Nvidia, a leader in AI semiconductors, reported record earnings for the second quarter of fiscal year 2027, exceeding market expectations. The surge in demand for AI infrastructure significantly boosted data center revenue.
According to reports from the Wall Street Journal and others, Nvidia announced after the market closed on August 26 that its Q2 revenue reached $96.22 billion (approximately 133.3 trillion won), a 106% increase from $46.7 billion in the same period last year.
During this period, earnings per share (EPS) were reported at $2.22, surpassing market forecasts. Prior estimates from market research firm LSEG projected revenue at $92.18 billion and EPS at $2.10.
The data center segment was the primary driver of this performance, with revenue soaring 117% year-over-year to $89 billion.
Notably, revenue from hyperscale customers, including Google, Microsoft, and Amazon, rose 102% year-over-year to $48.7 billion, accounting for about 55% of total data center revenue. Revenue from AI cloud and industrial customers reached $40.3 billion.
Nvidia described the demand for AI infrastructure as "remarkably large," noting that the growth rate of some AI companies is outpacing their own financial capabilities and long-term credit capacity. The company is entering contracts to assist some customers in securing the land, power, and data center capacity needed for AI expansion.
Looking ahead, Nvidia's revenue outlook also exceeded market expectations. The company projects Q3 revenue to reach $108 billion, surpassing the market estimate of approximately $104 billion to $105 billion. However, the gross margin for Q3 is expected to be 74%, slightly lower than the 75% reported in Q2, raising concerns about potential profitability declines due to rising costs.
Nvidia's long-term growth outlook remains optimistic. Management anticipates revenue for fiscal year 2028 to increase by about 70%, significantly above the previous market estimate of around 45%, indicating confidence in sustained investment in AI data centers.
CEO Jensen Huang emphasized that "AI has reached an inflection point" and is now performing genuinely useful tasks. He added that AI-related technologies are productive and profitable, stating, "Computing is revenue."
However, uncertainty looms over Nvidia's business in China. The company did not include revenue from Chinese data centers in its Q3 projections. CFO Colette Kress noted that the profitability of older Hopper chip products shipped to China is declining. Additionally, selling semiconductors to China requires Nvidia to pay 25% of its revenue to the U.S. government, which poses a further burden.
* This article has been translated by AI.
Copyright ⓒ Aju Press All rights reserved.