Overview

• Both companies announced a strategic agreement to co‑deploy at least 10 GW of AI data‑center capacity on Nvidia hardware, with finer terms to be finalized in the coming weeks.
• Nvidia committed up to $100 billion in staged investments tied to deployment milestones, and reporting indicates the first tranche would follow completion of the initial 1‑GW buildout.
• The first phase targets the second half of 2026 on Nvidia’s Rubin platform, and Jensen Huang said 10 GW equates to roughly 4–5 million GPUs.
• Nvidia told clients the OpenAI partnership will not change its strategic priorities or chip supply to other customers, even as antitrust experts warn the tie‑up could confer unfair advantages.
• Sam Altman said OpenAI will roll out compute‑intensive products in the coming weeks, with some features initially limited to Pro subscribers or subject to additional fees.