Story
Nvidia Affirms AI Roadmap and Broadening Demand in Investor Update
Summary
Nvidia has reaffirmed its AI product roadmap is on schedule and reports that demand is expanding beyond hyperscalers, according to an analyst note. The chipmaker also clarified its outlook on future energy efficiency and maintained its gross margin guidance.
Nvidia has confirmed to investors that its long-term AI infrastructure roadmap remains on track and that demand for its technology is expanding beyond its core hyperscale cloud customers. The update, detailed in a Citi analyst note following discussions with Nvidia's investor relations team, addresses key questions regarding the company's product pipeline, market dynamics, and financial strategy.
Product Roadmap Intact
Nvidia stated that its Rubin Ultra product roadmap is proceeding as planned, with no changes to the NVLink domain architecture unveiled at Computex. The company emphasized that its networking technology, a critical component of its platform, is seeing strong momentum.
- Co-packaged optics (CPO) for scale-out networking are already in production with its Spectrum-X platform, experiencing high customer adoption.
- Beginning with the Feynman platform in calendar 2028, customers will have the option to choose between NVLink implementations using either CPO or traditional copper interconnects.
Broadening Demand and Market Dynamics
While hyperscale cloud providers have been the primary drivers of AI infrastructure spending, Nvidia reports that demand has gained significant momentum over the past two years from other segments. The company expects AI laboratories, sovereign customers, and enterprise on-premise deployments to represent a larger share of the market going forward.
AdNvidia also addressed its role in the AI model ecosystem, stating that its in-house models like Nemotron and Cosmos are designed to support enterprise and government adoption, not to compete with leading frontier-model developers. The company declined to comment on the specific expansion plans of Meta Platforms but reiterated that overall AI demand remains "very strong."
Financial Outlook and Efficiency Gains
Nvidia maintained its guidance for gross margins in the mid-70% range and reiterated its target of returning 50% of cash flow to shareholders this year. A recent $25 billion debt offering was intended to provide greater financial flexibility, the company said.
Management also clarified recent comments from CEO Jensen Huang regarding energy efficiency. The projection that $100 billion of compute could eventually require just one gigawatt of power reflects expected efficiency improvements in future products, not current infrastructure costs. Nvidia estimates its current systems generate roughly $30 billion to $40 billion of revenue per gigawatt, noting that newer generations like Blackwell are significantly more power-efficient than the preceding Hopper architecture.