AI Agents Drive Millions Of Microtransactions On Base

Artificial intelligence agents have made more than 138 million payments via the x402 protocol on Base, according to a Chainalysis study published on October 14. However, this figure measures the number of transactions, not 138 million dollars: more than 87% of the analyzed operations are worth less than ten cents.

Tiny payments, but rapidly growing activity

The x402 protocol allows an autonomous software to pay directly for an online service without requiring human validation for each expense. Initially developed by Coinbase, it relies notably on stablecoins to settle amounts too small for traditional card networks.

The key figures of the study allow measuring the rise of this still experimental economy :

• More than 138 million x402 transactions have been recorded on Base since the protocol launch ;
• More than 87% of payments were between 0.01 cent and 10 cents ;
• The most frequent range was between 1 and 10 cents ;
• The median delay between funding a wallet and its first payment fell from 73 days to 18 minutes ;
• The share of wallets refilled within 30 days increased from 70.6% to 89.3%.

These operations are notably used to purchase data, query a programming interface, access an AI model, or have code reviewed. An agent assigned a task can thus independently acquire the digital resources necessary for its execution.

Stablecoins present an economic advantage here. A transaction of a few cents would generally be absorbed by card payment fees. On a low-cost blockchain like Base, it can remain viable.

Circle summarizes this usage distribution with a formula :

Cards for consumer-triggered agent payments, stablecoins for machine-to-machine settlements too small for the card economy.

The United States dominate AI model purchases

Wallets linked to the United States account for 30% of geolocatable payments made between agents and services on x402. Their dominance is even clearer in one specific category: they concentrate 88.1% of payments intended for using artificial intelligence models.

The share of wallets linked to China reaches only 0.1% in this category. Chainalysis suggests a possible explanation: Chinese developers might more often use open national models like DeepSeek or Qwen, while major American models are often billed by usage.

The situation reverses, however, for some specialized services. Chinese wallets are behind nine out of ten payments dedicated to security and code analysis. Most of this activity has been directed toward AgentLISA, a tool using a language model to audit Solidity smart contracts.

On-chain data therefore do not describe the entire global market. They only cover x402 transactions that Chainalysis could identify and geolocate. The strong American share does not mean the United States perform 88% of all payments between AI agents.

The number of transactions is not enough to prove adoption

The growth in payments and more frequent wallet refills suggest some projects are moving from simple tests to recurring uses. Hoolie Tejwani, head of Coinbase Ventures, thus considers that “AI agents are becoming economic actors, each agent effectively constituting a new blockchain user”.

However, this conclusion must remain cautious. A transaction recorded as coming from an agent does not necessarily prove that an AI system alone made a genuine economic decision. Classic scripts, automated tests, or operations between wallets controlled by the same actor can also inflate the statistics.

The cumulative amount transferred would moreover be a more meaningful indicator than the number of operations alone. With a majority of payments below ten cents, 138 million transactions do not necessarily represent a significant financial volume.

The most significant change lies rather in the possibility to monetize each request. Data, a query, or a few seconds of computation can be sold individually without subscription and without heavy banking infrastructure. Will Su, head of digital asset research at BlackRock, believes agents could thus create significant demand for frequent, low-value, and continuously executed payments.

The next step will be to determine whether these flows correspond to useful and sustainable services. For now, x402 mainly demonstrates that a blockchain can provide autonomous software with a means of payment suited to tiny expenses.