Jensen Huang: Sell-off in software company stocks fails to account for AI’s long-term impact

When someone whose company is worth trillions of dollars thanks to an artificial intelligence boom tells the market "you're wrong"—it's worth listening. Jensen Huang, the longtime head of Nvidia, publicly stated that the current sell-off of software company stocks does not account for the fundamental long-term advantages that AI will bring. According to him, the largest technology providers are already working to adapt artificial intelligence for specific improvements to their clients' businesses.

The last few months have been challenging for the technology sector. After several years of unbridled optimism around AI, investors began asking uncomfortable questions: where is the real return on multi-billion-dollar investments in infrastructure? Shares of a whole range of software companies, which previously soared on the wave of AI hype, have declined. The market, it seems, has grown tired of waiting and has begun taking profits. Skepticism was mounting against reports showing that AI implementation in the corporate sector is progressing slower than analysts had predicted, and the return on investment remains murky for many companies.

It was at this moment that Huang decided to offer a counter-argument. His thesis is simple and at the same time ambitious: the current correction is not a verdict, but a manifestation of market myopia. The head of Nvidia is convinced that the largest cloud and technology providers—those very hyperscalers who are buying Nvidia GPUs by the tens of thousands—are on the threshold of a qualitative transition. They are ceasing to simply sell computing power and are beginning to offer clients ready-made AI solutions adapted to specific business tasks. This is a fundamentally different monetization model, and it is this, according to Huang, that will determine the next phase of the AI revolution.

Technically, this is about a transition from horizontal infrastructure to vertical solutions. If previously Amazon, Microsoft, and Google sold raw computing power through their cloud platforms, now they are increasingly creating specialized AI tools for specific industries—from healthcare and finance to logistics and manufacturing. This means that every dollar invested in GPUs and data centers will generate significantly more revenue than simple computing rental. For Nvidia, this is, of course, an ideal scenario: the deeper AI penetrates into business processes, the greater the demand for its chips.

However, it would be naive to view Huang's words as objective market analysis. He is an interested party—perhaps the most interested one in the entire AI ecosystem. Every percent drop in software company stocks potentially threatens Nvidia as well: if the market decides that AI doesn't justify expectations, hardware manufacturers will be next under fire. Therefore, his statement is not only analytics but also strategic communication aimed at maintaining the narrative of AI transformation's inevitability.

Nevertheless, there is a rational kernel in his argument. The history of technological revolutions shows that there is always a time lag between the emergence of a breakthrough technology and its mass commercialization, which markets systematically underestimate. The internet went through the dot-com bubble, but companies that survived that period became trillion-dollar giants. Cloud computing seemed like a niche product for years before becoming the foundation of modern IT infrastructure. AI is probably going through a similar phase—a period of disillusionment after initial euphoria, followed by steady growth based on real business cases.

The key question is not whether AI will bring long-term benefits—few doubt that. The question is how accurately current valuations reflect the timing and scale of those benefits. Huang is essentially telling the market: you're seeing the direction right, but you're misjudging the speed. The largest providers are already building bridges between raw AI power and concrete business results. And when these bridges are completed, the current sell-off will look like a missed opportunity.

For the industry as a whole, Huang's statement is a signal that the focus is shifting. The era of "AI for AI's sake" is coming to an end. The time is coming when technology will have to prove itself not with presentations and benchmarks, but with lines in clients' financial reports. And it is precisely those companies that are the first to learn how to turn AI's capabilities into measurable business value that will determine the landscape of the technology industry for the next decade.