revenue Soars 117% – How Long Can This Rally Last?

Published on 2026-08-27Last updated on 2026-08-27

Abstract

Nvidia's revenue doubled year-on-year, with data center revenue skyrocketing by 117%. However, the market is beginning to question the sustainability of this growth amid AI capital expenditures, the Rubin generation transition, and competitive pressures.

Records Smashed Again, Market Begins to Question How Long Growth Can Last

Nvidia has once again delivered a stellar earnings report, with revenue, operating profit, and earnings per share all reaching record highs. For the second quarter of fiscal year 2027, revenue was $96.221 billion, a year-over-year increase of 106% and a sequential increase of 18%; net profit was $59.688 billion, up 126% year-over-year; diluted earnings per share were $2.46, a 128% increase from the prior year. Compared to $46.743 billion in revenue in the same quarter last year, Nvidia has managed to double its quarterly revenue in just one year.

However, following the release of the strong results, the stock price did not immediately rise but initially fell approximately 1.3% in after-hours trading. This reflects a shift in market focus: in the past, investors were primarily concerned with whether Nvidia could continue to exceed expectations; now, they are more interested in how long AI infrastructure construction can continue and whether next-generation chips and new business models can support future growth.

Data Center Remains Core Growth Driver, Customer Base is Expanding

Second-quarter data center revenue reached $89.023 billion, a year-over-year increase of 117% and a sequential increase of 18%, accounting for nearly all of the incremental revenue. Nvidia began categorizing its data center customers in the previous quarter into two groups: hyperscale customers, and AI cloud, industrial, and enterprise customers.

Among these, hyperscale customer revenue was $48.710 billion, up 102% year-over-year and 13% sequentially; revenue from AI cloud, industrial, and enterprise customers was $40.313 billion, up 138% year-over-year and 25% sequentially. This indicates that AI computing demand is expanding from a few major cloud giants to AI-native companies, traditional corporations, sovereign AI initiatives, and more industry-specific scenarios. However, investors are also beginning to focus on the profitability and long-term demand of these new customers, not just the size of the orders.

The Chinese market remains a significant variable. Nvidia stated that shipments of data center Hopper products to mainland China accounted for less than 1% of total data center revenue in the second quarter, and third-quarter guidance does not include any data center computing revenue from mainland China.

Blackwell Ultra Ramps Up, Vera Rubin Enters Full Production

The growth in the second quarter's data center segment was primarily driven by the large-scale deployment of Blackwell Ultra infrastructure. Meanwhile, the Vera Rubin platform has entered full production, with related rack systems already operational on partner platforms such as CoreWeave and Google Cloud. Rubin encompasses not only GPUs but also CPUs, networking, software, and system-level solutions.

Nvidia also launched the DSX platform for AI infrastructure builders and continues to expand its Agent Toolkit, PhysicsNeMo, and CUDA-X software libraries. For Nvidia, competition is no longer limited to the performance of a single chip but revolves around the complete ecosystem comprised of chips, networking, software, and systems. The market will now closely monitor the pace of the transition from Blackwell to Rubin and whether the new platform can continue to drive customers to expand their capital expenditures.

AI Infrastructure Enters the Financing Stage

As data center scale expands, Nvidia's involvement in infrastructure construction is also deepening. The company has established strategic partnerships with institutions such as Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs, and KKR, planning to mobilize over $500 billion in third-party capital for AI infrastructure development.

This model, on one hand, helps customers secure funding and expand GPU procurement; on the other hand, it also draws increased market attention to Nvidia's capital pressures, credit support, and potential recurring financing risks. The company issued $25 billion in senior unsecured notes in the second quarter and stated that future investments will primarily support the supply chain, infrastructure, and long-term growth.

Nvidia expects third-quarter revenue of $108.0 billion with a gross margin of approximately 74%. Future growth hinges on answering three key questions: Can Blackwell Ultra and Rubin continue to drive customer procurement? Can AI infrastructure investment maintain its current pace? And, amid competition from AMD, Google's self-developed TPUs, and more custom chips, can Nvidia maintain its market share? While near-term performance exceeding expectations supports the stock price, mid-term pricing will increasingly depend on order sustainability and the return on capital expenditure.

Related Reads

When Real Estate Ownership Goes Digital: What Happens to Your Rights, Risks, and Liquidity?

"Tokenizing Real Estate: Rights, Risks, and the Path to Liquidity" While tokenizing real-world assets (RWA) gains traction, real estate presents unique complexities. Beyond technical token issuance, critical challenges remain: enforcing legal rights, managing the underlying physical asset, and creating genuine secondary market liquidity. This article explores these issues through OneAsset, a Dubai-based commercial real estate (CRE) tokenization platform. OneAsset moves away from simply offering asset fragmentation. Instead, it focuses on institutional-grade infrastructure, prioritizing asset quality, legal enforceability, and operational fundamentals. Each property is held in an independent, single-asset vault, backed by a legally separate Special Purpose Vehicle (SPV) for bankruptcy remoteness. Investors acquire tokens representing the economic rights to a specific property, with precise legal claims defined by the underlying SPV structure. OneAsset emphasizes that tokenization cannot transform a poor-quality asset. Its initial focus is on institutional investors and quality Dubai-based CRE, selected for stable tenant cash flows and a clear regulatory environment. The platform integrates compliance by design, aiming to embed investor qualification and transfer rules directly into the token architecture. A core insight is that asset fragmentation does not automatically create liquidity. True liquidity depends on the asset's inherent quality—its location, cash flow, and valuation—as well as sufficient buyer demand. The goal is not just tradability, but making real estate rights more easily priced, verified, and reallocated. Looking ahead, the article discusses the potential for "AiFi" (AI-powered finance). For AI agents to autonomously allocate capital, investment assets like real estate tokens must become truly "machine-readable." This requires a high degree of standardization in legal rights, valuations, cash flows, and compliance data—a direction OneAsset is pursuing through its structured data reporting. In conclusion, real estate tokenization is shifting from a technology narrative to a focus on asset fundamentals. Blockchain can enhance efficiency and programmability, but it cannot replace sound underwriting, property management, or legal execution. The real work begins after the asset is on-chain.

marsbit21m ago

When Real Estate Ownership Goes Digital: What Happens to Your Rights, Risks, and Liquidity?

marsbit21m ago

After Affecting Two Generations, Meta Ordered to Pay $18 Billion in Damages

After more than two decades, a legal parallel has emerged. In 1998, major U.S. tobacco companies settled for $206 billion, leading to strict advertising bans and warning labels that significantly reduced smoking rates. On August 26, 2026, Meta reached a landmark settlement with U.S. attorneys general, agreeing to pay up to approximately $18 billion and implement mandatory changes to Facebook and Instagram. This historic settlement, one of the largest against a tech company, stems from allegations that Meta deliberately designed addictive features like infinite scroll and push notifications, harming youth mental health and violating child privacy laws. Facing a potential $1.4 trillion lawsuit and a series of unfavorable jury verdicts, Meta chose to settle on the eighth day of trial to avoid a catastrophic ruling. The core of the agreement is not just the financial penalty, which Meta will pay over 10 years, but a series of strict, 10-year product mandates for young users. These include a hard two-hour daily time limit (combined across apps), a default "nighttime block" from midnight to 6 AM, restricted notifications during school hours, hidden "like" counts, an optional non-algorithmic feed, and stronger age verification. An independent auditor will monitor compliance. Crucially, roughly 30% ($5.3 billion) of Meta's payment is contingent on YouTube and TikTok adopting similar measures and paying around $5 billion each. This move aims to create an industry-wide standard and prevent Meta from being competitively disadvantaged. The settlement is being likened to Big Tobacco's "tobacco moment." By legally framing addictive algorithm design as a "public nuisance," it sets a powerful precedent. Nearly 3,000 similar cases are pending against other social media giants, signaling a fundamental shift in regulatory pressure. The era where platforms could deny the addictive impact of their designs on children is effectively over.

marsbit27m ago

After Affecting Two Generations, Meta Ordered to Pay $18 Billion in Damages

marsbit27m ago

Trading

Spot
活动图片