Jensen Huang argues for a shift in how Wall Street prices “compute” for AI at Nvidia

Nvidia Chief Executive Jensen Huang is pushing Wall Street to rethink how it values AI “compute,” arguing that the market still treats the technology stack like interchangeable infrastructure rather than an evolving, monetizable capability, according to a Yahoo Finance report published Friday.

The remarks come as Nvidia’s growth has depended on demand for its data center graphics processing units (GPUs) and the surrounding software and platform that help enterprises build and deploy AI workloads. Nvidia’s long-running theme is that AI is not only a systems engineering problem but also a business platform, with compute serving as the core economic input that turns models into usable applications.

In the interview referenced by Yahoo Finance, Huang urged investors to consider compute as a “commodity” in the sense that buying and deploying it will be widespread and recurring, but he also suggested that the economic value should not be reduced to a narrow view of hardware margins. The message is that investors should look beyond the idea of compute as a cost center and instead focus on how organizations pay for the ability to train, fine-tune, and run AI systems.

That framing is particularly relevant for how markets compare cloud spending and enterprise IT budgets. If compute becomes easier to procure, the argument goes, companies will normalize the purchase of AI capacity the way they do for electricity or standardized services. In that world, the competitive advantage and revenue potential shift toward platform integration, performance-per-dollar, and the ability to accelerate end-to-end AI workflows.

Nvidia has sought to formalize that platform approach through its broader ecosystem, which typically includes accelerated computing hardware, software libraries, and tools aimed at reducing the time and expertise required to bring models into production. Huang’s “compute-as-commodity” pitch, as described by Yahoo Finance, appears aimed at aligning capital market expectations with that ecosystem strategy, not just with GPU units sold.

Still, the public reporting referenced here does not provide granular details on how Huang expects pricing or earnings to change, nor does it lay out a specific valuation framework for analysts. The company also did not, in the materials available for this story, disclose any new guidance, financial targets, or changes to its product roadmap connected directly to the comments.

Going forward, investors will likely watch whether Nvidia’s next earnings period shows stronger linkage between its platform narrative and reported results, including indicates on software and services contribution, customer adoption of standardized AI deployment workflows, and the durability of demand across training and inference workloads. If the “commodity compute” view spreads among Wall Street analysts, Nvidia’s share price sensitivity could shift toward evidence of higher-value usage patterns rather than only unit growth.