At the G20 meeting in North Carolina on September 3, 2026, NVIDIA CEO Jensen Huang urged global ministers to treat AI infrastructure as essential public works, comparable to water and electricity systems. The call, joined by OpenAI's Sam Altman, signals mounting pressure on governments to fund the data centers required to run next-generation AI models - or risk falling behind in economic growth tied to automation and machine learning.
Huang's remarks position NVIDIA's hardware as the backbone of that buildout. "AI is as fundamental as water and electricity," Huang said, advocating for nations to construct domestic infrastructure to spur local economic development. The company's GPUs already dominate the market for training large-scale AI models, a role that has pushed its market capitalization to roughly $5.42 trillion, making it the world's most valuable publicly traded firm.
The numbers behind the stock
Despite that dominance, NVIDIA's stock presents a mixed picture. GuruFocus calculates a GF Value of $382.77 per share against a current price of $224.41 - a 41.4% discount to intrinsic value. The trailing price-to-earnings ratio sits at 28.37, well below the five-year median P/E of 56.15. A GF Score of 95 out of 100 reflects top-tier profitability and growth metrics, though the valuation component lags behind other factors.
Insider activity tells a different story. Over the past three months, insiders sold $827,828,220 worth of shares. Among institutional investors tracked as premium gurus, 33 currently hold positions: 12 added shares while 20 trimmed their stakes in recent quarters. The pattern suggests caution even as the company's fundamentals remain strong.
What powers the demand
NVIDIA's core business spans gaming GPUs and data center accelerators, but the surge in interest comes from enterprise AI workloads. Training foundation models requires clusters of high-performance chips working in parallel - a workload where NVIDIA's architecture and CUDA software ecosystem hold a commanding lead. As governments and corporations scale up AI capabilities, the need for new data center capacity grows in lockstep, a dynamic that places NVIDIA's hardware at the center of national infrastructure conversations.
For developers and IT teams, this shift changes procurement and architecture decisions. The push for sovereign AI infrastructure means more on-premises and colocation deployments, not just cloud-based GPU rentals. Professionals working with AI for IT & Development will encounter pressure to specify hardware-accelerated environments earlier in project lifecycles, as organizations race to secure compute capacity before supply tightens further.
Why this matters for IT and development professionals
The G20-level attention on AI infrastructure means data center buildouts are likely to accelerate, creating demand for engineers who can design, deploy, and optimize GPU-accelerated systems. Skills in cluster networking, power-efficient cooling, and workload orchestration will move from niche to mainstream. Developers who understand how to right-size models for specific hardware - rather than treating GPU access as unlimited - will have an edge as organizations scrutinize compute costs more closely. For those looking to build that expertise, an AI Learning Path for Software Developers offers a structured way to move from application-layer AI use to understanding the infrastructure that makes it possible.
Your membership also unlocks:
