
Coinbase CEO Brian Armstrong rejects AI pivot calls, arguing AI agents need programmable money. Base network, USDC, and x402 protocol hit 100M agentic payments in nine months, per Chainalysis.
Alpha Score of 38 reflects weak overall profile with weak momentum, weak value, poor quality, strong sentiment.
Coinbase CEO Brian Armstrong is rejecting the idea that crypto companies should shift focus from blockchain to artificial intelligence. He argues the two technologies are complementary, not competing, because autonomous AI agents will need programmable money to operate without traditional bank accounts or checkout systems.
"I used to hear versions of this, and it's the wrong way to think about the world. It's zero sum, scarcity thinking," Armstrong wrote on X. "Crypto is a general purpose technology. It's infrastructure, the same way electricity or the internet is infrastructure."
Armstrong's vision of agentic finance depends on AI systems becoming active participants in the digital economy. Three Coinbase products sit at the center of that model: the Base network, the USDC stablecoin, and the x402 payment protocol. Together they form the company's current infrastructure for machine-to-machine commerce.
Base launched in 2023 as an Ethereum layer-2 network, originally designed to make onchain applications faster and cheaper rather than specifically serving machine payments. The network has since become the foundation for Coinbase's autonomous payment strategy. In 2025, Coinbase introduced x402, a protocol based on the HTTP "402 Payment Required" standard. It allows applications and AI agents to make automated stablecoin payments for resources like APIs and data without opening conventional accounts or navigating checkout pages.
USDC supplies the dollar-linked payment asset within that system. x402 provides the communication standard. Base handles the underlying transactions.
The clearest evidence of adoption is the rapid growth of agentic payments on Base. Chainalysis reported in June that x402-related activity surpassed 100 million transactions within roughly nine months. Transactions worth at least $1 represented 95% of the total value transferred, indicating the activity was not driven entirely by negligible test payments or empty technical demonstrations.
The wallet patterns behind that activity add another layer. Chainalysis found that wallets making agentic payments were generally newer, held a wider variety of assets, and maintained smaller balances than typical Base users. Those characteristics suggest an emerging user category built for frequent automated spending rather than passive asset storage.
Armstrong's broader argument is that AI agents will generate new demand for crypto services instead of displacing them. Coinbase must still demonstrate that transaction growth translates into durable economic value. The early figures give agentic finance a measurable foundation beyond promotional language, though developers are still testing whether autonomous commerce can scale reliably.
The bet comes as other crypto firms explore similar territory. Franklin Templeton's head of digital assets, Sandy Kaul, recently argued AI agents represent the next "killer" use case for blockchain. On the XRP Ledger, AI agents recorded 1.4 million transactions, according to on-chain platform t54.ai. Ledger launched an open-source Agent Stack toolkit that lets AI agents read balances and suggest transactions without executing them autonomously.
Armstrong's position places Coinbase in direct competition with those approaches, betting that a full-stack solution – layer-2 network, stablecoin, and payment protocol – will win the emerging machine-to-machine payment market. The 100 million transaction figure on Base gives him data to point at. Whether those transactions generate sustainable fee revenue is the open question.
Drafted by a large language model from the source reporting linked above, then screened by automated publishing checks. It is not read by a journalist before publication. Some articles cite our Alpha Score. Verify prices and figures against the original source. Educational coverage, not personalized advice.