NVIDIA surpasses expectations once more, forecasts $91 billion for Q2, and approves an additional $80 billion in buybacks.

      Q1 revenue reached $81.6 billion, reflecting an increase of 85% year-on-year, with data centre revenue at $75.2 billion and net income at $58.3 billion. The quarterly dividend has been raised from one cent to 25 cents per share. This buyback marks the second $80 billion authorization in just three quarters.

      On Wednesday, Nvidia announced that its first-quarter revenue was $81.6 billion, an 85% increase from the previous year and a 20% rise from the last quarter. The company expects second-quarter revenue to be approximately $91 billion, with a margin of error of 2%, compared to the analyst estimate of $86.84 billion.

      The board also approved an extra $80 billion for share buybacks and increased the quarterly dividend from one cent to 25 cents per share. Initially, the stock rose following this announcement before stabilizing near flat in after-hours trading.

      During the quarter, data centre revenue totaled $75.2 billion, surpassing the average analyst prediction of $72.8 billion. Net income for the February-April timeframe was $58.3 billion, a 37% sequential increase and over 200% growth year-on-year, coinciding with Al Jazeera’s report characterizing it as ‘record profit and revenue amid the AI chip boom.’

      This new buyback authorization is the second $80 billion program approved by the board in three quarters. In total, the company has now authorized more than $160 billion for share repurchases, with a market capitalization that has, as of May 20, remained above $4 trillion for the majority of the past quarter.

      The increase in the quarterly dividend from one cent to 25 cents, while minor in the context of the buyback, indicates a shift in the company’s financial strategy towards returning capital rather than solely focusing on funding growth with excess cash.

      The guidance for Q2 suggests a sequential growth of about 12% compared to Q1. Analysts believe this guidance may be conservative in light of the visible capital expenditure commitments from hyperscalers for the latter half of the year. AWS, Microsoft, Google, and Meta collectively project around $470 billion in capital expenditure for 2026, with a large portion expected to be fulfilled through Nvidia’s products.

      The significant capital expenditure from Meta exceeding $115 billion, the AWS GB200/GB300 NVL72 deployments, and the $25 billion TPU-cloud joint venture between Google and Blackstone illustrate the anticipated demand for Q2 and Q3.

      In terms of competition, Nvidia’s share in the data-centre market against AI-accelerator alternatives remains a key narrative. Discussions between Tenstorrent, Intel, and Qualcomm, as well as Alibaba’s T-Head Zhenwu M890 announcement, represent two notable alternatives to Nvidia, the US-based RISC-V/x86 option and the domestic accelerator route from China, respectively.

      However, neither alternative has yet reached a scale that would significantly impact data-centre growth. The licensing arrangement on H200 sales to Chinese clients remains a crucial factor for the fiscal year 2027's second-half guidance.

      Nvidia did not reveal the timeframe for the new $80 billion buyback authorization, the geographic division of the Q1 data-centre revenue (between the US and the rest of the world), specific Q2 shipment volumes for the GB300 NVL72 included in the guidance, or the H200 licensing revenues related to China.

      CEO Jensen Huang, in his prepared remarks during the call, indicated that the AI infrastructure development is still "in the early innings," a phrase commonly used by the company. The next significant data point will come with the fiscal year 2027 Q2 results, especially the Q3 guidance, which will be closely monitored to determine if the second-half capital expenditure commitments translate effectively into Nvidia’s revenue.