You do not want to give Jensen Huang a 16-year head start

Nvidia's incredible moat and how it's like AWS in the early days

AWS’ 7-year head start

In a 2018 interview with David Rubenstein from the Carlyle Group (YouTube), Jeff Bezos remarked:

AWS completely reinvented the way companies buy computation. Then a business miracle happened. This never happens. This is the greatest piece of business luck in the history of business as far as I know. We faced no like-minded competition for seven years. It’s unbelievable. When you pioneer if you’re lucky you get a two year head start. Nobody gets a seven year head start. We had this incredible runway.

In 2017, Warren Buffett commented, “You do not want to give Jeff Bezos a seven-year head start.”

Nvidia’s 16-year head start with AI

Nvidia introduced CUDA, its parallel computing platform and API model for GPUs, way back in 2007. This turned out to be a hugely consequential move, giving Nvidia a multi-year head start over alternatives like the nonprofit OpenCL (released in 2009).

In hindsight, Nvidia's foresight to release CUDA when GPU computing was still nascent was a masterstroke. The company invested heavily in making CUDA accessible and performant at a time when others were asleep at the wheel.

Fast forward to today, and it's clear just how big of an advantage Nvidia has built up. Some key areas where CUDA is far ahead of competitors:

Maturity and optimization - Nvidia has had over a decade to optimize and tighten integration between Nvidia’s GPU hardware and CUDA’s software stack. Everything from compilers to math libraries are tailored to run blazingly fast on Nvidia chips. CUDA software helps program “GPUs for new tasks, turning them from single-purpose chips to more general-purpose ones that could take on other jobs in fields like physics and chemical simulations.”

Tooling and ecosystem - Through CUDA, Nvidia provides developers great tools like debuggers, profilers, libraries and documentation. The ecosystem around CUDA for things like machine learning is rich and vibrant. It’s CUDA, coupled with Nvidia’s hardware, that helps give Nvidia a lead over its competitors.

Performance - There's a reason CUDA GPUs are the go-to for high performance computing and AI workloads. Nvidia's close coupling of hardware and software results in better real-world performance.

Nvidia’s biggest threats, Intel and AMD, are still years off from being able to put any significant pressure on the graphics chipmaker.

Developer mindshare - Thanks to its head start and Nvidia’s heavy investment in AI, CUDA has become synonymous with GPU programming for many developers. Communities like Stack Overflow have way more CUDA expertise than OpenCL.

"Customers will wait 18 months to buy an Nvidia system rather than buy an available, off-the-shelf chip from either a start-up or another competitor," said Daniel Newman, an analyst at Futurum Group. "It’s incredible."

Business strategy - Improving and investing in CUDA allows Nvidia to capture developer mindshare and sell more GPUs. It’s in their business interest in a way that vendor-neutral APIs like OpenCL can never be.

“No one even comes close”

Nvidia’s loyal customers and developers include DeepMind co-founder and head of applied AI, Mustafa Suleyman. He’s now co-founder of Inflection AI, which raised $1.3b at a $4b valuation in its most recent funding round.

Suleyman “said that there was no obligation to use Nvidia’s products but that competitors offered no viable alternative. ‘None of them come close,’ he said.”

Nvidia reaped tremendous benefits by betting on CUDA early. Its first-mover advantage has compounded over the years. Even as competitors play catch up, CUDA remains the gold standard for developing GPU-accelerated software. The company built an enduring platform that will continue to pay dividends for years to come.

Nvidia's competitors can't “compete with a company that sells computers, software, cloud services and trained A.I. models, as well as processors.”

"I believe the hardest part of the equation is software, a moat Nvidia has built with CUDA and its libraries and workflows." - Patrick Moorhead, CEO of Moor Insights and Strategy

How Nvidia’s moat with AI is like AWS

Nvidia today has the same advantage AWS has had in cloud computing for years because it's benefited from the same flywheel:

The more developers choose Nvidia, the more and the better CUDA libraries and workflows are for working with GPUs, the more companies will base their generative AI strategies around using Nvidia, and the more other developers will choose to learn CUDA over other frameworks.

Just as developers built libraries, workflows, and expertise around using AWS, they have done the same with CUDA over the past decade.

If anything, Nvidia’s moat is even deeper than AWS’, whose lead in the cloud has gradually been chipped away by Microsoft and Google.

Suffice it to say you do not want to give Jensen Huang a 16-year head start on building the critical tools for the most important industry of our generation.

---

Is NVDA still undervalued?

It's hard to say whether or not Nvidia is currently undervalued, but it has a phenomenal business with an incredible moat in a fast growing, increasingly important sector.

Disclaimer: this is not a recommendation to buy NVDA, only a comment on the quality of its business (not valuation)

If you’ve read so far, you might as well subscribe. It’s free and always will be.