When Nvidia releases quarterly results this week, market participants will be looking for evidence that the chipmaker continues to translate surging AI demand into accelerating profits. That hope rests in part on a collective capital spending push by large technology firms estimated at $630 billion, which investors expect will drive demand for Nvidia’s high-performance AI processors.
Yet the same wave of investment that has propelled Nvidia’s rise is also spawning competitive headwinds. Several hyperscalers are pursuing in-house silicon intended to lower their cost of running AI workloads, creating a direct challenge to the incumbent GPU supplier.
Over the last three years Nvidia’s shares were a major force behind the U.S. market rally, but in the early part of 2026 the stock has recorded only modest gains, rising roughly 2% so far this year. The market’s cautious tone reflects both the scale of expectations already embedded in Nvidia’s valuation and the new strategic moves by cloud and internet giants.
Alphabet’s Google has moved into the competitive set with its own tensor processing units, or TPUs, and recently agreed to supply Anthropic - the company behind the Claude chatbot - with those chips. Media reports also indicate that Google has engaged in talks to provide its in-house silicon to Meta, which is among Nvidia’s larger customers.
Advanced Micro Devices remains a close market rival and is preparing a new flagship AI server product slated for later this year. And in a bid to shore up its position in inference workloads - where trained AI models generate real-time responses - Nvidia last year reportedly licensed technology from Groq in a deal said to be worth $20 billion. The company also disclosed last week that it has agreed to sell millions of chips to Meta, without revealing the transaction’s value.
At the same time, Nvidia’s own capital decisions have raised questions about the sustainability of hyperscaler spending. The company has prolonged consideration of a previously suggested $100 billion investment in OpenAI, one of its biggest customers, and recent reporting indicated that Nvidia may substitute that plan with a smaller $30 billion commitment.
Investor anxiety over a potential AI spending bubble is palpable. "This earnings in particular is important because people are so concerned about AI spending - whether we’re in a bubble," said Ivana Delevska, chief investment officer of Spear Invest, which holds the company’s shares in an exchange-traded fund. "Showing that earnings are not really decelerating will be pretty important."
Analysts polled by data compilers expect Nvidia to report that quarterly profit rose more than 62% for the period ended in January, a deceleration from the prior quarter’s 65.3% growth. Consensus revenue estimates point to an increase of just over 68% to $66.16 billion. For the current quarter, analysts project revenue could climb another 64.4% to $72.46 billion.
Nvidia has outpaced sales expectations for 13 consecutive quarters, but the margin by which it has beaten consensus has been narrowing. Among analysts, views differ on the degree of upside the company might still show. RBC analysts anticipate management could guide April-quarter revenue at least 3% above estimates, while Delevska — who is bullish on Nvidia — suggested management could forecast sales up to $10 billion above consensus, potentially exceeding estimates by more than 13%.
Management commentary earlier in the year suggested ongoing discussions with customers about data center orders for the following year, and some analysts expect Nvidia to revise upward a previously cited order backlog figure of $500 billion first mentioned in October.
Constraints on growth may come not from demand but from supply. Nvidia and its rivals must secure production capacity at Taiwan Semiconductor Manufacturing Co.'s 3-nanometer lines, where chipmakers are competing intensely for wafer time. Supply bottlenecks at contract manufacturers like TSMC could limit how quickly AI processors reach customers and thereby cap near-term revenue upside.
"We think Nvidia will meet expectations, but it is hard to see them delivering much upside in light of TSMC capacity," Jay Goldberg of Seaport Research Partners wrote in a research note.
Another variable that could influence near-term sales is the potential easing of export restrictions to China. Nvidia’s CEO recently said he hopes China will permit sales of the high-end H200 AI chip and indicated a license was being finalized. Should the company regain fuller access to the Chinese market, sales could receive a notable lift. AMD, for its part, restored AI chip sales to its current-quarter outlook after receiving licenses that allow certain modified processors to be shipped to China.
On profitability, analysts expect Nvidia to report an adjusted gross margin of about 75% for the fourth quarter, a year-over-year increase of more than one percentage point. Market watchers generally do not anticipate that global memory shortages will materially harm Nvidia’s results; they point to the company’s pricing power and the likelihood that it has already secured allocations of high-bandwidth memory for the year, which should insulate margins from higher memory costs.
As Nvidia’s results arrive, investors will weigh the firm’s ability to sustain explosive profit and revenue growth against fresh competition from vertical integrators and the practical limits of foundry capacity. The coming quarterly report will be a key signal about whether demand from hyperscalers and Big Tech will keep fueling Nvidia’s dominant AI position or whether the market is shifting toward more diversified silicon strategies.