万事达最近做了一件相当富有创意的事情。昨晚，他们正式推出了一种名为机器代理支付（Agent Pay for Machines，简称 AP4M）的支付系统，专门为人工智能代理设计。简单来说，它允许人工智能程序独立进行支付和收款，而无需人类确认。为什么这很必要？因为它改变了人工智能的工作方式。之前，人工智能仅提供建议；现在它可以直接执行任务。例如，如果你拥有一家花店并想创建一个网站，你可以告诉你的人工智能：“在 100 美元的预算内，获取域名、主机、图片和结账页面。” 然后，人工智能将自动与多个服务提供商进行谈判、下订单并进行支付，每笔交易的成本可能仅为几美分。这种交易传统银行卡根本无法完成；频率太高，金额太小，并且需要反复的人类授权。万事达的首席产品官直言不讳：“AP4M 将真正使人工智能商业模式蓬勃发展。机器支付的规模、速度和金额与今天的人类支付完全不同。” 那么，加密货币在其中扮演什么角色？首先，授权记录存储在区块链上。我们为人工智能设定的权限，例如最大支出金额和可以购买的服务类型，将存储在 Polygon 区块链上。为什么要放在区块链上？因为在未来，可能会有多个系统来验证人工智能是否超越了其权限。通过将其放在公共账本上，机构可以相互验证，消除需要一个中心化机构来认可的必要性。万事达目前选择 Polygon 作为其初步部署。其次，结算支持稳定币。除了银行卡和账户外，该网络还允许使用稳定币（如 USDC）进行结算，参与的公司包括 Coinbase、Ripple 和 Solana 等加密领域的公司。RippleX 的一位高级副总裁表示：“自主代理已经能够独立结算，但只有在控制跟随的情况下，机构才能放心运营。链上加上稳定币实现了二级结算、可编程合规和全流程审计。” 该系统的基本逻辑并不复杂：每个代理都被发放一个可验证的身份（凭证化）；公司可以设定程序化规则（例如，每笔交易不超过 0.50 美元）；促进代理之间的高频交易；最后，实现跨渠道结算。当然，万事达也没有放弃其核心业务，开放了其全球清算网络，同时将加密技术嵌入作为信任层。目前，已有超过 30 家公司表示支持，包括 Adyen、Ant International、Stripe、Cloudflare 和 Checkout.com。这一意义在于，传统支付巨头并没有绕过加密系统，而是将其作为解决现实问题的工具。高频、小额、自动化和跨系统场景是传统渠道无法处理的领域，而加密可以填补这些空白。然而，仍然存在一些问题：如何防止滥用人工智能授权边界？如果代理出错，谁负责？但这些并不妨碍对趋势的判断：支付的未来正在从 “需要一个人按确认按钮进行支付” 转向 “程序根据规则自动进行持续支付”，而万事达作为全球第二大卡组织，正在引领这一潮流。
<img src="https://imageproxy.pbkrs.com/https://img.jinse.com.cn/7477052_watermarknone.png?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" width="600" height="335" original-src="https://imageproxy.pbkrs.com/https://img.jinse.com.cn/7477052_watermarknone.png"/>

万事达

CRCL

CRCG

CRCA

CRCD

万事达推出了 “机器代理支付”（Agent Pay for Machines，简称 AP4M），这是一种支付系统，使人工智能代理能够自主执行高频率的小额交易，无需人工干预。AP4M 利用 Polygon 区块链进行授权记录，并使用 USDC 等稳定币进行结算，旨在彻底改变机器对机器的商业模式。在主要科技公司的支持下，该倡议将传统清算网络与加密基础设施相结合，以促进可编程合规和自动支付

Mastercard recently did something quite imaginative. Last night, they officially launched a payment system called Agent Pay for Machines, or AP4M for short, specifically for AI agents. Simply put, it allows AI programs to pay and receive money independently, without requiring human confirmation. Why is this necessary? Because it changes the way AI works. Previously, AI only offered suggestions; now it can directly perform tasks. For example, if you own a flower shop and want to create a website, you can tell your AI, &#34;Within a $100 budget, get the domain, hosting, images, and checkout page all set up.&#34; Then the AI ​​will automatically negotiate with multiple service providers, place orders, and make payments, each transaction potentially costing only a few cents. This kind of transaction simply can't be done with traditional bank cards; the frequency is too high, the amounts are too small, and it requires repeated human authorization. Mastercard's Chief Product Officer put it quite bluntly: &#34;AP4M will truly make AI business models flourish. The scale, speed, and amount of machine payments are completely different from today's human payments.&#34; So what role does cryptocurrency play? First, authorization records are stored on the blockchain. The permissions we set for AI, such as the maximum amount to spend and the types of services it can purchase, will be stored on the Polygon blockchain. Why on the blockchain? Because in the future, there may be more than one system to verify whether AI has overstepped its authority. By placing it on a public ledger, institutions can verify each other, eliminating the need for a centralized institution to endorse it. Mastercard has currently chosen Polygon for its initial deployment. Second, settlement supports stablecoins. Besides bank cards and accounts, this network also allows settlement using stablecoins like USDC, with companies in the crypto space such as Coinbase, Ripple, and Solana participating. A senior vice president of RippleX stated, &#34;Autonomous agents are already settling independently, but only when control follows can institutions feel confident operating. On-chain plus stablecoins enable second-level settlement, programmable compliance, and full-process auditing.&#34; The underlying logic of this system is not complex: each agent is issued a verifiable identity (credentialization); companies can set programmatic rules (e.g., no more than $0.50 per transaction); high-frequency trading between agents is facilitated; and finally, cross-channel settlement is achieved. Of course, Mastercard hasn't abandoned its core business either, opening up its global clearing network while embedding cryptography as a trust layer. Currently, over 30 companies have expressed their support, including Adyen, Ant International, Stripe, Cloudflare, and Checkout.com. The significance of this is that traditional payment giants haven't circumvented encryption systems, but rather are using them as a tool to solve real-world problems. High-frequency, micro-amount, automated, and cross-system scenarios are areas traditional channels cannot handle, and encryption can fill these gaps. However, some issues remain: How to prevent abuse of AI authorization boundaries? Who is responsible if an agent makes a mistake? But these don't hinder the judgment of a trend: the future of payment is shifting from requiring &#34;a person to press a confirmation button to make a payment&#34; to &#34;programs automatically making continuous payments according to rules,&#34; and Mastercard, as the world's second-largest card organization, is leading the way.
<div class="lb-trans">万事达最近做了一件相当富有创意的事情。昨晚，他们正式推出了一种名为机器代理支付（Agent Pay for Machines，简称 AP4M）的支付系统，专门为人工智能代理设计。简单来说，它允许人工智能程序独立进行支付和收款，而无需人类确认。为什么这很必要？因为它改变了人工智能的工作方式。之前，人工智能仅提供建议；现在它可以直接执行任务。例如，如果你拥有一家花店并想创建一个网站，你可以告诉你的人工智能：“在 100 美元的预算内，获取域名、主机、图片和结账页面。” 然后，人工智能将自动与多个服务提供商进行谈判、下订单并进行支付，每笔交易的成本可能仅为几美分。这种交易传统银行卡根本无法完成；频率太高，金额太小，并且需要反复的人类授权。万事达的首席产品官直言不讳：“AP4M 将真正使人工智能商业模式蓬勃发展。机器支付的规模、速度和金额与今天的人类支付完全不同。” 那么，加密货币在其中扮演什么角色？首先，授权记录存储在区块链上。我们为人工智能设定的权限，例如最大支出金额和可以购买的服务类型，将存储在 Polygon 区块链上。为什么要放在区块链上？因为在未来，可能会有多个系统来验证人工智能是否超越了其权限。通过将其放在公共账本上，机构可以相互验证，消除需要一个中心化机构来认可的必要性。万事达目前选择 Polygon 作为其初步部署。其次，结算支持稳定币。除了银行卡和账户外，该网络还允许使用稳定币（如 USDC）进行结算，参与的公司包括 Coinbase、Ripple 和 Solana 等加密领域的公司。RippleX 的一位高级副总裁表示：“自主代理已经能够独立结算，但只有在控制跟随的情况下，机构才能放心运营。链上加上稳定币实现了二级结算、可编程合规和全流程审计。” 该系统的基本逻辑并不复杂：每个代理都被发放一个可验证的身份（凭证化）；公司可以设定程序化规则（例如，每笔交易不超过 0.50 美元）；促进代理之间的高频交易；最后，实现跨渠道结算。当然，万事达也没有放弃其核心业务，开放了其全球清算网络，同时将加密技术嵌入作为信任层。目前，已有超过 30 家公司表示支持，包括 Adyen、Ant International、Stripe、Cloudflare 和 Checkout.com。这一意义在于，传统支付巨头并没有绕过加密系统，而是将其作为解决现实问题的工具。高频、小额、自动化和跨系统场景是传统渠道无法处理的领域，而加密可以填补这些空白。然而，仍然存在一些问题：如何防止滥用人工智能授权边界？如果代理出错，谁负责？但这些并不妨碍对趋势的判断：支付的未来正在从 “需要一个人按确认按钮进行支付” 转向 “程序根据规则自动进行持续支付”，而万事达作为全球第二大卡组织，正在引领这一潮流。
</div><img src="https://imageproxy.pbkrs.com/https://img.jinse.com.cn/7477052_watermarknone.png" alt="" original-src="https://img.jinse.com.cn/7477052_watermarknone.png"/>

Mastercard has quietly pulled off a major move that could revolutionize the payment industry.

- 万事达卡推出 AI 支付系统 AP4M，实现机器自主支付。  
- 该系统利用区块链记录授权，支持稳定币结算。  
- 传统支付企业利用加密技术解决高频小额交易问题。

万事达卡悄然实施了一项可能彻底改变支付行业的重要举措