BlackRock: Autonomous AI Could Drive Stablecoin Demand

BlackRock’s report The Machine-Native Economy projects that increasingly autonomous AI agents will execute many small payments to buy data, software, APIs and computing power, creating steady demand for stablecoins. The report estimates more than $300 billion in stablecoins were in circulation and about $11.2 trillion of adjusted transaction volume in 2025.

The report describes a payment pattern that is small in size, frequent in timing and required around the clock. Software wallets can hold stablecoins and settle transfers instantly without a person approving each payment. Stablecoins are already being used to settle many of those transfers.

BlackRock calculates stablecoin transaction volume grew at roughly an 80% compound annual rate from 2020 to 2025. The firm compares that with about 8.5% growth for the U.S. Automated Clearing House network over the same period and notes ACH processed roughly $93 trillion last year. The report cautions that measurement methods differ across networks.

Several technical approaches are being developed to handle machine-originated payments. Coinbase’s x402 protocol uses the web’s HTTP 402 Payment Required status to request payment before returning a resource, allowing an agent to transfer USDC and receive a response without a human checkout. Stripe and Tempo are building the Machine Payments Protocol, which can settle with stablecoins or conventional methods. Stripe and OpenAI are working on an Agentic Commerce Protocol to connect agents with merchant systems, and Google and Visa are developing standards for agent identity and authorization.

The report states that machine payments might use blockchains, existing payment rails or hybrid systems depending on cost, speed and ease of integration. On Ethereum, more activity could increase demand for blockspace and for ETH used in fees and staking, but the connection depends on fee structures, staking economics and gas-sponsorship models. Circle’s Arc treats USDC as the native gas asset on its chain, so rising activity could strengthen the stablecoin’s role without necessarily boosting a separate platform token.

BlackRock highlights AI compute as a large potential market. The firm estimates cumulative investment in AI infrastructure could exceed $5 trillion between 2025 and 2030. Forecasts cited in the report put combined cloud revenue for Amazon Web Services, Microsoft’s cloud business and Google Cloud at about $1.1 trillion by 2030.

The report outlines how autonomous agents could compare providers on price, hardware, latency or performance, buy compute for a defined task and pay per job or per model token. Payments could become recurring machine-to-machine flows for inference and training.

If such commerce scales, markets for standardized claims on computing capacity could form, with those claims traded, used as collateral or hedged through futures. Establishing those markets would require standards that account for differences in chips, energy costs, geography and performance. The report notes agent-originated payments remain at an early stage and that traditional financial firms and crypto-native companies are building infrastructure in parallel.

Private and public infrastructure providers are competing for what the report calls the machine’s wallet. Stablecoin issuers aim for their tokens to become the default settlement asset while payment protocols seek adoption by agents and services. Blockchains face the task of converting higher stablecoin throughput into demand for their native economic assets, and traditional payment networks have incentives to keep activity on existing rails. As software gains delegated spending authority, payment infrastructure providers are preparing to serve automated customers that transact continuously and prioritize price, speed and programmability.