Nvidia says chip supply can meet strong AI demand as growth widens

Nvidia Chief Executive Jensen Huang said in Taipei that the company has enough supply to support robust growth in both CPUs and GPUs, a message that pushed the AI boom’s next bottleneck into sharper focus: not whether Nvidia can build chips, but whether customers can absorb them.

The timing mattered. Huang’s remarks came during Computex week, when Taiwan’s semiconductor supply chain is on display and investors are looking for signs that AI demand still justifies heavy capital spending. His comments did not announce a new forecast or a revised financial target. Instead, they reinforced the idea that the hardware cycle remains broad enough for Nvidia to keep shipping across multiple product lines, even as the industry continues to wrestle with packaging, manufacturing and deployment constraints.

That shift matters because the market has spent months worrying about whether AI expansion will outrun chip production, whether data centers can be built fast enough and whether customers can finance the next wave of systems. If Nvidia says supply is no longer the binding constraint, the pressure moves elsewhere: power availability, data-center build-outs, procurement budgets and the ability of cloud operators and enterprises to install systems on time. Nvidia’s position as a barometer for the AI market means investors are likely to read Huang’s confidence as a sign that demand remains strong enough to absorb more capacity.

Nvidia also used the Taipei trip to widen its footprint beyond data-center accelerators. On June 1, the company unveiled the RTX Spark superchip for PCs, with an initial plan to release more than 30 laptops and 10 desktops over time. Nvidia said the chip will debut later in 2026 on Windows PCs from Microsoft, Dell, HP, ASUS, Lenovo and MSI. The company also said the Vera CPU is in full production, with early adoption from OpenAI, Anthropic and SpaceX, underscoring a strategy that now spans both general-purpose computing and AI infrastructure.

Huang has tied that expansion closely to Taiwan. He said Nvidia plans to invest around $150 billion a year in the island, and he previously described Taiwan as the “epicentre” of the AI revolution. A February Taipei Times report said he called Taiwan’s supply chain a “record year” in 2025, naming Foxconn, TSMC and Wistron, and predicting an “incredible year” in 2026. The commercial stakes were visible in the market reaction: CNBC reported that Nvidia’s PC-chip push sent shares of AMD, Intel and Qualcomm lower.

For now, Huang’s message is clear. Nvidia says it can keep up. The harder question is whether the rest of the AI economy can keep pace with the chips.