According to DeepLearning.AI, OpenAI, Oracle, and SoftBank have unveiled an ambitious $1 trillion plan to expand AI data center capacity through five new U.S. sites and a major Stargate UK facility. The Stargate roadmap outlines a target of 20 gigawatts of AI computing power globally, with future projections reaching up to 100 gigawatts. Within the next 18 months, Ohio and Texas will receive 1.5 gigawatts of this capacity. Oracle is set to manage construction, while Nvidia will supply up to 31,000 GPUs to power these data centers. In addition, Nvidia has pledged a landmark $100 billion investment in OpenAI, signaling a significant acceleration in generative AI infrastructure and positioning these companies at the forefront of supercomputing for AI applications (source: DeepLearning.AI via Twitter, The Batch).

According to Greg Brockman (@gdb) on Twitter, OpenAI has announced a strategic partnership with NVIDIA to secure millions of GPUs—equivalent to the total compute NVIDIA is projected to ship in 2025. This collaboration aims to significantly expand OpenAI's AI infrastructure, enabling the rapid scaling and deployment of advanced AI models and services. The deal underscores growing enterprise demand for high-performance AI hardware and reinforces NVIDIA's leadership in the AI chip market. As cited in the official OpenAI announcement, this partnership is set to accelerate innovation in generative AI, large language models, and next-generation applications, offering substantial business opportunities for cloud providers, AI startups, and enterprise adopters (source: OpenAI, https://openai.com/index/openai-nvidia-systems-partnership/).