Overview

• Both companies signed a letter of intent to co‑deploy at least 10 gigawatts of Nvidia AI systems for OpenAI, with investment and buildout progressing over time and the first phase targeted for the second half of 2026.
• OpenAI will treat Nvidia as a preferred provider for chips and networking gear as the partners align hardware and software roadmaps for next‑generation models.
• Nvidia CEO Jensen Huang said the 10‑gigawatt plan equates to roughly 4 million to 5 million GPUs, highlighting the project’s unprecedented scale.
• Reporting indicates a possible two‑part structure in which OpenAI pays cash for systems while Nvidia takes a non‑controlling equity stake, with an initial $10 billion contingent on a definitive purchase agreement.
• Analysts said the partnership could invite antitrust scrutiny, and the companies noted it complements OpenAI’s ongoing work with Microsoft, Oracle, SoftBank and the Stargate initiative.