Nvidia Revenue Surges, Huang Renxun Declares AI Inflection Point
Nvidia’s fourth-quarter revenue surged 73% to $68.1 billion, exceeding analyst expectations and signaling continued strong demand for artificial intelligence technology, the company announced Wednesday, February 25, 2026. The results, which represent the period ending January 31, 2026, demonstrate the resilience of the AI boom despite recent investor anxieties fueled by product releases from Anthropic and a cautionary report from Citrini.
Net profit for the quarter reached $42.96 billion, a 94% increase year-over-year, with a gross margin of 75%, up 2 percentage points. For the full fiscal year 2026, Nvidia reported revenue of $215.9 billion, a 65% increase, and net profit of $120.07 billion, also up 65%.
“Computing demand is growing exponentially, and the inflection point for agentic AI is here,” said Nvidia CEO Huang Renxun in a statement accompanying the earnings release. He noted a rapid increase in enterprise adoption of AI agents. The company’s data center revenue led the growth, reaching $62.3 billion for the quarter, a 75% increase and a new historical high.
The strong performance prompted Huang Renxun to raise the company’s revenue outlook, now projecting $500 billion in revenue. Nvidia is forecasting revenue of $78 billion for the first quarter of fiscal year 2027, plus or minus 2%, excluding assumptions related to revenue from the China data center market. This guidance also surpassed analyst predictions, which anticipated revenue growth of 63% to 68% and $720 billion for the next quarter, according to LSEG data.
Even as data center revenue drove the overall growth, Nvidia’s gaming revenue also saw a significant increase, rising 47% to $3.7 billion, though it experienced a 13% decrease compared to the previous quarter. Professional visualization revenue increased 159% to $1.3 billion, and automotive and robotics revenue reached $6 billion.
Huang Renxun is scheduled to unveil “something the world has never seen” at the GTC 2026 conference on March 15 in Santa Clara, California. The announcement, made February 19, has generated considerable industry anticipation. The new chips are likely to come from either the Rubin series, which is already in full production, or the next-generation Feynman series, which is exploring wide integration with SRAM and potential 3D stacking technology.
Nvidia is adapting its chip development to meet shifting AI computing demands, moving from focusing on model pre-training with the Hopper and Blackwell series to inference scenarios with the Grace Blackwell Ultra and Vera Rubin series. The upcoming chips are expected to address bottlenecks in latency and memory bandwidth.