Jessy, Golden Finance. On September 16th, Google and Coinbase jointly launched the Agentic Payments Protocol (AP2), integrated with the X402 stablecoin payment rail. This move gave AI agents true payment capabilities. In other words, AI is no longer just able to converse and execute, but can now "spend money." At this moment, the machine economy has moved from imagination to reality. The curtain has officially risen on the machine economy. As early as August 2024, Coinbase pioneered a historic experiment: an AI bot used cryptocurrency to purchase tokens from another AI bot. Coinbase CEO Brian Armstrong publicly announced at X that this crypto transaction was initiated and completed entirely by AI, with no human intervention. The key infrastructure behind this was Coinbase's MPC wallet. It allows AI agents to directly generate and hold wallet addresses, making on-chain payments using stablecoins like USDC. Since AIs cannot open accounts in the banking system, MPC wallets become their key to entering the economic system. Coinbase has since continued to push the boundaries of AI payments. They demonstrated a demo of AI paying humans directly in cryptocurrency, launched the CDP AI Builder program, which supports wallet integration with AI agents, and collaborated with startups like Skyfire to explore AI applications in autonomous procurement, labor settlement, and cross-border payments. The logic is clear: if AI is to perform economic activities in the real world, it must first learn to trade, and blockchain and stablecoins provide the ideal platform. Coinbase's experiment proved feasible, but Google's involvement signals that this direction is being embraced by mainstream tech companies. Thanks to Google Cloud's Agent2Agent (A2A) protocol, AIs on different platforms can already communicate and collaborate. The newly released AP2 protocol and X402 payment rails give AI agents true payment capabilities for the first time. The agent not only proposes solutions but also autonomously completes settlement, requiring no human intervention. In a demonstration with the Lowe's Innovation Lab, the AI agent completed a complete shopping trip: diagnosing needs, recommending products, confirming the order, paying with USDC, and triggering fulfillment. No bank card information was entered, and no human placed the order manually. The human, as a natural person, simply issued instructions and expressed confirmation, and the transaction was finally completed on the blockchain. Google's role makes this development landmark. This isn't just a fringe experiment conducted by a blockchain startup; it's the world's largest cloud services and AI infrastructure giant, directly driving the standardization of AI agent payments. A Natural Fit Between Blockchain and AI Agents The core characteristics of AI agents are autonomous decision-making and automated execution. However, if they can't complete transactions, their role in the economic system will remain limited. The payment process is crucial in transforming "intelligent agents" into "economy." Blockchain and stablecoins provide the foundation for all of this. First, they eliminate the barrier to opening an account. AI can't open a bank account, but it can generate a wallet address in seconds. Second, on-chain payments are settled in real time, allowing funds to arrive within seconds, eliminating the need for complex clearing processes. Finally, blockchain payment costs are low enough to support micropayments of a few cents or even less.
This means AI agents can be truly applied in a wide range of scenarios. For example, research agents can automatically pay for database access and instantly generate reports; code review agents can charge a few cents based on the number of bugs found; customer service agents can instantly call and pay for translation agents; and purchasing agents in manufacturing can even automatically place orders and complete cross-border settlements.
This isn't just a functional complement; it's the beginning of a machine economy. Thousands of AI agents will exist as economic agents, with blockchain providing them with payment, identity, and property rights protections. Transactions between them will form a new micro-economy.
As AI agents take on increasingly complex tasks, from cross-platform collaboration to automated contract fulfillment, seamless payments will become a core driver. For developers, this presents a new economic testing ground. They can create services run by agents, explore pay-as-you-go business models, or build entirely agent-driven microeconomic systems. For the blockchain industry, this provides the most direct implementation scenario for "crypto-native AI." Blockchain provides the transaction and property rights foundation for AI, while AI provides real application needs for blockchain. The combination of the two may become a growth driver in the next tech cycle. As Google and Coinbase emphasized in their announcements, payments are just the beginning. As AI agents continue to integrate autonomous workflows, transactions will not just be ancillary activities but will become the core of the machine economy.
Joy
