Is Nvidia Stock Still a Buy After Its Incredible 1,100% Run? Here's the Honest Answer.

Nvidia (NASDAQ: NVDA), the beating heart of the artificial intelligence (AI) boom, has seen its shares gain more than 1,100% since ChatGPT-3 was released to the public at the end of 2022.

If you're wondering whether the stock still has room to grow, here's what you need to know.

Image source: Getty Images.

What Nvidia actually does

At the most basic level, Nvidia designs graphics processing units (GPUs). Originally designed for rendering video game graphics, these chips are extraordinarily good at a more lucrative task: running the math that makes AI work.

Training an AI model is, at its core, a massive math problem. You need to run billions of calculations simultaneously, and that's exactly what GPUs were designed for. While traditional chips handle tasks one at a time in sequence, a GPU processes thousands of operations at once, in parallel. That is just what AI needs.

The company's vision meant that it was years ahead when the current AI boom really took off, and its competition has been struggling to catch up ever since. Nvidia posted $215.9 billion in revenue for its fiscal year 2026 (ended Jan. 25), up 65% from the year before -- and a roughly ninefold increase from $27.0 billion only three years prior. You can see the parabolic growth below.

NVDA Revenue (TTM) data by YCharts

Data center revenue alone -- the segment driven almost entirely by AI demand -- hit $62.3 billion in Q4, growing 75% year over year. Gross margins are sitting above 75%. These are staggering numbers for any company, let alone a hardware company.

The real moat: How CUDA locks in customers

Now, Nvidia has real technical leadership -- its chips are the most powerful on the market -- but that isn't the essence of its moat.

Nvidia's Compute Unified Device Architecture (CUDA) is the software platform that actually enables GPUs to do more than render graphics -- like train and run AI models. Nvidia invested in creating CUDA back in the early 2000s, long before the current AI boom. And it has become the de facto standard throughout the AI industry.

Why does that matter? Because CUDA only works with Nvidia's hardware. It creates lock-in.

Millions of developers know CUDA inside and out, and major AI software like PyTorch and TensorFlow are optimized for it.

That means when a business or research lab builds its AI infrastructure, it's likely building on CUDA, and once you're built on CUDA, moving to a competitor's chip is difficult and expensive.

That developer lock-in is what Jensen Huang, Nvidia's CEO, has called the company's most important asset. Competitors like AMD and Intel have capable hardware, and big customers like Amazon and Alphabet's Google are designing their own AI chips. But none of them can replicate the software ecosystem that took Nvidia 20 years and billions of dollars to build.

The bear case you need to take seriously

This most certainly does not mean that Nvidia is a risk-free investment. But in my view, its future performance has much less to do with the competition and much more to do with the market.

The biggest risk is simple: AI might not deliver the promised returns.

The scale of spending on AI infrastructure is staggering -- unprecedented -- on the order of trillions. Just a handful of companies spent more on AI infrastructure last year than the U.S. government spent over 13 years on the entire Apollo program (adjusted for inflation, of course).

But despite the incredible scale of investment and the rapid and nearly universal buy-in from executives across the economy, AI has relatively little to show for it. The returns just aren't materializing -- in productivity or dollars. Numerous studies have shown minimal effects on worker productivity. That won't do. The investment is too large for AI not to be game changing.

And the revenues that the leading model providers, like OpenAI and Anthropic, are collecting pale in comparison to the cash moving out the door to pay for compute -- like Nvidia's chips.

Of course, many will argue this will soon change, that the issue is not the technology, but its implementation within organizations. And this is a fair argument. It has certainly happened before.

At the end of the day, Nvidia's fate comes down to who is proven right.

Is Nvidia stock still a buy?

Nvidia is a rare company, one that dominates both the hardware and the software layer of what could be the most significant technology shift in a generation. Its competitive position, anchored by CUDA, is stronger than any other chip company's by a long shot.

For investors who believe AI is what it is promised to be, Nvidia is still a strong buy. It's not going to be unseated by the competition anytime soon.

But Nvidia would be one of the stocks hit hardest in a bubble scenario, so for investors skeptical of AI's actual pay-off, I would advise limiting exposure.

Johnny Rice has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Advanced Micro Devices, Alphabet, Amazon, Intel, and Nvidia. The Motley Fool has a disclosure policy.