
NVIDIA is pursuing more than $750 billion(約120兆円) in new AI infrastructure deals globally, including a partnership with South Korean conglomerate SK Group expected to generate more than $500 billion(約80兆円) of business and potential financing arrangements with OpenAI worth up to $250 billion(約40兆円) in guarantees and $350 billion(約56兆円) in processor purchases. Investors are concerned that NVIDIA is increasingly financing its own major customers, creating circular demand and potential exposure if spending slows—a risk CEO Jensen Huang has downplayed, arguing NVIDIA's investments represent only a small portion of the money its customers ultimately need to raise.
Summaries like this, in your inbox every morning.
Sign up free →What happened
NVIDIA is pursuing more than $750 billion(約120兆円) of new and potential AI agreements globally. The company announced a partnership with SK Group expected to generate more than $500 billion(約80兆円) of business, including development of more than 2 gigawatts of AI data centers on the Korean Peninsula. NVIDIA is also considering a guarantee of as much as $250 billion(約40兆円) to help OpenAI lease a proposed $500 billion(約80兆円) data center hub in Ohio, and separately discussing financing approximately $350 billion(約56兆円) of OpenAI's purchases of NVIDIA processors.
Why it matters
These agreements may support further demand for NVIDIA's processors, but investors have raised concerns that the company is increasingly financing businesses and infrastructure projects that are also major customers of its chips—creating potential circular demand and exposure if spending slows or customers struggle to generate sufficient returns from their investments.
What to watch
The SK Group agreement includes development of an AI factory by SK Telecom expected to open next year. NVIDIA invested $1 billion(約1600億円) in Naver, a South Korean internet and cloud services provider, to support an AI data center being developed with Brookfield; Naver's shares increased more than 8% in Seoul following the announcement. Negotiations with OpenAI remain preliminary and could change or collapse.
NVIDIA, the dominant producer of artificial intelligence processors, is pursuing more than $750 billion(約120兆円) of new and potential AI agreements as it accelerates investments across the global computing ecosystem. The most substantial commitment is a partnership with SK Group, a South Korean conglomerate, expected to generate more than $500 billion(約80兆円) of business between the two companies. Under this agreement, the companies plan to develop more than 2 gigawatts of AI data centers on the Korean Peninsula, representing enough electricity to power roughly 1.5 million homes. SK Telecom, a South Korean telecommunications company within SK Group, will build the first AI factory, which is expected to open next year. NVIDIA will also collaborate with SK Hynix, SK Group's memory chip supplier, to develop future high-bandwidth memory products, potentially improving NVIDIA's access to components that remain in short supply due to expanding AI data center construction.
Separately, NVIDIA agreed to invest $1 billion(約1600億円) in Naver, a South Korean internet and cloud services provider, to support an AI data center being developed with Brookfield, a U.S. private equity firm. The investment is expected to allow Naver to more than triple the size of the facility, which will use NVIDIA's AI computing hardware. Following the announcement, Naver's shares increased more than 8% in Seoul.
Most controversially, NVIDIA is considering a guarantee of as much as $250 billion(約40兆円) to help OpenAI, the developer of ChatGPT, lease a proposed $500 billion(約80兆円), 10-gigawatt data center hub being developed in Ohio by a subsidiary of SoftBank Group, a Japanese technology investment company. The chipmaker is separately discussing financing approximately $350 billion(約56兆円) of OpenAI's purchases of NVIDIA processors for the project. However, these negotiations remain preliminary and could change or collapse. While such arrangements may signal strong long-term demand for AI infrastructure, investors have raised concerns that NVIDIA is increasingly financing businesses and infrastructure projects that are also major customers of its chips. This arrangement could increase NVIDIA's exposure if spending slows or customers struggle to generate sufficient returns from their investments. NVIDIA shares declined as much as 5.3% to $195.92 in New York trading on Monday, and the annual cost of protecting its debt against default rose by as much as approximately 0.14 percentage point to around 0.82 percentage point. CEO Jensen Huang has rejected claims that NVIDIA's agreements create circular demand, arguing that the company's investments represent only a small portion of the money its customers ultimately need to raise.
NVIDIA's pursuit of more than $750 billion(約120兆円) in new AI agreements reflects the company's strategy to accelerate investments across the global computing ecosystem while securing long-term demand for its processors. The SK Group partnership is particularly significant, committing both parties to develop more than 2 gigawatts of AI data centers on the Korean Peninsula—a scale equivalent to powering roughly 1.5 million homes. This partnership also includes a collaboration between NVIDIA and SK Hynix on high-bandwidth memory products, addressing a component shortage NVIDIA faces due to expanding AI data center construction and improving the chipmaker's access to critical supplies.
However, the scale and structure of these deals have triggered investor concern about circular demand dynamics. NVIDIA is not merely selling chips; it is guaranteeing or financing the infrastructure projects and processor purchases of its largest customers. The proposed OpenAI arrangement—combining a $250 billion(約40兆円) guarantee for a data center lease and $350 billion(約56兆円) in processor financing—exemplifies this pattern. Investors worry that if customers cannot generate sufficient returns from their AI investments or if spending growth slows, NVIDIA's exposure could mount. CEO Jensen Huang has countered this criticism by stating that NVIDIA's agreements represent only a small portion of the total capital its customers must raise, positioning the company's role as a partial enabler rather than the primary funding source.
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytime
No comments yet. Be the first to share your thoughts!
Log in to join the discussion





Get curated AI news from 200+ sources delivered daily to your inbox. Free to use.
Get Started FreeFree · takes 30 seconds · unsubscribe anytime