cloudflare captcha loop - Silicon Valley giants collaborate on AI-powered payment protocol x402
402, dormant for years, has been awakened by AI
Over the years, there has been an interesting contrast on the internet.
The internet has turned increasingly many tasks that were once cumbersome into nearly imperceptible actions—sending messages, hailing a ride, and ordering food delivery, all alike.

The act of "paying" has always been a bit of an exception.

Often, when you want to use a service that involves spending money, things become more complicated—you can’t use it directly; instead, you must first open an account, link a bank card, deposit funds, and then purchase a plan or service.
Information flows faster online, but money has never truly flowed like information.
This is not an issue that people have only just realized today. Long ago, the design of internet protocols actually left room for this very thing.
In HTTP, there has long been a status code called 402 Payment Required, meaning: the resource is not being denied to you—you simply need to pay first. This code has remained in the technical standard for many years, and in subsequent specifications, its description has remained deliberately cautious: keep it reserved, for future use.
The developer documentation offers a similar explanation, stating that this space was originally reserved for digital cash and micropayments, but over the years, it has never truly been put into practice.
Looking back now, this detail is actually quite interesting.
It illustrates that the internet long understood it would eventually need to add a layer—not about how information is transmitted, but how value flows. However, for all those years, this need never truly pressed itself upon the industry.
Until today, things suddenly became different.
On April 2, 2026, the Linux Foundation announced the formation of the x402 Foundation, which will receive the x402 protocol contributed by Coinbase. The real significance of this news isn't that an old status code has finally been "revived," but that the major players seated at this table are undeniably titans.
Cloud and network companies like Google, AWS, Microsoft, and Cloudflare have arrived. Payment and commerce infrastructure companies such as Stripe, Visa, Mastercard, American Express, Adyen, and Shopify have also joined. Players from the stablecoin and public blockchain ecosystems, including Coinbase, Circle, Base, Polygon Labs, and the Solana Foundation, are here too.
Just by looking at this list, you can tell this isn’t a casual tech meetup or nostalgic gathering by a few engineers.
How will we spend money on the internet of the future?
What new development has suddenly aligned these players?
The answer is simply an AI agent.
Over the past several years, software subscription services have become the default business model in the internet world. Whether you’re building software, a platform, or enterprise services, you’ll likely end up charging on a monthly, annual, or per-user basis. Why has this model worked so well in the past? Because it aligns with human behavior: a company purchases a system for its employees, creates accounts, goes through budgeting, contracts, and finance processes—all of which naturally evolved alongside human workflows.
But that's not how AI works.
An AI agent is not an employee sitting at a desk. It functions more like an automated execution node that continuously receives tasks, invokes services, and purchases resources. It might call dozens of APIs within a minute to buy compute power, browser instances, model inference, mapping services, CAPTCHA capabilities, and various other API services. It rarely thinks, “Should I upgrade to a premium membership?”—it simply evaluates, “To complete this task right now, should I spend an extra $0.00001?”
At this point, trying to make it fit today’s subscription and checkout processes, which are primarily designed for humans, starts to feel awkward.
When Stripe launched its machine payment protocol in March this year, they were straightforward: AI is evolving from chatbots into agents capable of planning, executing, and evaluating outcomes—and the existing financial system was not designed for such entities. The issue isn’t that payments can’t be made; it’s that the current payment system is too heavy, too coarse-grained, and too friction-filled.
It was precisely at this time that x402 was once again seriously discussed.
Coinbase’s designed logic is actually straightforward: the client requests a resource, the server responds with a 402 status and includes the payment requirement; after the client completes the payment, it resubmits the request along with proof of payment, and the resource is then released. Coinbase’s stance is equally direct: they want stablecoin payments to run directly over HTTP, enabling interfaces, applications, and AI agents to automatically settle payments without complex account systems or manual billing processes.
Now, many online resources don’t require you to buy a membership or subscription first—they let you pay only for what you use, one use at a time.
At this point, many people might think, isn't this just the micropayment concept that the internet has been talking about for years but never truly delivered?
Yes, and no.
In the past, micropayments were discussed as users paying a few cents or dimes for an article, a song, or a video. Why has this never taken off? It’s not because there’s no demand, but because humans themselves are unsuited to serve as high-frequency settlement nodes in this system. When you have to go through a transfer, enter a password, and confirm a payment just for content worth a few dimes, the process itself erases the very value of the transaction.
Machines are different. They don’t get annoyed or find things troublesome. If the protocol is designed correctly, they are naturally suited for high-frequency, small-amount, programmatic, and instant payments.
So 402 was truly awakened this time not because people suddenly remembered the old dream of micropayments, but because, for the first time, a group of genuine users who need it emerged on the internet.
Tech giants from Silicon Valley enter the scene
Next, what follows is not a technical issue, but a matter of industrial specialization.
Why are companies like Google, AWS, and Cloudflare involved? It’s not because they’ve suddenly developed an interest in protocol history, but because they understand better than anyone that the things intelligent agents will consume most frequently in the future aren’t content, but capabilities—compute power, bandwidth, browser instances, model inference, data APIs, and workflow services—these are the resources most likely to be procured by machines in the future.
In the past, these resources were often sold only in bundles, via subscription, or through pre-paid充值, not because they couldn’t be sold individually, but because the transactions were too fragmented. Each sale required a full process of account setup, card binding, billing, collections, and reconciliation—resulting in high costs and significant bad debt risk. Now, if layer 402 operates effectively, these resources will have the opportunity to sell many previously indivisible items incrementally, one transaction at a time.
So what they value is not "another payment option," but a lower-friction fee interface.
The same applies to established payment companies like Stripe, Visa, and Mastercard. They didn’t suddenly pivot to Web3; rather, they clearly understood that if machine payments were to scale, they couldn’t cling to the outdated logic of “human button-clicking.” Otherwise, the next wave of transaction volume might very well bypass them entirely.
So what they’re doing now isn’t rejection—it’s acceptance while repositioning themselves. Stripe is straightforward: merchants can still use its payment intent API, funds can still flow into their merchant balances, and systems for taxes, refunds, and reconciliation continue unchanged. Visa has also rolled out specific card payment standards and toolkits for machine-to-machine payment protocols, essentially saying: yes, machines buying things on their own is possible—but I need to remain in control of the critical layers like authentication, authorization, risk management, and the payment network.
As for Coinbase, Circle, and other Web3 players, this finally presents what may be the largest-scale application opportunity for crypto assets.
In the past, many people discussed stablecoins by focusing on aspects like dollar hegemony, on-chain finance, and crypto liquidity—these are certainly valid. But in the context of x402, the most straightforward and practical value of stablecoins is that they can serve as a programmable, micro-scale global settlement tool.
Especially when the trading entities shift from humans to machines, this value transforms from merely a financial narrative into the infrastructure of the entire internet industry.
So, having come this far, I believe this is no longer the old question of whether cryptocurrencies are legal or not. It’s more like a battle to define the operating rules of the next-generation internet: what interfaces will machines default to when invoking resources, purchasing services, or allocating budgets? What assets will they use for settlement? And whose rules will they follow by default?
Understanding this explains why the Linux Foundation took over x402.
Because anything intended to become a foundational standard is hard for others to fully trust if it remains under a single company. Placing it into a foundation essentially provides it with a neutral framework.
Why didn’t several institutions in the Web3 camp emphasize as much this time on making users aware of wallets and blockchains? Because truly mature infrastructure is ultimately not perceived by users. Protocols like TCP and content delivery networks are never consciously noticed— the deeper you go into the stack, the less need there is for people to constantly talk about it. If stablecoins and on-chain settlements are to truly serve the AI economy in the future, their ideal state is precisely to quietly disappear into the protocol layer.
This is actually very similar to many moments in internet history. What truly transforms an industry is often not the most compellingly marketed application, but rather a foundational standard that, at the right time, connects with real demand. At first, it may seem unremarkable—even dull—but once it integrates with the industry, something substantial soon grows from it.
402 already has this meaning now.
The pressure is on China's internet industry.
What this article truly wants to convey is not even the above.
All of the above are context; what truly matters is how China's internet should view this.
Many people’s first reaction here is perfectly normal: payment methods in mainland China and overseas have always been different. China has WeChat Pay and Alipay, while overseas relies on credit cards and PayPal—these differences are huge, yet the internet still thrived, and cross-border business was still possible. So why does it suddenly seem so significant when it comes to 402 and stablecoins?
This question is important and must be clearly explained.
Because if this point isn't clearly explained, the entire article becomes vague—as if the internet were rewritten simply because the protocol was written into HTTP and settlement assets were switched to stablecoins. The reality is far more complex. What will truly have a major impact is not the name of the protocol itself, nor how new the term "stablecoin" is, but rather how it begins to intersect with the internet’s pricing structures, settlement systems, and cross-border value flow mechanisms.
First, let’s explain why payment methods in China and overseas were different in the past, but the impact was not significant.
The reason is simple: over the past two decades, the mainstream internet, despite different payment methods, had essentially the same transaction structure.
Chinese users pay via Alipay and WeChat Pay, while overseas users use cards or link PayPal—fundamentally, all are initiating a purchase through the platform. The platform remains the organizer of the transaction, with payment being merely the final step. Over the years, the People’s Bank of China has repeatedly emphasized that China’s mobile payment system is mature—this is precisely what they mean: it is a highly efficient, highly localized platform-based payment ecosystem.
So in the past, the fact that "payment methods differed" was mostly about different checkout tools and experiences, but the mall itself remained the same mall. The platform simply needed to adapt to local payment methods. While this certainly affected conversion rates, costs, and user experience, it generally did not reshape the fundamental division of labor on the internet.
The 402 and stablecoin are not bumping into the checkout counter—they're bumping into the fee interface itself.
In the past, payments followed a sequence: first came the platform, then the checkout; first the service, then the payment. But if protocols like x402 are implemented, payment becomes an integral part of resource access. An interface can charge directly, a single browser call can charge directly, a model inference can charge directly, and a data request can charge directly. It’s no longer about opening an account, signing a contract, and then reloading—instead, each invocation is settled immediately.
Don’t underestimate this change. On the surface, it just looks like “more detailed fees,” but it will actually affect many things.
The most straightforward point is that the location of value capture will change.
Over the past two decades, China’s internet has been strong—strong in applications, strong in platform organization, strong in traffic operations, strong in localized payment ecosystems, and strong in delivering exceptional product experiences. Chinese companies excel at building business models centered around users and platforms.
But if the next round of value capture increasingly occurs at the level of automatic charging per capability invocation, the focus of competition will gradually shift from platforms and traffic to underlying capabilities, standard interfaces, settlement networks, and programmable fee systems.
In the past, Chinese companies competed over who could build better applications, better platforms, and better retain users within their ecosystems. In the future, part of this competition may shift to who can make their models, computing power, data, and APIs more easily and frequently invoked, instantly settled, and automatically integrated into global agent networks.
These are not the same set of skills.
The second change is that cross-border friction will be rewritten.
Today, Chinese companies going global—whether in software services, gaming, or e-commerce—are essentially adapting to each market individually: integrating local payment methods, partnering with local acquirers, establishing local merchant accounts, and using local settlement systems. In other words, cross-border internet operations in the past have largely followed a logic of "localized patchwork."
But what x402 and stablecoins aim to do is not to integrate a separate local payment system in each country—they aim to create a unified settlement layer that is closer to native internet infrastructure.
Why are stablecoins important here? Not because their name includes the word "coin," but because they function more like a neutral unit of account within machine networks than local payment tools in various countries. Coinbase explicitly defines x402 as a stablecoin for direct payment over HTTP; Stripe also clearly states that customers can pay globally using stablecoins, though currently, only primarily U.S.-based businesses can accept such payments.
This detail is actually critical.
What does this indicate? It shows that behind this seemingly "open" new payment protocol lies a strong geopolitical structure. While global agents can all make payments, the entities that can first承接, absorb, and interpret this settlement traffic are still primarily U.S. companies.
In other words, once a new payment protocol is tied to a new default settlement asset, it becomes not just a technical tool, but an instrument of industry allocation.
What does this mean for China’s internet industry? It means that if, in the future, the vast majority of machine-to-machine and interface-to-interface transactions default to running on this system, Chinese companies will no longer face the question of whether to integrate a payment channel—but a more pressing one: Will you become a charging node within this network, or merely a user node?
The fee node earns sustainable infrastructure income and holds a strategic position in the rules. What is gained through using a node is mostly application-layer business, which can be easily constrained by upstream interfaces and settlement rules.
The third change is the strategic issue of China's payment infrastructure, which will be further amplified by this event.
China's mobile payment system is undoubtedly powerful, and there's no denying this. However, the strength of China’s payment ecosystem lies primarily in the superior efficiency of its closed-loop systems between local platforms and local users. WeChat Pay and Alipay are exceptionally well-suited for consumer internet, lifestyle services, e-commerce retail, and platform-based transactions, but they were not inherently designed for global machine-to-machine or agent-to-agent programmatic value exchange.
That's why "in the past, different payment methods weren't an issue," but with "402 + stablecoins," things suddenly become much bigger.
Because that past difference was essentially just a variation in payment experience; now, this issue is beginning to touch the foundational settlement layer of the global machine economy.
So China may soon be forced to seriously consider a question: Do we have the capability to provide Chinese internet platforms, Chinese AI companies, and Chinese enterprises expanding overseas with a compliant, controllable, low-friction payment and settlement system that can be integrated into future agent networks?
This is no longer just a question of whether a payment tool is easy to use—it’s an issue of digital infrastructure.
The final change is that institutional boundaries will erupt before product changes.
This is also the area that most technical articles are least willing to elaborate on, yet it is precisely the most realistic one.
Today, the transaction logic of Alipay and WeChat Pay is clear in terms of roles, responsibilities, channels, and regulatory interfaces. It is clear who the platform is, who the merchant is, who the user is, and who the payment institution is—how funds flow and how anti-money laundering is conducted—all within a mature framework.
But once 402, stablecoins, and AI agents are combined, the situation becomes much more complex.
AI agents are not natural persons; they use programmatic authorization to continuously purchase resources, invoke services, and spend money; stablecoins are not localized, closed payment tools—they are inherently suited for cross-border movement; x402 is not merely a payment page within a platform, but rather aims to be embedded into resource access protocols.
So the immediate questions arise: Who authorizes? Who sets the budget? Who is accountable? Who handles customer identification and anti-money laundering? Who manages cross-border payments and foreign exchange boundaries? If the machine makes unauthorized purchases, repeats payments, or is exploited through an attack, who ultimately bears responsibility?
These issues won’t wait until the technology is fully mature—they often emerge as soon as commercialization begins to gain traction.
From the perspective of China’s internet industry, what makes 402 and stablecoins truly significant is not how novel they sound, but that for the first time, they bring together issues like “internet payment interfaces,” “default settlement assets,” “cross-border value flows,” and “machine payment liability boundaries”—problems that were previously scattered—onto the same table.
In the past, the difference between payment methods in China and overseas had less impact because they were merely different payment tools and did not alter the underlying structure of internet transactions.
The reason 402 + stablecoins may have a greater impact is that they are not merely altering the payment experience, but rather the future internet’s pricing, settlement, and industrial division of labor structures.
The former modifies the checkout action, while the latter may modify who collects payment, who settles transactions, and who serves as the default node.
These two levels are not on the same scale.
Don't stand on the sidelines of the game.
Looking back, the long-dormant 402 was revived not because Silicon Valley engineers suddenly felt nostalgic, nor because Web3 found a new slogan.
What it truly illustrates is that the old internet model—where humans click buttons, platforms charge monthly subscriptions, and settlements occur at month-end—still works today, but it’s increasingly unable to fully meet the demands of the new machine economy. AI isn’t here to simply replicate human payment methods; rather, it’s pushing the internet to finally implement the layer it has long neglected.
Seen solely from within the U.S. tech industry, this is a story about the reallocation of roles among protocols, payments, stablecoins, card networks, and cloud infrastructure.
But from a Chinese perspective, what truly matters is not that the old number "402" has finally found its purpose, but rather where China’s internet, China’s payment system, and China’s AI companies expanding overseas will stand on this new network when the global internet begins seriously building out its "layer of value flow."
Will you continue as a user of the application layer, or will you have the opportunity to help define fee interfaces, settlement standards, and infrastructure rules?
Will you become a charged node in the future global agent network, or simply act as a capability provider and caller?
Are you viewing this as just another piece of overseas tech news, or do you recognize that it reflects a fundamental shift in the underlying division of labor on the internet?
These questions still don't have final answers today.
But the direction has already begun to emerge.
At least one thing is clear now: the next major infrastructure shift on the internet may not begin with content or social media—it’s likely to start with billing and settlement.
Once this layer is truly rewritten, many issues that today appear to be merely product, payment, or international expansion challenges will ultimately become questions of industrial positioning.
In the past, the internet primarily addressed how information flows; now, some are seriously tackling how value can flow automatically. What China truly needs to understand is not whether to observe from the outside, but whether, when this network is built, we are inside it—or still standing outside.
Future payments, recommended courses
The arrival of the AI payment era is not an issue confined to any single step—each stage, from regulatory classification and licensing pathways to AML system construction, off-chain and on-chain risk controls, and compliant implementation in real business scenarios, directly determines whether AI can truly "step forward."
Leveraging Manqun Law Firm’s extensive expertise in Web3 legal practice, we are proud to introduce the “Crypto Payments Compliance: Exclusive Practical Workshop (Shenzhen Edition).”
April 18–19, Shenzhen, limited small-group session, multi-perspective approach: Legal + Compliance + Security.
Instructors come from Manqun Law Firm, Beosin, KUN Kun, COBO, Xinhuo Group, and seasoned professionals from licensed institutions.
Coverage: Regulatory pathways and licensing frameworks, compliance governance frameworks, anti-money laundering practices, on-chain security, and firsthand experience from licensed institutions.
If you're developing a product related to crypto payments or exploring the compliance path for AI agents in payment scenarios, scan the QR code to consult with us.
Source:Show originalDisclaimer: The information on this page may have been obtained from third parties and does not necessarily reflect the views or opinions of KuCoin. This content is provided for general informational purposes only, without any representation or warranty of any kind, nor shall it be construed as financial or investment advice. KuCoin shall not be liable for any errors or omissions, or for any outcomes resulting from the use of this information.Investments in digital assets can be risky. Please carefully evaluate the risks of a product and your risk tolerance based on your own financial circumstances. For more information, please refer to our Terms of Use and Risk Disclosure.This detailed match analysis covers key moments, player performances, and tactical insights.
Key Statistics
Possession: 55% - 45%
Shots on target: 6 - 3
Pass accuracy: 88% - 82%
Corners: 5 - 2
Player Ratings
- Home MVP: John Doe (9.2)
- Away MVP: James Smith (8.7)
Post-Match Analysis
The manager praised the team's resilience after coming from behind. "We showed great character," he said.
This win moves them to the top of the league table with 45 points.