Nvidia CEO Jensen Huang remains optimistic about the company’s future, asserting that the emerging DeepSeek technology will not affect sales during the recent earnings call. Concerns arose last month when speculation suggested that DeepSeek’s R1 model required significantly fewer chips for training, leading to a notable drop in Nvidia’s stock price.
However, Huang praised the R1 model as an “excellent innovation,” highlighting that it, along with other reasoning models, is beneficial for Nvidia because it requires much more computational power. He noted that reasoning models can demand up to 100 times more compute resources, and future models are expected to require even more.
Huang emphasized that the DeepSeek R1 has sparked global excitement and has made a world-class reasoning AI model accessible through open-source, with nearly every AI developer now utilizing R1.
Nvidia’s financial performance continues to impress, reporting a record revenue of $39.3 billion for the last quarter, surpassing both its own forecasts and Wall Street expectations. The company anticipates that revenue will increase again in the next quarter, projecting around $43 billion.
Sales in Nvidia’s data center segment nearly doubled in 2024, reaching $115 billion and showing a 16% increase from the previous quarter. During the earnings call, Huang highlighted the latest Blackwell chip, designed specifically for reasoning tasks, noting that demand for it is “extraordinary.”
Looking ahead, Huang stated confidently, “We will grow strongly in 2025.” Despite the recent concerns regarding DeepSeek, the AI chip market remains robust. Major tech companies like Meta, Google, and Amazon have announced substantial investments in AI infrastructure, collectively committing hundreds of billions of dollars for the upcoming years.
