Today’s breakthrough has the potential to reshape the path of artificial intelligence development, cloud computing, and semiconductor markets. NVIDIA ($NVDA), which trades on the NASDAQ in the Semiconductor & Semiconductor Equipment sector, has unveiled its plan to invest up to $100 billion in a strategic partnership with OpenAI. This isn’t just about money—it’s about constructing at least 10 gigawatts (GW) of NVIDIA systems to power OpenAI’s next-generation AI infrastructure.
The deal, formalized through a letter of intent, aims to deploy these systems in phases, with the first phase coming online in the second half of 2026, using NVIDIA’s Vera Rubin platform. This milestone moves NVIDIA further into the driver’s seat of global AI infrastructure, a domain previously dominated by cloud providers. Investors noticed immediately: NVIDIA’s stock jumped, reflecting growing confidence that this move could deepen its moat in both hardware and AI compute services.
But the implications go well beyond NVIDIA. Oracle Corporation ($ORCL – traded on NYSE, in the Technology – Cloud Computing / Software sector) and SoftBank Group ($SFTBY – OTC, Telecommunications & Technology sector) are already heavily involved in the broader Stargate AI infrastructure initiative, valued at $500 billion over four years. Microsoft ($MSFT – NASDAQ, Technology sector) likewise stands to be affected, given its deep ties to cloud infrastructure and AI model deployment.
Why 10 GW? Because the scale of compute required for cutting-edge AI models is exploding—both for training and inference. OpenAI’s push to have millions of GPUs powered by NVIDIA systems means there’ll be enormous demand not only for chips, but for data centers, power supply, cooling, networking, and energy sources. All these are expensive and complex to scale.
Yet with excitement come questions. Is the return on investment going to be commensurate with the risks? Deploying 10 GW of AI systems means grappling with energy costs, potential regulatory constraints, supply chain bottlenecks, and the possibility of overbuilding. Some analysts worry that expectations may already be baked into current stock prices for NVIDIA, Oracle, and Microsoft. If the technical or logistic challenges delay rollout, the excitement could wane. Still, many believe that this is a turning point, not just for NVIDIA or OpenAI, but for the whole semiconductor + cloud infrastructure ecosystem. It may accelerate the race for sovereign AI infrastructure around the world.
All eyes will be on how this unfolds in the next year: what happens in late 2025 through 2026 around financing, deployment, and whether the Vera Rubin platform delivers as promised. Stakeholders from governments, energy sectors, and AI companies will be watching closely—this isn’t just a corporate strategy story, it’s increasingly a story about national infrastructure, competitiveness, and who leads the next era of AI.