• Jensen Huang predicts AI infrastructure will drive multi-trillion-dollar investment.
  • Nvidia's dominant GPU position fuels record revenue and strong margins.
  • Analysts see sustained demand despite potential policy and supply-chain headwinds.

AI's Economic Engine

Nvidia CEO Jensen Huang told CNBC that artificial intelligence will become a multitrillion-dollar industry, reinforcing the chipmaker's central role in the AI boom. “The buildout of AI infrastructure is just getting started,” he said, citing demand from hyperscalers, cloud providers, and enterprise customers. Nvidia’s latest earnings reflect that surge: revenue more than tripled year-over-year, with gross margins in the high-70% range, propelled by sales of its H100 and upcoming Blackwell GPUs.

A person familiar with Nvidia’s supply chain noted that “orders remain robust through 2025,” despite occasional product cycle dips. The company’s CUDA software ecosystem continues to lock in developers, creating a competitive moat that rivals AMD and Intel have struggled to match.

Policy and Market Realities

While Huang’s optimism is broad, export controls on advanced chips to China remain a constraint. Nvidia has developed compliant variants, but lost sales in that market could trim growth. “Geopolitical risks are baked into the outlook,” said a semiconductor analyst who asked not to be named. “But the global AI capex cycle is powerful enough to offset near-term headwinds.”

Economists at Goldman Sachs recently estimated that AI-related investment could contribute 0.5 percentage points to US GDP growth annually through 2027. Nvidia, as the primary hardware supplier, is poised to capture the lion’s share of that spending.

The Long View

Huang’s multitrillion-dollar forecast builds on a trajectory of accelerating compute demand. “We’re moving from AI experimentation to full-scale deployment,” he said. Nvidia’s roadmap includes next-generation chips, networking gear, and software for robotics and autonomous vehicles. “Every industry will be reinvented by AI,” Huang added.

Meanwhile, competition is intensifying. Microsoft, Google, and Amazon are designing their own AI chips, and startups like Groq are challenging Nvidia’s inference dominance. But for now, Nvidia’s integrated hardware-software stack remains the gold standard, and Huang’s vision of a trillion-dollar AI economy is increasingly seen as plausible.

Correction: An earlier version of this article misstated the timing of Nvidia’s Blackwell GPU launch. It is expected in late 2024.