AI consolidates U.S. dollar hegemony through stablecoins

Dedollarization debate is raging, artificial intelligence may become a key variable

As the dedollarization debate intensifies, artificial intelligence may change the rules of the game. This is at least the view of two economists at the ASEAN and China, Japan and South Korea Macroeconomic Research Office (AMRO). In an analysis report released on July 24, 2026, they raised a novel argument: instead of threatening the dominance of the dollar, artificial intelligence will strengthen it. What are the key factors? The answer is: the large-scale use of stablecoins backed by the US dollar.

Brief overview

Artificial intelligence may strengthen the global dominance of the US dollar.
Computing power will create a continuous demand for US dollar liquidity.
Artificial intelligence agents may prefer U.S. dollar stablecoins.
Although this scenario is reasonable, its commercial application still remains uncertain.

How artificial intelligence reshapes global financial infrastructure-Analysis by Chen Xufu and Huang Xianguo

The core of the artificial intelligence analysis by two AMRO economists is the "computation-dollar cycle"(energy-computation-dollar cycle). This economic mechanism consists of four steps:

First, electrical energy drives data center operations.
Second, data centers convert electricity into billable computing power.
Third, this computing power has become the recurring operating expenses of global companies denominated in dollars.
Fourth, automatic payments generated by artificial intelligence agents are settled through U.S. dollar stablecoins, and their reserves will be invested in U.S. Treasury bonds.

If computing power becomes an indispensable input factor in the economy, the influence of its denominated currency will increase accordingly. Artificial intelligence may replicate the mechanism of the oil market-the pricing of oil in dollars has long supported the international market's demand for dollars.

How can artificial intelligence infrastructure contracts create structural dollar demand?

Artificial intelligence is not an "invisible" industry. It consumes electricity, water, semiconductors and industrial real estate. These resources are increasingly secured through long-term dollar-denominated contracts.

To support this view, Chen Xufu and Huang Xianguo cited Anthropic's case. The American artificial intelligence company just signed a 20-year data center lease contract with TeraWulf. This is not just a simple real estate lease, but also a supply-side bet on the continued need for computing power.

Interpretation: Anthropic promises to pay in U.S. dollars over 20 years to ensure access to its energy and computing resources.

At the same time, OpenAI also launched DeployCo. The consulting subsidiary stationed engineers within client companies to assist in deploying artificial intelligence systems. Once these systems are integrated, computing power will become a recurring operating expense like power and telecommunications.

These contracts thus create structural dollar demand, as AI costs are recorded on the balance sheets of global companies in dollar currencies. Whether companies in Japan, Indonesia or Germany want to use artificial intelligence for logistics, accounting or customer relationship management, they need to continue and increasingly access to dollars, rather than just occasional use.

According to Chen Xufu and Huang Xianguo, this is the first channel to strengthen the dominant position of the US dollar. They estimate that dollar-denominated computing power will create a permanent demand for dollar liquidity.

The second key element of AMRO's argument: payment settlement

As artificial intelligence agents become ubiquitous in logistics, inventory management, and corporate cash management, they will require programmable currencies. The dollar stablecoin just meets this demand. In addition, industry players such as Circle have built infrastructure that supports programmable transfers around the clock, eliminating the need for traditional banking intermediaries.

OpenAI has also signed a cooperation agreement with Visa to specifically build these programmable payment channels. At the same time, the release of Open USD shows that the ecosystem is ready for this transformation.

Therefore, the question is no longer whether stablecoins are the only option, but whether they can be the first to achieve large-scale and efficient cross-border operations. If so, agency businesses will naturally rely on programmable dollars.

Is the dollar dominance brought by artificial intelligence really inevitable?

For many analysts, the scenario proposed by the AMRO economists is reasonable but not certain. They believe that artificial intelligence agents are already conducting experimental operations, but there is no evidence that they will quickly become the main payers in the real economy. In short, technology demonstrations are far from commercial applications.

stablecoins are not without weaknesses. The main issues include: unanchoring risk, issuer concentration, dependence on blockchain, cybersecurity and compliance. Indeed, if an compromised artificial intelligence agent generates a large number of erroneous payments in an instant, the speed of automatic settlement may be dangerous.

Last but not least, artificial intelligence will not spontaneously choose the dollar. It will use the lowest cost, most liquid and easiest to integrate tools. The United States currently has this advantage thanks to the depth of its market and the dominance of the USDT and USDC.

As a result, the artificial intelligence economy is not only transforming businesses, it is also redrawing the foundations of the international monetary system. The next global game may be played out between the "wallets" of autonomous agents. More follow-up developments deserve attention.