Artificial intelligence is no longer just a software story. In recent days, NVIDIA CEO Jensen Huang has argued that AI is triggering the biggest buildout of physical infrastructure in human history, a project he says will require trillions of dollars and reshape labor markets from chip fabs to plumbing. His pitch is simple but sweeping: AI is becoming a foundational utility, and the world is only at the start of constructing the hardware, power, and data center backbone it needs.
Huang’s comments, delivered around the World Economic Forum in Davos and in follow-on interviews, frame AI as a multi-layered “computing platform” that will sit beneath the global economy. In his telling, the stakes are not just about who trains the best model, but which countries and companies can finance, build, and operate the vast physical stack that makes those models possible.
The “largest infrastructure buildout in human history”
Jensen Huang has started describing the AI surge as the “largest infrastructure buildout in human history,” a phrase that signals how far he believes the technology has moved beyond experimental labs. At Davos, NVIDIA CEO Jensen Huang cast AI as a generational construction wave that will span data centers, chip factories, and power systems, arguing that this buildout will touch almost every part of the industrial economy. In his view, AI is not a niche upgrade to existing IT, but a new layer of infrastructure that rivals the historical rollout of railways, highways, and the internet, with NVIDIA CEO Jensen Huang positioning it as a once-in-a-century transformation.
That framing has been amplified by NVIDIA itself, which has highlighted AI as the foundation for this unprecedented expansion of physical assets. In a recent post, NVIDIA emphasized that AI is becoming the base of the “largest infrastructure buildout in human history,” underscoring how the company sees its chips and systems at the center of a global wave of data center and network construction. The message from NVIDIA Data Center is clear: AI infrastructure is not a side project, it is the new backbone around which digital and physical industries are being reorganized.
Trillions of dollars and a five-layer “AI cake”
To justify calling AI the biggest infrastructure project ever, Huang points to the sheer scale of investment he believes is required. He has said that “trillions of dollars of AI infrastructure needs to be built,” arguing that the current wave of spending is only the beginning of a much longer cycle. In one interview, Nvidia CEO Jensen Huang stressed that the world is still in the early innings of deploying specialized chips, systems, and data centers, and that the capital flowing into these assets will be measured in the trillions as enterprises and governments race to modernize. Technology Editor Daniel Howley has highlighted how Huang links this spending to a broader shift toward “AI native” companies that are built around these capabilities from the ground up.
Huang has also tried to make this abstract capital cycle more concrete by describing AI as a “five-layer cake.” At the bottom is energy and physical plant, followed by data centers and cloud infrastructure, then AI models, and finally the applications and services that sit on top. He has argued that the real economic payoff will come from that top layer of software, but only if the lower layers are built out at massive scale and with careful coordination. In Davos conversations, he walked through how this stack is emerging around the world, explaining that “this layer on top ultimately is where economic benefit will happen,” while warning that the platform only works if the underlying hardware and power systems are robust. That layered view of AI, laid out in detail in a conversation with Jensen Huang, has been echoed in analysis of “The Five Layer Cake That Changes Everything,” which describes how energy, cloud infrastructure, models, and applications must align for AI to deliver on its promise.
From Davos to Washington: AI as national infrastructure
Huang’s argument is not just about corporate balance sheets, it is also about national strategy. Speaking from the World Economic Forum in Davos, he has framed AI infrastructure as a competitive asset that countries must build if they want to lead in the next phase of the digital economy. In one televised discussion, he said the infrastructure around artificial intelligence will need trillions of dollars in additional investment, describing how data centers, networks, and power grids are being scaled up around the world to support this new computing platform. In that exchange, Nvidia CEO Jensen tied the AI buildout directly to the World Economic Forum’s annual meeting in Davos, Switzerland, underscoring how central the topic has become to global economic planning.
In the United States, Huang has explicitly linked this infrastructure wave to policy choices in Washington. In a recent interview, he said the AI boom is fueling the “largest” infrastructure buildout in history and noted that “President [Donald] Trump wants to make sure America is the place that we do it,” casting federal support as a catalyst for domestic data center and chip factory projects. Speaking with FOX Business’ Maria Bartiromo from the World Economic Forum, he argued that the United States economy is uniquely positioned to benefit from this surge if it can mobilize capital and regulatory support quickly enough. That message dovetails with his broader claim that AI infrastructure is now a matter of national competitiveness, not just corporate strategy.
Six-figure jobs and the new blue-collar AI economy
One of the most striking parts of Huang’s pitch is his insistence that the AI buildout will be a jobs engine, particularly for skilled trades. He has said that the trillions of dollars in investment for AI will boost wages of construction workers involved in the projects, predicting that many of these roles will reach “six-figure” pay as demand for data centers, chip plants, and power facilities accelerates. In a discussion about the labor impact, Huang argued that the United States economy is the “best place in the world” to see a significant boom in this area, pointing to the scale of planned projects and the need for electricians, steelworkers, and other trades. Those comments were highlighted in coverage of how Huang said that the trillions in AI spending would flow into construction and the applications that sit on top, and in a separate report noting that, during his discussion with Fink, the Nvidia CEO went into detail about the different levels of AI infrastructure and how they rely on skilled labor.
Huang has gone further, arguing that this is “a good time to be a plumber” and that workers do not need computer science degrees to benefit from the AI wave. He has described how data centers require extensive plumbing, cooling, and electrical work, and how those projects will create “a lot of jobs” as companies race to expand capacity. In one account, Huang described the AI boom as the largest infrastructure buildout in human history that would create a surge in construction and maintenance roles, emphasizing that people do not need to study computer science to participate. Another report on the AI infrastructure buildout noted that, during his discussion with Fink, the Nvidia CEO explained how the different levels of AI infrastructure depend on tradespeople, reinforcing his message that the AI economy will be built as much by workers in hard hats as by coders. That perspective was captured in coverage that, during his remarks with Fink, the Nvidia CEO outlined how the US economy could see a significant boom in this area.
More from Morning Overview