Nvidia CEO Points to $1 Trillion Build Out That's Bullish for Several Holdings

In recent years, when Nvidia NVDA CEO Jensen Huang delivers his much-expected keynote event at the GTC conference, it has been met with widespread enthusiasm by the stock market. This time around, the initial reaction to Huang’s GTC 2025 keynote was less enthusiastic, as evidenced by the fall off in the shares.

Some of that was likely due to the market but also because folks have become accustomed to Jensen delivering big at these events. However, we are seeing NVDA shares recover some of Tuesday's lost ground as folks digest what was shared and consider the implications.

Our view is that we remain in the early innings of AI adoption with Big Tech currently capacity-constrained. As that adoption accelerates, data center and network capacity will once again become constrained given the ensuing data consumption explosion and related power consumption. With that in mind, Huang shared that by the end of the decade, and potentially sooner, data center build-out should top $1 trillion. That keeps us bullish on the shares of Nvidia as well as Marvell MRVL and Eaton ETN.

Jensen’s Keynote

Much like Apple’s AAPL WWDC event, Nvidia’s Graphics Technology Conference (GTC) is aimed more at developers than investors, but the company showed it continues to focus on being a leader in the AI space.

With that in mind, Huang unveiled Nvidia's next two generations of chips, including Blackwell Ultra and Rubin. Blackwell Ultra, which has more memory and twice the bandwidth of Blackwell, will be coming in the second half of 2025, while Rubin, which has HBM-4 memory, will be coming in the second half of 2026. Huang went on to tease that Rubin Ultra will be available in the second half of 2027.

In terms of the nearer term, Huang shared that the top four cloud service providers — Amazon AMZN Web Services, Microsoft’s MSFT Azure, Google GOOGL Cloud Platform and Oracle ORCL Cloud — have bought 3.6 million Blackwell GPUs. Remember, they only started shipping in late 2024, and for comparison they purchased 1.3 million Hopper GPUs last year. While that is a big increase, we have to remember the dramatic upswing in AI/data center capital spending is a known-known coming off the December quarter earnings season. So, from that vantage point, those comments from Jensen were ones of confirmation rather than breaking new ground.

In Tuesday’s Daily Rundown, we shared that we wanted Huang to unveil new partner wins that would speak to continued AI adoption across a widening array of industries. He did not disappoint:

Nvidia would team with Cisco CSCO, T-Mobile TMUS, and Cerberus ODC to build full-stack radio networks in the U.S.

GM GM has selected Nvidia to help it build its self-driving autonomous fleet, Huang added.

NVIDIA, Alphabet and Google announced new initiatives to advance AI, democratize access to AI tools, speed the development of physical AI and transform industries including healthcare, manufacturing and energy.

Yum! Brands YUM announced it is partnering with NVIDIA to accelerate the development of AI technologies for Yum! restaurants.

Oracle announced a first-of-its-kind integration between Nvidia-accelerated computing and inference software with Oracle's AI infrastructure and generative AI services to help organizations globally speed the creation of agentic AI applications.

Dell Technologies DELL announced new AI PCs, infrastructure, software, and services advancements to accelerate enterprise AI innovation at any scale.

Several other partnerships were announced, including ones from IBM IBM, Micron MU, GE Healthcare GEHC, Super Micro Computer SMCI and ServiceNow NOW.

More Pro Portfolio:

At the time of publication, TheStreet Pro Portfolio was long NVDA, MRVL, ETN, AAPL, AMZN, MSFT, GOOGL and NOW.