Jensen Huang Recently Delivered Incredible News for Nvidia Investors

Nvidia (NVDA -7.62%) supplies some of the world’s most advanced graphics processing units (GPUs) for data centers — hardware that developers use to power and train artificial intelligence (AI) software. Demand for its chips far exceeds what it can currently supply, which helps explain how the company has added over $2.3 trillion to its market capitalization since the start of 2023.

At Nvidia’s annual GPU Technology Conference (GTC) last month, CEO Jensen Huang laid out some incredible catalysts that could accelerate the company’s already rapid growth. With its stock currently trading down 27% from its record high amid the sharp sell-off in the broader market, this could be a significant buying opportunity.

New AI models need 100 times the computing power of their predecessors

Large language models (LLMs) sit at the foundation of every AI application. These models are trained on mountains of data, and the more data an LLM can access, the “smarter” the resulting tool will be. However, training them requires massive amounts of computing power — particularly parallel processing power — which is why there is so much demand for Nvidia’s data center GPUs.

Up until recently, LLMs delivered “one-shot” responses, meaning a chatbot would rapidly generate a single output for every prompt input by the user. While this method was fast and effective, it failed to weed out inaccuracies, which detracted from their value and the user experience. Now, top developers like OpenAI, Anthropic, and DeepSeek are focusing on an entirely different approach called test-time scaling, or “reasoning.”

Rather than simply ingesting endless amounts of data, these models spend more time “thinking” before rendering responses to inputs. In other words, they make better use of the data they already have, and are more apt to clear up any inaccuracies behind the scenes before releasing the final output. This approach has been wildly successful, producing some of the most advanced AI models to date, such as OpenAI’s GPT-4o series, DeepSeek’s R1, Anthropic’s Claude 3.7 Sonnet, and Alphabet‘s Gemini 2.5 Pro.

However, reasoning models require significantly more computing power. Huang says each response consumes 10 times more tokens (words, punctuation, and symbols) because of how much “thinking” goes on in the background, and as a result, the models are also much slower to render a final output. Huang says GPUs will need to be 10 times faster to offset this, and he estimates that developers will soon need a staggering 100 times more computing power to deploy reasoning models with a satisfactory user experience.

Nvidia’s new Blackwell GPU architecture is a step in that direction. In some configurations, a Blackwell GB200 GPU can perform AI inference 30 times faster than the company’s previous generation of chips, which were based on its Hopper architecture. Plus, last month, Nvidia revealed its new Blackwell Ultra architecture, which will be capable of delivering 50 times more performance than Hopper because it’s specifically designed for reasoning models.

A $1 trillion annual opportunity by 2028

The continuing shift toward reasoning models could be a significant tailwind for Nvidia’s GPU sales. At GTC last month, Huang said the top four providers of cloud infrastructure services (and thus, the world’s largest operators of data centers) have ordered a whopping 3.6 million Blackwell GPUs already, which is almost triple the number of Hopper chips they purchased last year.

Those four cloud providers are Amazon Web Services, Microsoft Azure, Google Cloud, and Oracle Cloud Infrastructure. That list doesn’t include other big spenders that are developing AI for their own purposes, like Meta Platforms, Tesla, and OpenAI, so the total number of Blackwell orders is almost certainly much higher.

This could just be the beginning: Huang predicts AI infrastructure spending will top $1 trillion annually by 2028, and much of that will go toward AI accelerator chips such as those that Nvidia provides.

Nvidia’s data center business generated $115.2 billion in revenue during its fiscal 2025 (which ended Jan. 26). That was up 142% compared to the prior year. If Huang’s forecast is right, the company’s sales likely have substantial room to grow.

Nvidia stock looks like a bargain right now

The 27% drop in Nvidia stock from its recent all-time high has created an opportunity for investors to buy it at an attractive valuation relative to its history. It currently trades at a price-to-earnings (P/E) ratio of 36.9. That’s its cheapest level in three years, and also a 38% discount to its 10-year average P/E ratio of 59.5.

Data by YCharts.

Moreover, Wall Street’s consensus estimates (as provided by Yahoo! Finance) suggest that Nvidia’s earnings per share (EPS) for fiscal 2026 will come in at $4.53. That gives the stock a forward P/E ratio of just 23.9. In other words, Nvidia would have to soar by 149% by the end of this fiscal year just to trade in line with its 10-year average P/E ratio of 59.5 (assuming Wall Street’s EPS estimate proves to be accurate).

With that said, I think investors should look beyond the next 12 months because, if Huang is correct, Nvidia shareholders’ best returns might be realized over the next three to five years instead.