As the AI industry enters the stage of large-scale commercial deployment, the market’s focus on chip companies is shifting from pure high-speed growth to growth sustainability and profitability. Against this backdrop, NVIDIA, with its technological leadership, ecosystem barriers, and scale effects, continues to be regarded as one of the most competitive players in the AI infrastructure space.
Although major cloud providers are actively seeking to diversify their AI chip supply, Wall Street analysts believe that NVIDIA’s (NVDA) dominant position in the AI data center market will remain difficult to challenge in the short term, and its high profit margins are expected to last through the end of the decade.
Gil Luria, head of technology research at DA Davidson, stated that due to the lack of sufficiently mature alternatives in the AI chip space among hyperscale cloud providers, NVIDIA still holds extremely strong pricing power. Luria pointed out that these large cloud providers actually have few options and remain almost entirely dependent on NVIDIA’s chips, making it highly sustainable for NVIDIA to maintain gross margins above 70%.
In recent years, driven by the rapid development of generative AI, major tech companies such as Microsoft (MSFT), Amazon (AMZN), Google (GOOG, GOOGL), and Meta (META) have continued to expand their AI data center investments, making NVIDIA the biggest beneficiary of the current wave of AI infrastructure build-out. Although these tech giants are attempting to reduce their reliance on NVIDIA by purchasing custom AI chips from Broadcom (AVGO) and AI accelerators from AMD (AMD), the market generally believes that these alternatives are still in the early stages of development. Luria believes that neither AMD nor Broadcom has yet truly formed a competitive advantage capable of challenging NVIDIA, and that competitors remain in a very early phase of development.
According to financial data, NVIDIA’s latest quarterly revenue increased by 85% year-over-year to $81.6 billion, with adjusted gross margins reaching 75%, far higher than most semiconductor companies. Luria maintains a “Buy” rating on NVIDIA with a price target of $300, implying about 37% upside from Thursday’s closing price.
However, as AI investment scales continue to expand, some investors have become more cautious about the short-term growth prospects of large chip companies. Despite NVIDIA’s stock price having risen more than 1,300% over the past five years, and its latest quarterly results overall beating market expectations, the company’s stock still pulled back after the earnings release. This reflects that market expectations for AI leaders have reached extremely high levels—simply exceeding expectations is no longer enough to sustain further stock price increases. A similar situation has occurred with Broadcom. On Thursday, Broadcom’s stock recorded its biggest one-day drop in more than 16 months, as its AI chip revenue outlook failed to meet investors’ rising expectations. Luria believes the market reaction was somewhat excessive, stating that Broadcom delivered a very solid performance with revenue still growing strongly, but investors have been trained to expect even higher growth rates.
