NVIDIA is close to finalizing a $30 billion equity investment in OpenAI, which effectively replaces or scales back from a previously announced larger commitment involving up to $100 billion.
Background on the Original Deal
In September 2025, NVIDIA and OpenAI announced a strategic partnership; framed as a memorandum of understanding or letter of intent to deploy at least 10 gigawatts of NVIDIA-powered AI infrastructure involving millions of GPUs, starting with the Vera Rubin platform in late 2026.
Register for Tekedia Mini-MBA edition 19 (Feb 9 – May 2, 2026).
Register for Tekedia AI in Business Masterclass.
Join Tekedia Capital Syndicate and co-invest in great global startups.
Register for Tekedia AI Lab.
As part of this, NVIDIA committed to invest up to $100 billion progressively to support OpenAI’s purchases and deployment of its chips in data centers. This was tied to milestones, with an initial tranche planned once definitive agreements were reached.
However, that multiyear investment framework proved more complex and took longer than expected to finalize, and it appears to have stalled or been abandoned in its original form.
The New Arrangement
NVIDIA is now shifting to a direct $30 billion equity stake in OpenAI (no milestones or deployment contingencies attached). This forms part of OpenAI’s massive ongoing funding round, which aims to raise more than $100 billion overall and values the company at around $730 billion pre-money or up to ~$830 billion in some reports, post-money estimates vary.
OpenAI plans to reinvest much of the new capital into NVIDIA hardware anyway, so the partnership for GPU supply remains strong, but the structure changes from a conditional infrastructure investment to straight equity ownership.
This move allows NVIDIA led by Jensen Huang to gain ownership in a key customer, while reducing the capital commitment and risk compared to the original $100 billion plan. The shift was first prominently reported by the Financial Times.
NVIDIA has declined to comment directly, but sources describe it as a strategic pivot toward taking a stakeholder position rather than just an “arms dealer” supplying chips. This reflects the intense capital needs of frontier AI development—OpenAI’s valuation and fundraising scale are staggering—and ongoing close ties between the two companies despite the scaled-back investment vehicle.
The shift from NVIDIA’s originally announced up to $100 billion multi-year infrastructure investment commitment to a streamlined $30 billion direct equity stake in OpenAI carries several significant implications across strategy, finance, market dynamics, and the broader AI ecosystem.
The original $100B plan involved conditional, milestone-tied funding for massive data center builds. This exposed NVIDIA to execution risks like delays, cost overruns, power constraints, and regulatory hurdles in physical infrastructure.
The new $30B equity investment is a straightforward cash-for-shares deal with no such contingencies, reducing balance-sheet strain and direct operational exposure while still securing a meaningful ownership position in a top customer. NVIDIA gains equity in one of the highest-valued private AI companies ~$730B pre-money / up to ~$830B post-money in the ongoing round, potentially benefiting from OpenAI’s future growth, valuation appreciation, or eventual liquidity events.
At the same time, OpenAI plans to reinvest much of the fresh capital into NVIDIA hardware anyway—maintaining strong GPU demand and reinforcing NVIDIA’s “arms dealer + stakeholder” role in AI. Bullish long-term signal for investors: Analysts view this as strategically positive for NVIDIA stock, trimming risk while locking in a gigantic customer relationship.
It addresses earlier market concerns from January 2026 reports of the deal being “on ice” that had pressured AI/tech stocks. Short-term volatility is possible, but the pivot underscores confidence in sustained AI capex without overcommitting capital.
The $30B from NVIDIA forms a key part of a blockbuster fundraising round targeting > $100B total, providing runway for frontier model training, compute scaling, talent, and operations. Unlike the milestone-dependent original structure, this equity comes without deployment hurdles, accelerating OpenAI’s roadmap under Sam Altman.
OpenAI remains a “gigantic customer” for NVIDIA GPUs (potentially millions more units), ensuring access to cutting-edge hardware amid competition from AMD, custom chips from hyperscalers, or in-house efforts. At ~$730-830B, this round cements OpenAI’s status as a mega-cap private entity, signaling investor belief in its path to AGI-level capabilities despite enormous cash burn.
The deal exemplifies tightening ties among AI leaders—chip suppliers (NVIDIA), model developers (OpenAI), cloud providers (e.g., Microsoft, Amazon), and investors (SoftBank, others). This creates a more integrated (some say “circular”) financing loop: investors fund OpenAI ? OpenAI buys NVIDIA chips ? NVIDIA profits and reinvests ? repeat.
Critics highlight bubble risks if growth expectations falter or if returns don’t materialize. Even the scaled-back version shows frontier AI requires staggering funding. OpenAI’s round potentially >$100B and ongoing chip purchases underscore that infrastructure spend isn’t slowing—it’s evolving in structure for efficiency.
The pivot eases earlier overhangs that weighed on AI stocks. It may stabilize or boost sentiment around NVIDIA and the sector, while highlighting how megadeals are being restructured amid complexity. This isn’t a retreat from the AI boom—it’s a pragmatic recalibration.
NVIDIA secures strategic ownership and demand with less risk; OpenAI gets faster, cleaner capital to fuel its ambitions; and the ecosystem demonstrates resilience in adapting massive commitments to real-world execution challenges. The partnership remains rock-solid, just in a more shareholder-friendly form.



