Nvidia CEO Jensen Huang told G20 innovation ministers that building a single gigawatt of AI computing infrastructure would cost between $50 billion and $60 billion. The Chapel Hill gathering produced a consensus statement on AI policy but no binding funding commitments, while Nvidia itself remains dependent on outside manufacturers to build the chips behind that infrastructure.
Jensen Huang told officials gathered at the G20 Innovation Ministerial in Chapel Hill, North Carolina that building a single gigawatt of AI computing infrastructure would run between $50 billion and $60 billion. One gigawatt is roughly the output of a large nuclear plant, and Huang's message to the ministers was that compute capacity now belongs in the same category as roads, water systems, and the electrical grid.
The Chapel Hill meeting produced consensus, not funding
The G20 Innovation Ministerial, held September 1-2, 2026, was co-hosted by the US Department of Commerce and the White House Office of Science and Technology Policy, bringing together innovation ministers from G20 member states alongside tech executives including OpenAI's Sam Altman and Anthropic's Tom Brown. The group produced two documents, the Carolina Principles for Emerging Technologies and the AI Prosperity Objectives Compact, built around four pillars: research investment, commercialization, trusted adoption, and public-private partnerships.
However, a consensus statement emerged without any formal funding commitments or specific enforcement mechanisms. That stance is expected to carry through to the G20 leaders' summit scheduled for Miami in December 2026.
Huang frames infrastructure as durable, fungible, and upgradeable
Huang's pitch went beyond a general case for AI's importance. He argued the infrastructure needs to be durable, fungible, and software-upgradeable: built to survive multiple generations of AI models rather than becoming obsolete every 18 months, flexible enough to run different workloads, and improvable through software rather than full hardware replacement.
Nvidia still depends on outside chipmakers
Nvidia's own production still runs through other companies. As a fabless chipmaker, Nvidia outsources manufacturing to Taiwan Semiconductor Manufacturing and buys memory chips from Micron Technology and Samsung, according to the Motley Fool. Those chips depend in turn on ASML's extreme ultraviolet lithography machines, which print the circuits needed for the 7-nanometer and smaller processes that Nvidia's latest designs require.
Sources: Crypto Briefing, Motley Fool
Trading involves risk.
Risk Disclaimer
