AI Capital Hub?NVIDIA FY2027 Q2 Earnings: Growth Still Accelerating, AI Supercycle Enters Multi-Track Parallel Phase

By now, we've grown accustomed to a gradually solidifying narrative: the money in AI infrastructure is flowing from "chip purchases" to electricity, to cooling, and to cloud providers' own custom silicon. Nvidia, as the most expensive link in this chain, is supposedly seeing its data center growth momentum progressively diluted. Over the past few quarters, bears have repeatedly used "high base effect" and "peak growth" to set a ceiling on Nvidia, compounded by the accelerating push toward custom ASICs among hyperscalers and persistent doubts about whether the AI capex cycle has peaked. As a result, market expectations for Nvidia have been steadily tempered.

Heading into this earnings report, the consensus estimate from forty-one analysts was only 1.2% above the company's own guidance—almost a signal that Wall Street merely wanted confirmation that Nvidia's data center growth hadn't stalled. Nvidia had beaten expectations for five consecutive quarters, but the magnitude of each beat had been narrowing, leading markets to increasingly wonder whether this curve would eventually hit its inflection point.

After the US market close on August 26, Nvidia delivered results that completely shattered all expectations. The data center business remained the unequivocal growth engine, with both revenue and Q3 guidance significantly surpassing consensus estimates. Jensen Huang's opening line in the earnings statement set the tone for the quarter: "AI has reached its inflection point. It's doing useful work. Its tokens are productive and profitable. Now, compute is revenue."

A Year of Transition from Supplier to Capital

Over the past year, Nvidia has gradually become a large-scale AI infrastructure investment firm. Gaming GPUs have long become a footnote in its history, and "data center supplier to the AI industry chain" is now just one of its many identities. On September 18, 2025, Nvidia invested $5 billion in Intel as a strategic move—a rare instance of investing in a competitor that made the relationship between the two old rivals even more delicate. This came just one month after the US government agreed to acquire approximately 10% of Intel's shares for $8.9 billion. The back-to-back positive news of government equity participation and Nvidia's injection ignited market imagination about the beleaguered chip giant's return to the table. Intel's stock surged 23% that day, marking its largest single-day gain since 1987. On December 29, 2025, the deal formally closed, with Nvidia purchasing over 214.7 million Intel shares via private placement.

In the eleven months that followed, Nvidia launched a near-unlimited investment spree. On September 22, 2025—just four days after the Intel investment—Nvidia committed up to $100 billion to OpenAI. In January 2026, Nvidia invested $10 billion in xAI. In July 2026, it added to its OpenAI commitment, bringing the total to $250 billion. Most recently, on August 20, Nvidia completed its most complex investment transaction yet, doing three things simultaneously: first, paying $6 billion to license Poolside's AI model, Model Factory; second, making an additional $1 billion investment at a $12 billion pre-money valuation; and third, extending job offers to over 100 Poolside employees, who will join Nvidia's Nemotron open-source model project.

In effect, Nvidia has once again extended its capital influence deep into a frontier AI model company. Notably, Poolside specifically stated in its letter to investors: this is neither an acquisition nor an acqui-hire. The company will remain independently operated, with all three co-founders staying on. But if it continues to independently compete in open-source model development, it will need more Nvidia hardware than might realistically be available otherwise. In other words, it was the scarcity of compute that ultimately pushed the company into Nvidia's embrace.

The relentless stream of capital moves is blurring the boundaries of the entire AI industry: supplier, shareholder, lender, and customer—roles that used to be clearly delineated—now simultaneously apply to Nvidia, which sits at multiple seats on the poker table at once. This closed capital loop has also made the "circular financing" criticism more concrete. Nvidia invests money in customers, who use it to buy Nvidia GPUs. That money then returns to Nvidia's income statement as revenue, which boosts the stock price, giving Nvidia more capital to invest in the next customer. The oft-repeated claim—"Nvidia invests $100 billion in OpenAI, and then OpenAI gives that money right back to Nvidia"—describes precisely this cycle. Looking at the broader ecosystem's operational mechanics: Nvidia, Microsoft, and Oracle all invest in AI developers, who subsequently become major buyers of their cloud services, causing the same capital to circulate among multiple companies and be booked as revenue at each stop.

If the ultimate commercial returns on AI infrastructure remain elusive, this seemingly efficient capital cycle will snap first at its most fragile link. The risk lies in the fact that every link in this chain rests on the assumption that "the next link can keep paying": whether OpenAI can earn back its data center costs with its models, whether cloud providers can rent out their compute, and whether AI startups can find a profitable model before burning through their funding. If any node's revenue underperforms—say, a customer cuts procurement—Nvidia's orders will decline, leading to writedowns in the book value of its equity investments, while the funding and guarantees previously extended to support these customers could turn into potential bad debts.

The defense's logic, however, runs like this: these investments are incremental and tied to actual deployment milestones. Take the OpenAI deal as an example: each new investment tranche is triggered only upon deployment of one gigawatt. The first gigawatt triggers a $1 billion investment, with subsequent tranches priced based on OpenAI's prevailing valuation at that time. This means that if deployment doesn't actually happen, the investment doesn't happen either—there's no room for "manufacturing revenue out of thin air."

This is the ongoing market debate over Nvidia's capital strategy that has persisted for months, with concerns about AI infrastructure leveraging even spilling into the bond market. Since the start of 2026, hyperscalers and Nvidia-affiliated entities have issued $225 billion in bonds, a staggering 973.7% year-over-year surge! And regardless of whether these risks ultimately materialize, they will all become stumbling blocks on the staircase of Nvidia's stock price.

Another Quarter of Beating Expectations

Nvidia's FY2027 Q2 earnings report looks impeccable by any traditional standard. Revenue hit $96.221 billion, up 106% year-over-year and 18% quarter-over-quarter—the highest annual growth rate since Q2 FY2025. This came in $5.2 billion above the midpoint of the company's own guidance and more than 4% above analyst expectations. Adjusted EPS was $2.22, up 120% year-over-year and nearly 6% above market consensus. Non-GAAP operating income reached $63.956 billion, up 124% year-over-year, corresponding to an operating margin of approximately 66%, beating the analyst estimate of $61.19 billion. Operating expenses were $8.232 billion, below the expected $8.32 billion—revenue outperformed while cost control also beat expectations. However, after the earnings release, Nvidia's stock initially ticked up in after-hours trading, then quickly turned lower, at one point falling 4%.

Data center revenue reached $89 billion, up 117% year-over-year and 18% quarter-over-quarter, exceeding the analyst range of $85.8–85.9 billion and representing approximately 92.5% of total company revenue. Revenue from hyperscalers came in at $48.71 billion, significantly beating the expected $43.55 billion by nearly $5.2 billion. AI cloud, industrial, and enterprise applications revenue was $40.31 billion, slightly below the market expectation of $41.96 billion. Edge computing revenue was $7.2 billion, up 27% year-over-year and 13% quarter-over-quarter, above the expected $6.61 billion. This breakdown indicates that the quarterly beat was primarily driven by hyperscalers, while the expansion pace of enterprise, industrial, and some AI cloud-related demand ran slightly behind what the market had anticipated.

This is a signal that warrants caution. Previously, the market (including many analysts) leaned toward the view that Nvidia's customer base was rapidly diversifying, with enterprise demand poised to take over the baton. But this quarter's data shows that the four major cloud providers remain the absolute primary growth engine, while enterprise-side ramp-up hasn't fully caught up. Compute & Networking revenue of $88.3 billion also beat the expected $84.69 billion, reflecting the same logic: large-scale cluster construction remains the core form of current demand.

On gross margin, Q2 GAAP and non-GAAP gross margins both came in at 75.0%, flat sequentially and up approximately 2.5 to 2.6 percentage points year-over-year. Against the backdrop of nearly $100 billion in quarterly revenue and an extremely tight data center supply chain, maintaining a 75% gross margin demonstrates that Nvidia's pricing power in the AI accelerated computing market remains exceptionally strong. However, Q3 guidance calls for 74.0%, down about 1 percentage point from Q2 and below the analyst expectation of 75%. Several distinct factors are at play, including yield fluctuations and cost disruptions typical during the early production ramp of new platforms, as well as rising supply chain costs—such as sustained price increases in critical components like HBM, advanced packaging, and substrates. This shares the same root cause as the memory crisis Apple previously faced.

Changes in product mix may also make this decline the starting point of a longer-term trend. As the share of full-rack system deliveries increases, the cost structure of products differs from selling GPUs alone. The most significant product signal in the report is that Vera Rubin is accelerating toward full-scale production. Racks are already running at partners including CoreWeave, Google Cloud, Microsoft Azure, Oracle Cloud Infrastructure, and Nebius. This detail matters—it means Rubin is already running in real customers' data centers. This new system-level architecture, combining Vera CPUs, Rubin GPUs, Spectrum-6 networking, BlueField, security, storage, software toolchains, and the DSX platform, will serve as the core pillar of the growth narrative for coming quarters. Nvidia repeatedly uses the term "AI factory" to emphasize that it's not supplying chips but rather the complete systems for building, operating, and scaling AI compute infrastructure. This is crucial for the valuation narrative: one of the market's biggest concerns had been whether Nvidia could smoothly transition to the next platform after the Blackwell demand peak. Rubin's successful ramp-up directly extends the visibility of the growth cycle. But the platform transition also brings short-term uncertainties—new platform ramps typically involve changes in supply chain, delivery cadence, customer acceptance, and cost structure, which is one possible reason for the lower Q3 gross margin guidance.

Looking at this earnings report, the quality of Nvidia's core business growth is genuine—possibly even better than the headline numbers suggest. Excluding $7.77 billion in equity investment gains, non-GAAP net income was $53.954 billion with a 66% operating margin, indicating that the core business itself is highly profitable and hasn't been flattered by investment gains. Cost control also beat expectations. However, ACIE falling $1.6 billion short of expectations indicates that customer diversification is proceeding more slowly than the market had imagined. This means the risk of over-reliance on the four major cloud providers hasn't eased in the near term. Declining gross margins, sharply reduced free cash flow, surging accounts receivable, and an expanding investment portfolio paint a picture of a company using a heavier balance sheet and longer cash conversion cycles to support faster growth.

AI Narrative, Competitive Landscape, and Supply Chain: Three Parallel Threads

Nvidia's fundamentals remain solid, but each of the three pillars supporting those fundamentals is undergoing its own transformation. First, the AI narrative. As we know, Wall Street's focus on the AI investment cycle has shifted from "compute expansion" to "return validation." Jensen Huang's long-term assessment is that AI infrastructure spending could reach $3–4 trillion annually by the end of the decade, driven by the widespread adoption of agentic AI—no longer simple Q&A chatbots, but AI systems capable of sustained decision-making, tool invocation, and multi-step task execution.

This assessment itself is hardly controversial. The disagreement lies in the timeline and the path to returns. Nobody today is questioning whether investing in AI is investing in the future; cash flow and the urgency of returns are what the market cares about. As discussed earlier, risk is amplifying in the bond market. The real, rapid realization of ROI is the core proposition for the entire AI industry chain—it's the endgame of this business model.

Additionally, the accelerating custom ASIC track will continue to erode the ceiling on Nvidia's share of the AI chip market. At the data level, custom silicon's share of the overall AI chip market is projected to rise from 20.9% in 2025 to 27.8% in 2026—the fastest-growing competitive threat in the current AI chip market. Google's TPU, Amazon's Trainium, and Meta's MTIA—all three major cloud providers are accelerating their in-house chip efforts. Broadcom, one of the biggest beneficiaries in this space, has already reached approximately $10.8 billion in quarterly AI-related revenue.

What makes this threat unique is that its driving force isn't performance—it's bargaining power. Cloud providers fully understand that custom chips can't match Nvidia in terms of generality and software ecosystem. But they're willing to sacrifice some performance to reduce dependence on a single supplier, gain supply chain autonomy, and secure leverage in procurement negotiations. This is also why this thread won't upend Nvidia's position in the near term, but it warrants sustained tracking over the long run. The key isn't whether TPU can beat Blackwell; it's what proportion of capital expenditure cloud providers are willing to divert away from Nvidia.

On the supply chain front, the data center supply shortage—including scarcity across the entire product lifecycle—shows no signs of easing. In fact, the Q1 earnings report already set the tone for the entire year on supply tightness. Nvidia raised total supply to $145 billion in Q1, with management explicitly stating they're not immune to supply challenges but are confident in supporting customer growth. Management expects Nvidia to remain supply-constrained throughout the entire Vera Rubin lifecycle.

This statement can be interpreted in two ways, and these two interpretations have entirely different investment implications. For bulls, demand is so robust that the chipmaker can't keep up with capacity—supply shortages mean entrenched pricing power and exceptional order visibility. On the other hand, supply scarcity also functions as a marketing narrative: it continuously creates a sense of product scarcity and urgency to buy, prompting customers to lock in orders early rather than wait on the sidelines. Regardless of which interpretation holds, the supply chain—especially TSMC's CoWoS advanced packaging capacity and HBM memory capacity—remains the most critical variable determining the ceiling on Nvidia's growth.

This also explains why, in this AI hardware cycle, upstream suppliers occupy the most certain beneficiary positions: TSMC controls advanced process nodes and packaging, while Micron and SK Hynix control HBM capacity. Their capacity allocation determines how much Nvidia can ship, and how much Nvidia can ship, in turn, determines the revenue ceiling for the entire AI infrastructure chain.

Putting these three threads together—business fundamentals, earnings expectations, and the three underlying trends—a clear structure emerges: demand, capacity, and capital are the core variables sustaining the entire system's growth rate.

Nvidia remains Nvidia. Having completed the leap from chip supplier to builder of the compute world, it is now laying the groundwork for an era that hasn't fully arrived. The growth story continues, and it still stands in the most dazzling position of this age. But the brighter the spotlight, the longer the shadow cast. The unresolved questions it has left for the market will become one of the most important storylines for US equities in the coming quarters.