According to McKinsey's latest research report, by 2030, AI related infrastructure investment alone will reach as high as 5.2 trillion US dollars. This is not a small number - to help you understand the scale, 5.2 trillion US dollars is equivalent to: 1.2 billion working hours (equivalent to 6 million people working full-time for an entire year) 3 million miles of fiber optic cable (enough to wrap around the Earth 120 times) 150 to 200 gigawatts of electricity (enough to supply 150 million American households with electricity for one year) At the same time, the entire GPU market is also moving towards a scale of over 1 trillion US dollars. These numbers tell us that computing power has become one of the most important infrastructures for humanity. But there is a big blind spot here, which is also the core issue I wanted to emphasize yesterday. The traditional commodity market has long solved the problem of price fluctuations. Starbucks can hedge the price of coffee beans to ensure that their large latte doesn't go up too much. McDonald's can hedge the price of potatoes and lock in the cost of large potatoes. Airlines can hedge against crude oil (WTI) to protect themselves from the impact of fuel price fluctuations. Gold, silver, copper - almost all important commodities have mature futures markets. However, these GPU manufacturers and data centers holding thousands of H100 and H200 cards do not have an effective way to short or hedge their computing assets. What problems will this cause? Super recommendation code http://app.lighter.xyz/?referral=MRBLOCK If the price of computing power collapses, they will face huge risks. They are unable to lock in future income, hedge supply chain risks, and manage risks like traditional commodity producers. And this problem will become increasingly serious with the expansion of AI investment scale. Why is this happening? Because GPUs are not standardized like traditional products. A barrel of crude oil is a barrel of crude oil, and a bushel of wheat is a bushel of wheat. But GPU? NVIDIA's H100 alone has over 50 different configurations. There are also H200, B200, RTX 5090, and so on. The price and performance of each model are different. Combined with the fast pace of technological updates and the high concentration of the market in the hands of NVIDIA and AMD, these factors have made it exceptionally difficult to establish a GPU futures market. But the good news is that the market is finally starting to address this pain point. About a month ago, @ OrnnExchange launched the Ornn Compute Price Index (OCPI). This index is based on actual trading data (not estimates, not manufacturer quotes), tracking hourly real-time rental prices for various GPUs such as H100, H200, B200, RTX 5090, etc. This index has been launched on Bloomberg Terminal and has become a standard reference for institutional investors. More importantly, Ornn is partnering with the Intercontinental Exchange (ICE) to launch the world's first GPU computing futures contracts. These contracts will be standardized products priced in US dollars and delivered in cash. Also collaborate with @ Lighter_xyz. At the same time, @ Silicon-Data has also launched its own GPU price index and partnered with the Chicago Mercantile Exchange (CME) to launch computational futures. CME is one of the world's largest futures exchanges, and their entry means that this market is gaining recognition from mainstream financial institutions. The launch of these futures contracts will allow market participants to go long or short on GPU computing power. Data centers can hedge their cost risks, GPU manufacturers can hedge their revenue risks, and investors can express their views on the future direction of the computing power market. Some even predict that the market size of computing power futures may eventually surpass that of oil futures. Why? Because AI will become the most important infrastructure for humans, and GPU computing power is the fuel for that infrastructure. With the popularization of AI applications, the demand for computing power will only continue to grow. This is not just technological progress, it is the evolution of financial infrastructure. We are witnessing the transformation of computing power from an 'invisible cost' to a 'tradable commodity'. The era of financialization of computing power is really coming. Next is compute+robotic finance