Original title: Agents will pay like locals, not tourists
Original author: Sam Broner, a16zcrypto
Translation: Peggy, BlockBeats
Editor's note: As AI Agents gradually evolve from auxiliary tools to "digital executors" capable of independently completing tasks, the payment system is also undergoing changes. In the past, online transactions primarily revolved around the retail process of "user clicks - checkout - payment," but in the Agent era, the subject of transactions is no longer just humans, but intelligent systems that can operate continuously and establish long-term cooperative relationships.
This article presents a vivid metaphor: Agents will not make one-off payments like "tourists," but will be more like "locals," completing transactions through stable supplier relationships, credit, and pre-negotiated business terms. In this model, the traditional card-centric payment system may only bear part of the transactions, while programmable payment tools like stablecoins are expected to play a greater role in new payment scenarios.
The following is the original text:
Walking into a marketplace, if you are a tourist, what you often see is a lively scene: crowds weaving between stalls, examining products, comparing prices, sampling goods, bargaining with vendors, pulling out coins or swiping cards to complete transactions. It seems that each interaction is like an independent business deal. A moment of negotiation, trust settled in cash or card instantly.
But in fact, most transactions do not happen this way.
If you observe closely, you will find that there are more locals in the marketplace. They purposefully approach familiar merchants. Restaurant owners go to their known butchers, fishmongers, and farmers; tailors seek out repairmen, weavers, and craftsmen. There is rarely any further bargaining among them, and many transactions are even completed on credit.
When we discuss how Agents will make payments, we often subconsciously start from the "tourist" perspective. But the behavior of Agents is more akin to locals.
The differences between Agents and humans, such as unlimited replication, flexible resource scheduling, and near-zero startup costs, mean that a small number of Agents can establish advantages in specific fields. Even if the threshold for building Agents continues to lower in the future, network relationships, partners, and trust mechanisms will remain key factors determining the quality of user experience.
Truly dominant Agents do not need tourist-like payment channels; what they need are supplier relationships, operating funds, and credit lines.
Agents will complete transactions alongside "tourists" (the users).
So, what form will this model take?
As Agents gradually evolve into platforms resembling businesses, their payment model will shift from retail payment networks to pre-negotiated B2B terms and credit systems. The existing payment infrastructure does not adequately meet this demand.
This provides an opportunity for the new generation of payment networks, such as stablecoins. However, this depends on entrepreneurs being able to build solutions around new payment scenarios, such as Agent payments, streaming payments, and high-frequency, low-value, global business transactions.
This article will expand on this perspective from three aspects: first, what key differences exist between Agents and humans, and how those differences shape future payment models; second, why existing payment systems struggle to meet the needs of Agents; third, what capabilities the next generation of payment infrastructure needs to possess to succeed in future competition.
Differences Between Agents and Humans
To understand Agents and payments, we need to answer two questions:
1. Do Agents behave more like individuals or more like businesses?
2. Do Agents' decisions lean towards short-term transactions or long-term partnerships?
The answer is: Agents are more like businesses and will establish long-term relationships.
Agents are often "lightweight instances" built on larger business systems. For example, a "smart tour guide Agent" supported by a major travel platform, or a franchisee that is fine-tuned according to local market demands within existing supply chains.
Why do Agents behave like businesses?
First, excellent experiences often come from advanced design rather than on-site negotiations.
Users do not want their Agents to start comparing prices, contacting merchants, or renegotiating terms only at checkout. An ideal Agent should have already completed this work: it knows which suppliers are reliable, prices have been negotiated, and it can directly complete transactions.
This represents a business relationship, not a one-time tourist transaction.
In fact, similar models have long existed in human society. Travel agents, literary agents, entertainment agents, watch dealers, and real estate agents all fit the definition of "Agents." These agents establish long-term cooperative relationships with publishers, production companies, watch distributors, or lending institutions, and each transaction is customized based on this foundation.
Second, Agents can be infinitely replicated, but the advantages of scaled enterprises cannot be replicated.
The most successful Agents will leverage the advantages brought by scale: lower computational costs, better vendor pricing, deeper system integration, and more stable technical components.
Scale will continually reinforce itself. A travel agent that books a million tickets a year will inevitably have better terms from airlines than an agent that books only ten tickets a year.
This trend has already emerged. Only products like ChatGPT have enough user distribution power to establish collaborations with platforms such as Shopify, Amazon, and Expedia. Small startups often have to rely on automated browsers or reverse API interfaces while bearing retail-level cost structures.
This is also why Agents will ultimately trend towards centralization, or at least most Agents will be built on large platforms.
Agents themselves are easy to develop, but economic laws dictate that only a few core Agents will ultimately exist in each vertical field; they will have deep supplier relationships and will be able to use profits to continuously optimize experiences.
At the same time, specialized Agents in vertical fields can also collaborate with user-side Agents to provide more complete services.
Two Types of Payment Relationships
If Agents' behavior is more akin to businesses, then two types of payment relationships need to be designed: User → Agent; Agent (or Agent platform) → Supplier
Users pay fees to Agents, which can take various forms: subscription fees, task-based fees, credit lines, authorized Agent use of user accounts
Agents then pay suppliers through B2B terms, such as: pre-negotiated prices, volume discounts, Net-30 invoices, sub-agent settlements
From the structure of current business expenditures, Agents occasionally still use retail payment channels, but this only accounts for a small portion of overall expenditure.
In fact, this resembles today's credit card system quite closely. Credit card issuers establish retail relationships with consumers, bearing risks while providing credit and rewards; while acquirers establish commercial relationships with merchants, completing transactions through negotiated rates, scaled settlements, and operating funds arrangements.
Agents and Credit Cards: Seemingly Compatible
Many believe that credit cards are actually a quite suitable payment tool for Agents.
Reasons include: widely accepted globally, suitable for transaction ranges from $20 to $1000, built-in arbitration and refund mechanisms, offer monthly billing, particularly important as it helps users understand their spending situations.
In the future, when Agents replace children and iPads as the primary source of "unexpected bills," this could become even more important.
But in reality, there are two issues: 1. Credit card technology is not suitable for Agent scenarios; 2. The charging model of credit cards puts the industry into a typical "innovator's dilemma."
Challenges in Upgrading Credit Card Technology
Almost all credit card systems assume human involvement: human approval, user interface interactions, traditional payment types (one-time or subscription)
Virtual card technologies like Stripe Link or Visa 3D have taken over 15 years to mature. However, the speed of Agent development far exceeds the pace of upgrades in payment infrastructure. Thousands of PSPs, POS systems, merchant backends, and client interfaces cannot be adapted in a short time.
Credit Cards Cannot Cover Extreme Payment Scenarios
For example: Agents make real-time streaming payments to computing services, Agents pay micro-fees for API calls, and these transactions are difficult to complete via credit cards.
The reasons are simple: Visa does not support transactions under 1 cent, and the credit card economic model relies on approximately 30 cents in fixed fees.
Technically, Visa can support micropayments, but it would directly impact its business model. More complex is that payment scenarios for Agents often exceed the traditional monetary range of credit cards. For example, many early Agent scenarios involve API service fees that are difficult to refund and hard to resell. Credit cards can still play a role, but the innovator's dilemma often restricts the speed of change within existing systems.
Traditional Payments Still Have Their Role
As Agent platforms gradually evolve into business-like systems, a large volume of high-frequency expenditures will be processed via B2B terms: invoices, Net-30, discounts, credit lines.
In this model, the "payment network" itself is not critical. Settlements may be made via wire transfer, ACH, or batch transfers. Traditional payments remain effective in mature business relationships. However, Agents will not only exist in such environments.
Agents are emerging rapidly, and they often operate in the scenarios where traditional payments are least efficient: initial cooperative relationships, cross-border payments, complex reconciliations, new Agent–Vendor models, instant payments, and micro-loans.
In these scenarios, stablecoins are a superior payment tool. More importantly, building new features on programmable currency is far easier than on traditional payment infrastructure.
Once new business relationships are established on stablecoins, these relationships often maintain that form for the long term. Over time, the proportion of stablecoins in the payment system is likely to increase continuously.
Opportunities for New Payment Technologies
Stablecoins are essentially a new financial platform.
They possess the following characteristics: faster, lower cost, globally available, supported 1:1 by high-quality liquid assets.
More critically, stablecoins are programmable. Functions such as arbitration, billing, credit, escrow, and conditional payments can be flexibly implemented within the same system.
Compared to banks or credit cards, stablecoin payments are easier to embed: API, database, Agent checkout processes.
This significantly simplifies reconciliation, approval, and system integration processes, which is particularly important for entrepreneurs building Agent business ecosystems.
In terms of economic models, stablecoins also solve the efficiency problems of credit cards at both ends: no 30-cent minimum fee, and large transfers are not eroded by interchange fees.
Therefore, whether it's Agents paying $0.001 per second for computing power, or businesses settling $50,000 supplier invoices, the same payment network can be used.
Building More Stablecoin Infrastructure
A common critique is that the entry and exit costs for stablecoins are relatively high.
For "tourists," this is indeed an issue. But when users are guided by Agents, this friction will quickly decrease.
Agents can help users complete currency exchanges and only execute necessary transactions to save on fees. If we also layer in billing and arbitration mechanisms, we come closer to a complete system.
We can imagine a scenario where users browse multiple brands in a department store, select products, and only need to check out once. The store backend is responsible for allocating payments to each merchant, and Agents need a similar model. What users see is: "Your Agent wants to book flights, hotels, and rent a car for you." Instead of handling three independent checkout processes.
The Agent platform is responsible for supplier relationships, while users only need to confirm transaction intents.
Conclusion
Agents do not make payments like tourists. They conduct transactions like locals, through relationships, credit, and long-term cooperation. This means that the true scale of payments in the future will flow through pre-negotiated B2B terms rather than card payments.
But we are currently in a critical window period. Agents are emerging, and entrepreneurs are building new business systems; they need payment tools that can be used today.
Credit cards are not ready: micropayment costs are too high, reconciliation is complex, technical debt is heavy, and reliance on manual risk control.
In contrast, stablecoins are already qualified. They are programmable, global, easy to integrate, and can support Agent payments from day one.
Payment relationships have strong path dependence. Once new business relationships are established on stablecoins, they often persist for the long term. In the coming years, as ecosystems mature and entry and exit frictions decrease, a batch of startups will build new capabilities around this infrastructure: billing systems, arbitration mechanisms, credit systems, bulk approvals, and cross-system interoperability.
The new payment era may well be starting from here.
免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。
