- Nvidia will be worth $10 trillion by 2030 as its next-generation Blackwell GPU drives massive revenue gains.
- That's according to I/O Fund tech analyst Beth Kindig, who said Nvidia has a massive moat around its business.
Nvidia stock will surge 258% from current levels and hit a $10 trillion market valuation by 2030, according to I/O Fund tech analyst Beth Kindig.
Such a gain would come after Nvidia stock has more than doubled so far this year and more than tripled in 2023 thanks to the fast-growing adoption of its H100 AI chips.
Nvidia has a $2.8 trillion valuation and is the third-largest company in the world.
According to Kindig, Nvidia's next-generation Blackwell GPU chip will drive another leg of massive growth for the chip maker, along with its CUDA software platform and its exposure to the automotive market.
Kindig estimates that the Blackwell GPU will surpass its predecessor, the H100, and generate data center revenue of $200 billion by the end of Nvidia's fiscal year 2026.
Blackwell "will empower and enable trillion-plus large language models, which is exactly where big tech is trying to go. Those components altogether equal a very large hardware data center segment, then we have software coming... and then the third one is automotive, so we have a lot coming. This is very, very early for Nvidia," Kindig told CNBC on Tuesday.
Kindig's long-term bullishness on Nvidia stems from the idea that the company has an "impenetrable moat" around its GPU business.
Kindig said that estimates for the AI data center market's total addressable market rising to $400 billion by 2027 and $1 trillion by 2030 will largely be captured by Nvidia, not its biggest competitors, AMD or Intel.
"Nvidia will take the lion's share of that," Kindig said, and that's largely because of Nvidia's software offerings it is integrating into its hardware products.
"The CUDA software platform is what developers learn on. So, similar to iOS is really what locked people into the iPhone because developers were developing applications for the iPhone. The same thing is happening with Nvidia, which is that the CUDA platform is what AI engineers are learning in order to program GPUs, so that helps lock them in, and that combination, right now, I'm calling it an impenetrable moat," Kindig explained.
Finally, the custom silicon AI chips being developed internally at mega-cap tech companies like Amazon and Alphabet will never directly compete with Nvidia, according to Kindig.
"They're not going to commercialize and sell chips the way that Nvidia does, so Nvidia has an open runway there," Kindig said.
Business Insider reported last week that Amazon has struggled in its attempt to make AI GPU chips that can compete with Nvidia's over the past four years.