In a recent dinner meeting that sounds like something out of a Silicon Valley power play, Oracle CTO Larry Ellison, along with Tesla and SpaceX CEO Elon Musk, were reportedly pleading with NVIDIA CEO Jensen Huang for more GPUs. Ellison described the scene, saying, “I went to dinner with Elon Musk, Jensen Huang, and I would describe the dinner as Oracle and me and Elon begging Jensen for GPUs. Please take our money… no no, take more of it. You’re not taking enough. We need you to take more of our money, please.”
This desperate plea highlights the massive demand for NVIDIA’s GPUs, particularly in the burgeoning field of artificial intelligence. Oracle, known for its database software, is investing heavily in building an AI supercluster – an ambitious project that will require over 130,000 of NVIDIA’s latest Blackwell AI GPUs and an estimated 2000+ data centers. To power this supercluster, Oracle estimates needing the equivalent of three nuclear power plants, showcasing the sheer scale of their ambitions.
The story paints a picture of NVIDIA’s dominant position in the AI hardware market. Everyone wants their GPUs – from tech giants like Oracle and Microsoft to research organizations like OpenAI and even individuals looking to build their own AI projects. With this high demand, NVIDIA has its hands full, relying on TSMC (Taiwan Semiconductor Manufacturing Company) to produce as many chips as possible. This shortage has driven up the price of NVIDIA GPUs, making them a precious commodity in the AI world.
It remains to be seen if NVIDIA can meet the escalating demand for its AI hardware. But one thing is clear: the company is at the heart of the AI revolution, with everyone from Oracle to Elon Musk vying for its powerful GPUs to power their ambitious AI projects.
