Coinbase targets AI payments as Armstrong sees agents driving most transactions within a decade
As Coinbase pushes to diversify beyond trading fees, the company is building payment infrastructure for artificial intelligence agents that it says could reshape digital commerce. Chief executive Brian Armstrong says those agents may account for well over 99.9% of all transactions within a decade, with much of the early activity already happening on Coinbase's Base blockchain using USDC.
Highlights
- Coinbase is developing payment tools and accounts tailored for AI agents to automate micro-transactions without human involvement, diversifying beyond core crypto trading revenue.
- Prediction markets and derivatives now contribute approximately 25% of Coinbase’s trading fee revenue, with stablecoins, staking, and custody further stabilizing results as market activity fluctuates.
- CEO Brian Armstrong stated he believes bitcoin has likely bottomed in the current cycle, maintains a $400,000 price target by 2030, and attributes recent weakness to capital rotation into major tech IPOs like Anthropic.
AI payments strategy expands beyond crypto trading
As first reported by CoinDesk, Armstrong said AI agents will soon outnumber humans and will need ways to pay each other for goods and services, especially for small and frequent transactions that traditional card networks may not handle efficiently.Coinbase is building financial accounts, payment tools and marketplaces designed to let AI agents coordinate work and pay for services without human involvement in each transaction. Armstrong described a use case in which one agent creating an advertisement hires separate AI services for voiceovers, video production and other tasks, paying each provider as work is completed.
The initiative fits Coinbase's broader effort to reduce its dependence on crypto trading, which remains exposed to swings in digital asset prices and investor activity. Armstrong said prediction markets and derivatives now generate about 25% of the company's trading fee revenue, while stablecoins, staking and custody also help support results when market activity slows.
Competition and market outlook shape expansion
Coinbase's move into AI-linked payments comes as competition intensifies across crypto, brokerage and prediction markets. Robinhood is expanding its crypto business while offering stocks, options and digital assets in one platform, and exchanges such as Kraken and Binance remain key rivals, particularly in derivatives trading.Armstrong said he is already using several AI agents to trade crypto with his own money, with mixed performance so far. He said Coinbase is developing tools that would let users create trading strategies, publish results and allow others to copy successful approaches, while adding safeguards such as spending limits, account permissions and testing before real funds are committed.
On the market outlook, Armstrong said he believes bitcoin has probably reached the bottom of the current cycle despite the recent selloff. He attributed some of the weakness to investors shifting money toward major technology listings, suggested a potential Anthropic IPO may be having a similar effect, and said he still sees bitcoin reaching $400,000 by 2030 as crypto adoption broadens into stablecoins, tokenized stocks and blockchain-based financial services.
As we previously wrote, the proposed Onchain Rights Standard aims to standardize tokenized stock shareholder notices in digital wallets.