Elon Musk & Larry Ellison 'begged' Jensen Huang to allot them NVIDIA GPUs at private dinner

The meeting underscored the high demand for NVIDIA’s GPUs, which dominate the AI chip market with an 80 per cent share. These chips are essential for training AI models and powering supercomputers, making them a highly coveted resource in the rapidly growing AI industry

featured-image

In a recent meeting with analysts, Oracle cofounder and tech billionaire Larry Ellison revealed that he and Elon Musk, CEO of Tesla and SpaceX, had a private dinner with NVIDIA CEO Jensen Huang where they desperately requested more GPUs. Ellison humorously described the meeting at Nobu Palo Alto, saying that he and Musk were essentially begging Huang to allocate them more of NVIDIA’s sought-after graphics processing units (GPUs), which are critical for artificial intelligence (AI) applications. Ellison recounted the dinner with a touch of levity, stating, “Please take our money.

Please take our money. By the way, I got dinner. No, no, take more of it.



” Despite the light-hearted tone, the meeting underscored the high demand for NVIDIA’s GPUs, which dominate the AI chip market with an 80 per cent share. These chips are essential for training AI models and powering supercomputers, making them a highly coveted resource in the rapidly growing AI industry. Ellison, whose personal wealth is estimated at $206 billion, is no stranger to spotting significant technological shifts.

As Oracle’s executive chairman, he has consistently positioned the company to capitalise on emerging technologies. Oracle, a $475 billion enterprise, has made significant investments in cloud-based infrastructure and AI, maintaining a strong partnership with NVIDIA. This relationship has helped Oracle stay at the forefront of AI advancements, as the company continues to build out its GPU infrastructure for large-scale AI applications.

At the analysts’ meeting, Ellison also reflected on the intense competition in the AI space, comparing it to a Formula One race where many compete but only one can be the ultimate winner. Oracle is heavily investing in GPU technology, with plans to house acres of NVIDIA GPU clusters in its global data centres. The company reported strong financial results in the first quarter of fiscal 2025, with revenues up 7 per cent to $13.

3 billion and profits reaching $2.9 billion. Oracle’s growing cloud infrastructure now includes 162 data centres worldwide, and the company signed $3 billion worth of GPU cloud contracts in the first quarter alone.

Tesla, under Musk’s leadership, is similarly reliant on NVIDIA GPUs to power its neural networks for autonomous driving technology. The race to build the world’s most advanced neural networks has intensified, with Ellison noting that many AI executives are scrambling for access to NVIDIA’s cutting-edge chips. Ellison also highlighted the immense cost of developing frontier AI models, estimating it at $100 billion over the next three years.

He pointed out that only a limited number of companies, countries, and individuals will have the resources to participate in this AI arms race, further emphasising the critical role NVIDIA plays in this high-stakes competition..