Bold claim: Nvidia hasn’t closed its $100 billion OpenAI deal yet, despite high expectations and a market-ready collab that could reshape AI infrastructure. And this is the part most people miss: the path from a non-binding letter of intent to a definitive agreement can be lengthy and nuanced, affecting timelines, pricing, and control over future deployments.
Here’s the core situation in plain terms. Nvidia’s chief financial officer, Colette Kress, told the UBS Global Technology and AI Conference that the proposed investment in OpenAI remains unsettled. A framework exists, including a plan to deploy at least 10 gigawatts of Nvidia systems for OpenAI—a capacity sufficient to power millions of homes in the United States—but no final contract has been signed. The companies are actively negotiating, but a definitive agreement is still out.
Why this matters: OpenAI is a cornerstone customer for Nvidia’s chips, alongside other large cloud providers. The potential deal would lock in substantial chip demand for years, contributing to Nvidia’s already large order book, which CEO Jensen Huang has described as worth about $500 billion for advanced chips through 2026. Importantly, the chips OpenAI would receive after a finalized deal are not included in those bookings yet, meaning the final number could rise when the agreement is complete.
Market impact and controversy: Nvidia stock rose about 2.6% as the news circulated, signaling investor sensitivity to deal progress and the broader AI funding environment. In the past year, Nvidia has pursued multiple partnerships with AI startups and invested in firms that also count as major customers. This practice—often termed “circular deals”—has drawn scrutiny about potential overleveraging and over-concentration in a fast-growing ecosystem.
OpenAI’s status as a central player in the generative AI boom further elevates the stakes. Nvidia’s willingness to commit up to $100 billion—and in other moves, up to $10 billion to Anthropic—reflects a strategy of deep integration with leading AI developers. Yet the reality of finalizing these agreements remains a hurdle, potentially affecting pricing, deployment timelines, and competitive dynamics among chipmakers and cloud providers.
Bottom line: The definitive agreement is still in progress. While the collaboration looks strategically compelling—promising large-scale Nvidia deployments for OpenAI—the official, binding terms have yet to be sealed. This means investors and industry watchers should monitor for further updates on the contractual details, milestones, and any shifts in chip-availability commitments as negotiations continue.
Would you like a concise explainer outlining what a final open-ended technology deal typically includes (goals, milestones, pricing, governance, and compliance), plus potential scenarios if the agreement stalls or accelerates?
