Investing in 'AI Computing Power' Like Oil or Electricity... GPU Futures to Be Launched
CME Group to Launch Two Types of Futures Contracts in October
New Futures to Track NVIDIA Chip Rental Prices
Will the Financial Market for AI Infrastructure Ignite?
As early as October, a new era may begin in which computing power, the core of artificial intelligence (AI) technology, can be bought and sold like oil or electricity. Thanks to the AI boom, the first-ever financial products based on the market value of graphic processing units (GPUs) are set to be launched.
On August 11 (local time), the U.S. financial media outlet CNBC reported that CME Group, the world’s largest derivatives exchange, plans to join forces with GPU market information provider SiliconData to introduce two types of futures contracts based on AI computing power prices on October 5. The contracts are expected to begin trading after receiving regulatory approval.
NVIDIA logo, a global semiconductor company. Photo by Reuters and Yonhap News Agency
View original imageThe new futures contracts will track the rental prices of NVIDIA’s AI semiconductor chips: the H100 and the latest Blackwell B200. The contracts will be traded based on an hourly GPU rental price index calculated by SiliconData, representing the cost to lease a GPU for one month.
Pete Kieby, Global Head of Energy and Environmental Products at CME, said, “Computing has become the currency of the AI era,” and added, “Just as oil, which powered the 20th-century economy, evolved from spot trading to a global derivatives market, we will standardize computing power and make it into a tradable commodity.”
GPU ‘Public Price Table’ Emerges
Since the generative AI boom, GPUs have become a core resource over which companies fiercely compete. However, unlike commodities such as oil or electricity, there has been no common reference price in the market. Even for the same GPU computing power, the price paid by companies could vary widely depending on the provider or the timing of the contract.
Carmon Lee, CEO of SiliconData, explained, “For years, two companies could buy exactly the same GPU capacity yet pay completely different prices, with no way to verify who got the better deal.” He continued, “GPU futures will provide a public, tradable benchmark price that the market has lacked until now for resources used in all AI systems.”
Once the futures market opens, AI developers or data center operators will be able to lock in costs or revenues in advance, protecting themselves against potential spikes in GPU rental fees. For example, if an AI company anticipates a significant increase in computing power prices, it can use futures contracts to hedge against the risk of rising prices.
"A Catalyst for Expanding the AI Infrastructure Financial Ecosystem"
Jensen Huang, CEO of NVIDIA, is delivering the keynote speech at Computex 2026 held in Taipei, Taiwan on June 1. Photo by AFP Yonhap News
View original imageCNBC assessed that this product launch could catalyze the expansion of the AI infrastructure financial ecosystem. So far, investors hoping to ride the AI investment wave have mostly bought shares of semiconductor companies such as NVIDIA or invested in data centers and related businesses. With GPU futures, investors no longer need to directly own semiconductors or data centers.
Some experts suggest that if AI computing power is established as an asset class, the price of these futures could become a new indicator of investment enthusiasm in the AI industry. If expectations rise for increased AI model development and data center investment, GPU rental prices are likely to rise in the future as well.
Hot Picks Today
"Two Spend 500,000 Won in a Flash, Yet Wait Two Hours?" ... 'Dreamland' Shunned by Kids and Foreign Tourists
- The Era of Getting 200 Million Won After 30 Years Is Over... A Japanese Bank's "Perpetual Department Head" Who Just Clocked In Cries Out
- Spain Hit by Its Worst Tragedy... Lee Kangin, Moreno and Real Madrid Join Soccer World in Mourning
- They Said, “Make It Your Home for Life” - Then Suddenly Kicked 200 Passengers Off... What Happened to the Around-the-World Cruise?
- "130 Rice Bowls Sold Out in a Day"... 'Unfiltered Face' Adds 8 Million Won to Night Market Sales: China's Bungeunwoo
Meanwhile, this futures launch has come as NVIDIA announced its strategy to link AI infrastructure with the financial market. On August 10, Jensen Huang, CEO of NVIDIA, announced a plan to establish a $500 billion (approximately 710 trillion won) AI infrastructure financial platform with six Wall Street institutions, including Goldman Sachs and BlackRock. This move focuses on the significant financing burden companies face from the massive initial investment—ranging from billions to tens of billions of dollars—required to build large-scale data centers. The market expects a range of new derivatives such as project financing (PF), loans, and asset-backed securities (ABS) to be developed.
© The Asia Business Daily. All rights reserved. Unauthorized AI training and use prohibited.