NVIDIA and OpenAI have announced a letter of intent for a strategic partnership focused on building next-generation artificial intelligence infrastructure. The collaboration involves deploying at least 10 gigawatts of NVIDIA systems, backed by a potential NVIDIA investment of up to $100 billion in OpenAI.
Key Takeaways
- NVIDIA and OpenAI plan to deploy a minimum of 10 gigawatts of AI computing infrastructure.
- NVIDIA intends to invest up to $100 billion in OpenAI, tied to the deployment milestones.
- The first 1-gigawatt phase is scheduled for the second half of 2026, using NVIDIA's Vera Rubin platform.
- The partnership aims to accelerate the development of OpenAI's advanced AI models, including superintelligence.
A Landmark Infrastructure Agreement
In a significant move for the artificial intelligence sector, NVIDIA and OpenAI revealed plans on September 22, 2025, to dramatically expand AI computing capacity. The agreement outlines the creation of a massive AI infrastructure powered by NVIDIA's hardware.
The core of the partnership is the deployment of at least 10 gigawatts of NVIDIA systems. This substantial amount of computing power is designed to support the training and operation of OpenAI's future AI models. The initiative represents one of the largest planned investments in AI compute to date.
To facilitate this large-scale build-out, which includes data centers and power resources, NVIDIA has stated its intention to invest up to $100 billion into OpenAI. This investment is structured to be released progressively as each gigawatt of computing power is brought online.
"This investment and infrastructure partnership mark the next leap forward—deploying 10 gigawatts to power the next era of intelligence," said Jensen Huang, founder and CEO of NVIDIA.
Project Timeline and Technology
The first phase of this ambitious project is set to begin relatively soon. The companies are targeting the second half of 2026 to launch the initial one gigawatt of computing systems.
This first deployment will be built on NVIDIA’s upcoming Vera Rubin platform, signaling that the partnership will utilize the latest in GPU and networking technology. The collaboration will also involve close coordination between the two companies to align their future roadmaps.
By the Numbers
- 10 Gigawatts: The minimum planned computing power deployment.
- $100 Billion: The maximum potential investment from NVIDIA into OpenAI.
- 2026: The target year for the first 1-gigawatt phase to go online.
- 700 Million+: OpenAI's reported number of weekly active users.
According to the announcement, OpenAI and NVIDIA will work to co-optimize OpenAI's model and software infrastructure with NVIDIA's hardware and software stack. This deep integration is expected to maximize performance and efficiency for training extremely large AI models.
Strategic Goals and Industry Impact
Both companies view this partnership as a foundational step toward achieving artificial general intelligence (AGI). Sam Altman, CEO of OpenAI, emphasized the central role of computing in the future economy.
"Compute infrastructure will be the basis for the economy of the future, and we will utilize what we’re building with NVIDIA to both create new AI breakthroughs and empower people and businesses with them at scale," Altman stated.
The collaboration builds on a long-standing relationship between the two technology leaders. Greg Brockman, President of OpenAI, noted their history of working together since the company's early days.
Expanding an Existing Ecosystem
This new agreement complements OpenAI's existing partnerships with other major technology firms. The company already collaborates with Microsoft, Oracle, SoftBank, and Stargate partners to build out its global AI infrastructure. The NVIDIA deal significantly expands its dedicated compute resources.
OpenAI has seen rapid growth, with its tools now used by over 700 million people weekly and widely adopted by businesses of all sizes. This new infrastructure is intended to support that expanding user base and enable the development of more powerful and beneficial AI systems.
Future of the Partnership
The letter of intent marks the formal beginning of this new phase in the NVIDIA-OpenAI relationship. The companies will now work to finalize the specific details of the strategic partnership in the coming weeks.
As a preferred strategic partner, NVIDIA's systems will be central to OpenAI's growth plans for its "AI factories." This term reflects the shift toward treating large-scale computing clusters as industrial-level production facilities for intelligence.
The scale of the planned deployment underscores the immense energy and hardware requirements needed to push the boundaries of current AI capabilities. The 10-gigawatt target represents a power consumption level comparable to that of a small country, highlighting the resources being marshaled in the race to develop superintelligence.