China is moving to create an entirely new class of financial derivative built around artificial intelligence. According to an exclusive Reuters report published on May 28, the Shanghai Futures Exchange is in the early stages of designing futures contracts tied to AI tokens, the smallest unit of information processed by large language models, in what sources describe as a direct response to the United States’ own push to financialize AI infrastructure.
The initiative is preliminary and no launch timeline has been confirmed. The Shanghai Futures Exchange and China’s securities regulator, the China Securities Regulatory Commission (CSRC), did not respond to Reuters’ request for comment. Nevertheless, the report signals that Beijing views the financialization of AI compute not merely as an economic opportunity but as a strategic front in the broader US-China technology rivalry.
What AI Token Futures Would Actually Do
To understand why this matters, it helps to understand what AI tokens are and why they have financial value. Tokens are the discrete units of text — roughly three-quarters of a word on average, that AI models consume when generating a response. Every query to a model like DeepSeek, Qwen, or ChatGPT consumes tokens, and the cost of those tokens is how AI companies price their services. As AI adoption scales, token consumption has become a meaningful economic variable.
Xiao Feng, Chairman and CEO of HashKey Group, described tokens as the “digital fuel” or “raw material” that powers AI models. China’s daily token usage has surged 1,000-fold since the start of 2024, reaching more than 140 trillion tokens per day by the end of March 2026, according to official data. That explosive growth has created real economic exposure for companies that depend on AI services — and where there is price exposure, there is demand for hedging instruments.
A futures contract tied to token prices would allow AI companies, cloud providers, and enterprise users to lock in future token costs, protecting themselves against price spikes caused by compute shortages or surging demand. The Shanghai exchange’s product would be tied to AI tokens used for pricing AI services, distinguishing it from the approach being taken in the United States.
How China’s Approach Differs from the US
The contrast with the American strategy is instructive. CME Group and Intercontinental Exchange (ICE) are both preparing to launch GPU compute futures — derivatives tied to the cost of renting the graphics processing units that power AI model training. The US approach is infrastructure-level, pricing the hardware layer. China’s proposed approach is service-level, pricing the consumption layer.
Both products serve the same ultimate purpose: giving companies along the AI supply chain a way to hedge against the cost of computing power. But the different architectures reflect different market structures. China’s AI industry is dominated by large model providers and enterprise software companies that consume tokens at scale; the US market is more hardware-centric, with Nvidia’s GPU dominance creating a natural benchmark for compute futures.
Yilei Shao, dean of the Shanghai AI-Finance School at East China Normal University, argued that China should move quickly. “The United States and China are the only two nations capable of mass-producing artificial intelligence,” she told Reuters. “The maxim ‘Whoever masters AI, rules the world’ is hardly an exaggeration.” She described token futures as key to both the technological and financial competition between the two countries.
The Infrastructure Already Exists
The groundwork for a token futures market has been quietly laid over the past year. In December 2025, China’s official commodity index company published a series of indices tracking the country’s compute supply, instruments that could serve as underlying benchmarks for futures contracts. Zhang Yunquan, a computing technology researcher at the Chinese Academy of Sciences, formally proposed the launch of compute futures to China’s parliament in March 2026.
Brokerage Baocheng Futures estimated in a research note earlier this month that China could debut compute futures within three to five years, though it cautioned that the current fragmented market structure remains an obstacle. The note reflects a broader consensus that the infrastructure and regulatory framework are not yet ready for a live product, but that the direction of travel is clear.
BlackRock CEO Larry Fink added international weight to the idea at a conference earlier in May, arguing that surging demand for tokens could spawn an entirely new asset class in buying futures of compute. His comments, made in a different context, nonetheless validate the thesis that token markets are a serious financial frontier.
This development fits into a pattern of Chinese financial innovation aimed at giving domestic markets a stake in the AI economy. As EastFrontier has reported, China’s AI startup funding tripled to $16.2 billion in Q1 2026, and Alipay launched an AI wallet and token pay system to power agentic commerce. A token futures market would add a third layer, a derivatives market that allows the entire AI supply chain to manage financial risk as the industry scales.