Can traditional banks survive the neobank and fintech onslaught, or will they become regulated utilities
Will Your Bank Still Exist in Ten Years? What a Map of Banking's Future Actually Shows
Based on analysis of a 117-node, 419-edge knowledge graph exploring the structural dynamics of banking disruption, neobank competition, and regulatory capture.
The Big Question, Simply Put
Imagine your town has one big old grocery store that has been there for a hundred years. Then a bunch of shiny new food delivery apps show up and start undercutting it. Everyone asks: is the old grocery store finished?
That is roughly the question this knowledge graph tries to answer, but for banks. Traditional banks have been around for a long time. In the last decade, a wave of new companies — called neobanks and fintechs — showed up with slick apps, no fees, and promises to do everything better. So what does the map of this competition actually show?
The short answer the graph gives: the old grocery store is probably not finished. But it is going to get a lot bigger, a lot fewer, and a lot more powerful — while many of the delivery apps either fail quietly or end up turning into grocery stores themselves.
The Most Contested Prize: Your Checking Account
The single most important node in this entire graph — the thing everything else in the map connects to — is called “deposit franchise stickiness.” That is a fancy way of saying: once your paycheck lands in a bank account, you almost never move it.
Think about how annoying it would be to switch your bank. You would have to update your direct deposit, change all your automatic bill payments, get a new debit card number, and explain to your employer where to send your pay. Most people never do this. They stay put.
This inertia is enormously valuable. It means banks have cheap, stable funding. It also means that despite a decade of shiny competitor apps, most people still have their “real” money at a traditional bank.
The graph shows more than fifty separate connections flowing into and out of this one node. Every major threat in the map — from cryptocurrencies to government digital currencies to Gen Z avoiding branches — ultimately aims at this stickiness. And almost every major defense mechanism — from mortgage lock-in to credit card rewards programs to Zelle — works by reinforcing it.
The whole graph is, in a sense, a map of a single contest: can anyone actually pry your main account away from your bank?
Why the Big Banks Are Likely to Get Bigger
The graph identifies something the analysts call the “barbell outcome.” Picture a barbell at a gym: two heavy weights at opposite ends, and almost nothing in the middle.
That is the predicted shape of banking. A small number of enormous banks — the JPMorgans, the Bank of Americas, the Citigroups — at one end. A large number of tiny community banks serving local relationships at the other. And a hollowed-out middle tier, where regional banks of medium size face pressure from all directions and slowly disappear through mergers or failures.
What is striking about this finding is not just that the graph predicts it — it is that six separate and independent chains of cause-and-effect all point to the same outcome. Commercial real estate loans going bad. Mergers accelerating. Artificial intelligence giving advantages to whoever has the most data. Regulatory rules that are expensive enough to squeeze mid-sized banks but that big banks can afford to treat as a fixed cost. Each of these, on its own, points toward the same barbell shape. Together, they look less like a guess and more like a structural forecast.
The Feedback Loop You Cannot Easily Break
Here is one of the most important patterns in the map, explained simply.
Big banks have more customers, so they have more data. More data lets them build better artificial intelligence — better fraud detection, better loan approvals, better personalized offers. Better AI means they keep customers and attract new ones. More customers means more data. And so on.
This is a reinforcing loop: each turn of the wheel makes the wheel spin faster. The graph shows this clearly, and it also shows a second gear attached to the first: when big banks get AI advantages, they use them to acquire smaller banks. Acquiring smaller banks gives them more data. More data spins the AI wheel faster.
What is notable is that nothing in the graph breaks this cycle from the outside. There is no competitor, no regulation, no technology that the map shows interrupting the loop.
The Surprising Failure That Proved the Moat Was Real
Here is something the graph encodes that is genuinely counterintuitive.
Goldman Sachs — one of the most sophisticated, best-funded financial institutions on earth — tried to build a consumer bank from scratch. It was called Marcus. They spent years and billions of dollars on it. It failed. They shut it down.
The graph treats this not as a story about Goldman making mistakes, but as evidence that the deposit franchise moat is taller than anyone thought. If the most capable possible challenger could not breach the wall, then the wall is probably very high. The failure does not lower our confidence in the moat — it raises it.
The Neobank Trap: Disrupting Your Way Into Becoming a Bank
Most American neobanks — the Chimes, the Currents, the Varos — built their business on a specific trick. Every time you swipe your debit card, the merchant pays a small fee. Neobanks captured a larger share of that fee by working around traditional bank structures. This was clever, but it turned out not to be a business.
The problem: this model has thin margins, does not compound, and is easily regulated away. The graph shows a node called “Neobank Unit Economics Crisis” — meaning these companies are not making money — and then tracks what happens next.
What happens is not collapse. It is transformation. Some neobanks try to get actual bank charters and become, essentially, banks. Some get acquired by traditional banks at bargain prices. Some try to expand into a “superapp” — adding insurance, investments, and lending to their payment tools.
The most successful neobank in the world, Brazil’s Nubank, did something different from the start: it made loans. It used credit creation, which is the core thing that makes a bank a bank. The graph shows that Nubank’s model contradicts the neobank crisis, precisely because Nubank operates more like a bank than a neobank. The implication is uncomfortable: the most successful “disruptor” succeeded by not disrupting the fundamental model.
The Regulation Question: Who Writes the Rules Wins
One of the clearest structural patterns in the graph is that the outcome for traditional banks depends less on competition and more on who controls the rules.
The graph shows a loop called “regulatory capture,” which means: large banks become large enough that regulators and politicians find it easier to accommodate them than challenge them. Large banks use that accommodation to shape rules that are expensive for smaller competitors. Smaller competitors struggle. Some fail, some merge. The survivors are bigger. Now they have even more political weight. The cycle continues.
This is not a conspiracy theory — the graph presents it as a documented structural dynamic visible in the data, and shows it being exemplified by specific regulatory events: the rollback of open banking rules, the deregulatory posture of the 2025-2026 political environment, and the way capital requirements and compliance costs scale differently for large vs. small institutions.
The same dynamic appears in the stablecoin story. The GENIUS Act — proposed legislation that would create rules for stablecoins (a type of cryptocurrency designed to hold a stable value) — appears in the graph as a node that simultaneously constrains the threat from stablecoins and creates a regulatory moat that benefits whoever can afford to comply with it. Regulating a threat can also be a way of controlling who is allowed to compete in the space once the threat is legitimized.
What the Map Does Not Resolve
The graph is honest about several things it cannot settle.
The open banking question is genuinely unresolved. In the UK and Europe, rules require banks to let customers share their financial data easily with other services. This makes switching banks easier. In the US, similar rules (called Section 1033) were partially rolled back. The graph contains edges pointing in both directions — data portability as a threat to incumbent banks, and the rollback of data portability as a defense — and marks the outcome as dependent on political choices that have not yet been finalized.
The AI question is also unresolved in an interesting way. The graph shows both “AI concentrates advantage at large banks” and “AI tools will become cheap enough for small community banks to use too.” These are opposite predictions, and the graph holds both of them without declaring a winner. Which one dominates probably depends on how quickly AI tools become commodities — available to anyone for low cost — versus remaining the province of institutions with large proprietary datasets.
The Bottom Line: What the Structure of the Map Shows
A few things stand out as structurally robust findings — meaning they are not just one analyst’s opinion, but are encoded across many independent paths in the graph:
The deposit franchise is the actual moat. Everything else in banking competition is secondary. Whoever controls where people keep their everyday money controls the funding advantage, and the funding advantage is what makes banks banks.
The middle tier of banking is under the most pressure. Large banks have regulatory, data, and AI advantages. Small community banks have local relationship advantages. Mid-sized regional banks have fewer of each. The consolidation wave is most likely to run through them.
Neobank disruption is more likely to produce new banks than to replace banking. The dominant exit path from the neobank unit economics crisis is convergence — toward bank charters, toward acquisition, toward models that resemble traditional banking. The disruption trajectory bends back toward the structure it was disrupting.
Regulatory choices are at least as determinative as competitive dynamics. Who wins in banking is substantially determined by who writes the rules, not just who builds the better product. This is not unique to banking, but the graph shows it operating with unusual clarity here: the preservation paths for credit creation and deposit stickiness run through regulatory decisions far more than through competitive innovations.
The DeFi and cryptocurrency path is structurally different from other threats. Every other threat to traditional banking in the graph can be addressed through lobbying, regulatory capture, or acquisition. The graph identifies decentralized finance as the one path that does not run through a regulator who can be influenced. Whether that matters depends on whether DeFi remains small and marginal or grows to systemic scale — a question the graph marks as open.
The structural story the graph tells is not that traditional banks are safe and disruption has failed. It is that the most likely outcome of disruption is a banking system that looks like today’s system, but with fewer and larger institutions, with more of the competition routed into regulatory arenas, and with the survivors having absorbed many of the tools and some of the talent from the disruption wave.