Jensen Huang: The Software Stock Selloff Ignores AI's Long-Term Effect

When the head of a company worth trillions of dollars thanks to the AI boom tells the market "you're wrong," it is worth listening. Jensen Huang, Nvidia's longtime chief, has said publicly that the current selloff of software stocks fails to price in the fundamental long-term benefits AI will bring. According to him, the biggest technology providers are already working to adapt artificial intelligence to concretely improve their clients' businesses.

The past few months have been difficult for the tech sector. After several years of unbridled AI optimism, investors have started asking uncomfortable questions: where is the real return on multibillion-dollar infrastructure spending? Shares of a range of software companies that had previously soared on the AI hype have turned down. The market seems to have grown tired of waiting and begun taking profits. Skepticism built as reports showed that enterprise AI adoption is moving more slowly than analysts forecast, and return on investment remains murky for many companies.

That is the moment Huang chose to offer a counterargument. His thesis is simple and ambitious at once: the current correction is not a verdict but a symptom of market myopia. The Nvidia chief believes the largest cloud and technology providers, the very hyperscalers that buy Nvidia GPUs by the tens of thousands, are on the verge of a qualitative shift. They are ceasing to simply sell compute and starting to offer clients ready-made AI solutions tailored to specific business tasks. That is a fundamentally different monetization model, and in Huang's view it will define the next phase of the AI revolution.

Technically, this is a move from horizontal infrastructure to vertical solutions. Where Amazon, Microsoft and Google once sold raw computing power through their cloud platforms, they are now increasingly building specialized AI tools for specific industries, from healthcare and finance to logistics and manufacturing. That means every dollar invested in GPUs and data centers would generate considerably more revenue than simple compute rental. For Nvidia, of course, this is the ideal scenario: the deeper AI penetrates business processes, the greater the demand for its chips.

Still, it would be naive to take Huang's words as objective market analysis. He is an interested party, arguably the most interested in the entire AI ecosystem. Every percent that software stocks fall potentially threatens Nvidia too: if the market decides AI is not living up to expectations, hardware makers will be next in line. His remarks are therefore not only analysis but strategic communication, aimed at sustaining the narrative of AI transformation as inevitable.

Nonetheless, there is a rational kernel in his argument. The history of technological revolutions shows that a time lag always separates a breakthrough technology from its mass commercialization, and markets systematically underestimate it. The internet went through the dot-com bubble, but the companies that survived that period became trillion-dollar giants. Cloud computing seemed 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 built 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 see the direction correctly, but you misjudge the speed. The largest providers are already building bridges between raw AI power and concrete business results. And when those bridges are finished, today's selloff will look like a missed opportunity.

For the industry as a whole, Huang's statement signals a shift in focus. The era of "AI for AI's sake" is coming to an end. The time is arriving when the technology will have to prove itself not with presentations and benchmarks, but with line items in clients' financial reports. And the companies that first learn to turn AI capabilities into measurable business value will shape the technology industry's landscape for the next decade.