Look at your bank card. You'll likely see the blue and white "Visa" logo in the bottom right corner on the back or the bottom left on the front. This emblem is so common it's almost invisible. Yet, behind this unassuming symbol lies a "payment empire" processing an average of 900 million transactions daily and generating $20.1 billion in annual profit.
Visa does not issue cards, extend credit, or assume risk. It simply takes a tiny, almost imperceptible "toll" each time a transaction is completed. In this way, it has quietly built a market capitalization exceeding $600 billion, defining the rules of modern electronic payments. However, after running smoothly for over half a century, this "toll road" is encountering new challenges.
This story begins on a morning in 1958. In September of that year, in Fresno, California, approximately 60,000 households found an unstamped envelope in their mailboxes. Inside was a credit card with a $300 limit, requiring no application, no approval, and usable immediately upon signature. This financial experiment, known as the "Fresno Drop," was the brainchild of local banker Joseph Williams.
At the time, middle-class American families relied on cash, checks, or local store credit for purchases, with large payments requiring difficult negotiations with banks. The process was cumbersome, slow, and filled with restrictions. Williams realized the post-war middle class wasn't unwilling to spend; they found it too troublesome. He believed that providing simple, respectable, and instantly available credit would lead to more frequent and less burdensome spending—even a willingness to spend in advance.
The method to test this bold idea was straightforward: mail 60,000 unsecured credit cards directly to households. The result was disastrous. Over 20% of cardholders had no intention of repaying, and various crimes like forged signatures, organized fraud, and black-market card trafficking were rampant. This led to a loss exceeding $8 million in the first year, and Williams was subsequently fired.
Amid the chaos, U.S. financial institutions noticed a truth obscured by the disorder: despite rampant fraud, people were still using the cards frequently. They realized people were willing to spend; the real problem was not demand but the lack of rules. Financial institutions began tightening issuance standards and restructuring transaction processes. By 1961, the nearly defunct project miraculously turned a profit.
As the model proved successful, financial management authorized it to banks nationwide, forming a loose alliance called "BankAmericard." However, disorder resurfaced. Rapid expansion outpaced the formation of unified technical standards, with banks operating independently. Even basic risk defenses, like shared customer blacklists, were absent. This led to nationwide "criminal arbitrage," where fraudsters successful in California could use their cards unimpeded in New York, with banks powerless to stop them. Unsurprisingly, banks faced massive bad debts. The nascent national credit card network, yet to reap the benefits of scale, was again on the brink of collapse.
This uncontrolled experiment validated the huge demand for "frictionless payment" but exposed a fundamental problem in wild growth: without rules, even the best ideas cannot be realized. Dealing with bad debts became a major headache for U.S. banks.
In 1968, Dee Hock, an ordinary clerk at a small Seattle bank, was tasked with cleaning up this "bad debt" mess. In the elite world of finance, Hock, in his twenties with only an associate degree, was an unremarkable bank employee. While untangling the mess, he identified a core contradiction: every bank blamed system failure, yet none were willing to relinquish their own control. Hock realized the root of the credit card system's chaos was not technology or poor management but a structural deadlock: each bank wanted control over the payment "highway" but lacked the ability to manage the entire pipeline alone.
A turning point came during a heated BankAmericard alliance meeting. Hock, as a representative, proposed a bold idea: "Instead of complaining here, let's rewrite the rules ourselves." Surprisingly, the proposal was accepted. Following the principle that those who identify a problem should solve it, Hock presented a prepared plan. His idea was simple: establish a new organization owned by all member banks within the alliance—a "no boss, just rules" model.
This new organization would not issue cards, compete with banks for customers, or assume risk. Its sole purpose was to establish unified technical standards, risk controls, and settlement rules—a set of "traffic regulations" all banks must follow. This meant that while these banks were competitors in the market, on the payment "road," everyone had to drive by the same rules. In 1970, a new organization, "National BankAmericard Inc.," was born.
At the founding meeting, to resolve the contradiction between competition and cooperation among banks, Hock proposed his management philosophy, naming it "chaordic"—a blend of "chaos" and "order." He believed the most viable systems in nature are not centrally controlled but emerge from member interactions, like an ant colony operating efficiently without a queen ant directing it. The core of "chaordic" theory is that the best systems are not strictly controlled but allow order and chaos to collide and grow naturally within certain rules.
In other words, the new organization did not rely on a single central command but on members adhering to common rules and collaborating autonomously, making it more flexible and resilient. In 1976, this organization adopted the name that would become globally recognized: Visa. The word itself has no meaning but was chosen for its ease of pronunciation in most languages. From its inception, Visa was anchored to a fundamental principle: not to be a bank, but to be a pipeline.
Specifically, Visa does not issue cards, accept deposits, or engage in direct lending with customers. It merely acts as a neutral channel for fund and information flow between banks. This pure intermediary role—"no card issuance, no lending, no credit risk assumption"—became Visa's strongest armor against cyclical storms but also, due to its purity, the source of its fundamental challenge of being bypassed by new technologies.
So, how does a "pipeline" that assumes no risk make money? The secret lies behind every seemingly simple card swipe. Take a $100 cup of coffee: when you pay by card at a POS terminal, the transaction completes almost instantly. However, your money doesn't immediately enter the café's account. It first goes through your issuing bank, then Visa's network, followed by the acquiring institution, before finally reaching the merchant. Therefore, the café ultimately receives about $97 to $98, not the full $100. The few dollars that "evaporate" are known in the industry as the "merchant discount fee."
In each such transaction, this $2-$3 fee is typically distributed among three parties in set proportions: the issuing bank gets about 75%, the acquirer about 18%, and Visa takes a fixed cut of about 7%. For that $100 coffee, Visa earns roughly $0.30. This seems small, but consider that Visa processed 257.5 billion transactions in fiscal 2025. This means that from these seemingly minuscule per-transaction fees, it generated $40 billion in net revenue and $20.1 billion in net profit for fiscal 2025. Its net profit margin has long remained above 50%, roughly double Apple's net profit margin (about 26.9%) during the same period.
Crucially, this money is earned with remarkable ease. Since the issuing bank bears the cardholder's credit risk and the acquirer bears the merchant's operational risk, Visa remains detached and unaffected. Visa doesn't need to know if you're buying coffee or paying a gambling debt, nor does it care if the café goes bankrupt next month. Its task is to verify transaction compliance in milliseconds and then take its cut and exit.
Because of this purity, the business possesses incredible scalability. Visa could focus all its energy on building and expanding the "pipeline" itself, eventually permeating the capillaries of global commerce like water, electricity, and gas, becoming an omnipresent yet often overlooked foundational payment infrastructure. Often, users choose Visa not out of preference but because they have no alternative.
In fact, Visa's high stickiness with the payment system benefits from a "flywheel effect" operating for decades: the more merchants that accept it, the more cardholders rely on it; the more cardholders there are, the more merchants dare not refuse it. The interdependent, mutually reinforcing force between merchants and cardholders pushed Visa into a self-reinforcing growth loop. This pure pipeline model—"no card issuance, no lending, no risk assumption"—formed an extremely stable closed loop, becoming the foundation that kept Visa standing firm through industry storms.
The 2008 global financial crisis served as the ultimate test of its "three-no" model. When institutions like Lehman Brothers collapsed due to "toxic" assets, causing the credit system to crumble, Visa remained unscathed due to its pure neutrality. Even in the same year market confidence hit rock bottom, Visa went public with this clear model, achieving the largest IPO in U.S. history at the time. This revealed a profound pattern: when systemic trust collapses, the party that does not engage in speculation and only provides transparent infrastructure becomes the most trustworthy safe harbor in the storm.
Based on this trust, Visa continued to expand globally after its IPO. As of 2025, Visa covered over 150 million merchants worldwide, connected nearly 15,000 financial institutions, and, together with Mastercard, controlled over 80% of the global bank card market share. However, just as traffic flowed smoothly on this seemingly indestructible payment highway, new challengers quietly emerged. They are not trying to "rebuild" a road but to create a way to travel without needing this highway at all.
The new competitors come from a type of digital currency called "stablecoins." What exactly is a stablecoin? Essentially, you can think of it as a "digital dollar" operating on the internet's blockchain. It is pegged 1:1 to the real U.S. dollar, meaning you can theoretically exchange 1 stablecoin for $1 cash at any time. Its core design allows it to completely bypass the traditional currency exchange system, showcasing a brutally simple advantage: it supports global instant transfers at near-zero cost and entirely circumvents the old network of banks and credit card organizations.
Put simply, if Visa built an efficient global toll road, stablecoins enable a form of "instant value transfer" that nearly ignores geographical and financial borders. The efficiency and cost transformation brought by this "transfer" is staggering. For example, a $100 cross-border remittance via traditional wire transfer takes days and costs tens of dollars in fees; using the Visa network costs the merchant a 1.5% to 3.5% fee; using stablecoins takes less than a minute and can cost less than 1 cent.
More critically for Visa, the emergence of stablecoins not only represents cheaper payment costs but challenges the very logic of Visa's existence. The real threat of stablecoins is not that they are "cheaper" but that they render the entire fee structure unnecessary. In the stablecoin transaction model, the core chain of issuing bank, card network, and acquiring bank that has operated for decades is entirely bypassed. Money "jumps" directly from the buyer's digital wallet to the seller's digital wallet, with no "middlemen taking a cut," and thus no opportunity to collect any "toll."
This isn't just about "saving money"; it makes Visa's meticulously maintained toll road suddenly seem irrelevant. In 2025, U.S. businesses paid over $200 billion in swipe fees. Faced with such a massive cost burden and the direct pursuit of profit, merchants naturally favor stablecoin settlement. By the end of 2025, the global stablecoin market cap approached $300 billion, with predictions suggesting it could surpass $2 trillion by 2028.
Beyond the pressure from stablecoins, another, perhaps more formidable, transformation simultaneously threatens Visa. With the rapid development of AI technology, more AI agents are taking over human payment decisions. Imagine this scenario: after you instruct your AI assistant about a trip, you wake up the next day to find it has already booked your round-trip flights and hotel. The entire payment process—the AI assistant automatically comparing prices, completing the payment, and syncing the itinerary—requires no action from you. This means the act of payment through the banking system, along with the choice of payment method, the confirmation click, and even the perception of the "payment" moment, disappears.
This "autonomous payment" is no longer science fiction; it's already a reality in some cutting-edge applications. For over half a century, Visa's existence has been rooted in human payment habits and psychological trust in the brand. We rely on it out of familiarity and choose it out of trust. However, AI is completely immune to this trust system built on "human nature." It has no habits, recognizes no brands, and operates on a cold, efficient decision logic: choose the fastest and cheapest path, provided it's secure. In other words, once AI calculates that the stablecoin channel cost is far lower than Visa's, the switch will be made automatically in milliseconds, and you might never know.
This means the thought "I want to use Visa"—born from human habit and emotion—will never appear in the algorithm's decision tree. Stablecoins provide a transmission pipeline at near-zero cost, while AI agents act as a cold, tireless rational navigation system. When a free shortcut meets an automatic navigation system that always chooses the optimal route, the outcome is self-evident: Visa's "toll road" faces not just a competitor next door, but a scenario where all traffic is silently redirected onto a new road that completely bypasses the toll booth.
Facing a highway about to be circumvented, Visa must contemplate: how should this road proceed in the future? In 2025, a symbolic event occurred: Silicon Valley tech companies approached Visa, asking it to open its global settlement network to help process stablecoin transactions like USDC. This created an interesting situation: a payment giant built on "collecting tolls" was invited to participate in building a new road designed for "free passage."
Thus, Visa embarked on a profound self-reinvention. It is no longer content with being just the pipeline for card payments. Instead, it announced its intention to become the underlying platform for all payment methods—the "network of networks" connecting all forms of payment. From Visa's perspective, since it cannot stop people from building new roads, it's better to make itself the indispensable gas station and checkpoint on all roads. Simply put, Visa's new goal is: no matter what form the funds flow in, they need to pass through its system for security verification.
To achieve this strategic goal, Visa is rapidly acting on two fronts simultaneously. On one hand, it continuously expands its network boundaries outward. For example, its $1 billion acquisition of Brazilian payment company Pismo aimed to secure a foothold in high-growth emerging markets, capturing future digital payment traffic inlets. On the other hand, it is continuously reinforcing its core moat—the security system.
Over the past five years, Visa has invested over $13 billion to build an AI-powered risk control "brain" continuously "fed" by data from trillions of global transactions. This system, with its precise real-time analysis capabilities, can identify fraudulent behavior at the millisecond level, successfully intercepting $40 billion in risky transactions in 2024 alone. This is perhaps the key factor that keeps Visa ahead of competitors: technology can be caught up with, but the "risk intuition" honed over half a century of global transactions—an almost instinctive sense—cannot be easily replicated. It's not just algorithms and data; it's the "muscle memory" formed from trillions of transactions, the "conditioned reflex" developed through countless attacks and defenses.
However, all of Visa's transformation efforts essentially declare one thing: the digital age still needs its "trust cornerstone." But for this statement to hold true, a prerequisite is needed: players in the new era are still willing to acknowledge and connect to this "cornerstone" provided by a specific company. The underlying philosophy of stablecoins is precisely the opposite. Its design初衷 is to use the certainty built by mathematics and code to replace any human credit intermediary. On the other hand, the trust in blockchain built by AI stems from preset algorithms and automatically executed rules, not the credit endorsement of any institution.
Therefore, when Visa extends its hand to the stablecoin or AI system, saying, "I can provide more secure trust," it might only receive a polite but firm reply: "There is no place for you in our design blueprint." And this is the core of the problem. The real challenge is not that Visa is not strong enough, but that its role as a "trust intermediary" is gradually losing its foundational reason for existence within the technical logic of the new world.
History here forms a cycle full of tension. Back then, Dee Hock's wisdom of "no boss, just rules" succeeded because all banks implicitly agreed they needed a recognized referee. But in the new arena where algorithms and code autonomously set the rules, Visa, as the "referee" itself, may no longer be needed. For this fundamental question, Visa may not have found an answer yet. But it has no way back; it must go all out to meet the new challenges. Because time is running out.
Comments