Over the past month, Nubank has been making headlines, and it looks increasingly likely that the Brazilian digital banking giant has ambitions beyond Latin America.
In September 2026, Nubank moved closer to entering the US retail banking market through a partnership with Lead Bank. With around 140 million customers globally, it is already the world’s largest digital banking platform by customer base, with roughly twice as many customers as Revolut.
That’s an impressive starting point but cracking the US is another matter.
Nubank has also received conditional approval from the OCC for a US national bank charter, with the aim of securing full approval by 2027. A full charter could give the company greater control over its US operations and a stronger foundation for competing in one of the world’s most lucrative banking markets.
Then came the Monzo rumours.
Reports suggested Nubank could acquire Monzo as part of its European expansion strategy. For Nubank, that would mean an established foothold in the UK. For Monzo, it could bring the financial backing and international reach to revisit its own US ambitions, where earlier expansion efforts struggled to gain traction.
There’s an interesting irony here. Monzo and Revolut were both founded in 2015, with Monzo getting started around five months earlier. Monzo also secured its UK banking licence years before Revolut. Yet Revolut has since built a much broader international business, while Monzo remains more closely tied to its home market.
Could Nubank be the international heavyweight Monzo needs? Perhaps. But Nubank was quick to dismiss the acquisition speculation, saying it had no plans to buy the British digital bank.
I can understand why.
Nubank has built its customer base by focusing on markets it knows. Buying its way into Europe through Monzo might accelerate expansion, but it would also mean taking on another company’s strategy, regulatory challenges and growth ambitions. An acquisition brings scale, but it doesn’t automatically bring success, particularly when Nubank is already working to establish itself in the US.
So where does that leave Nubank?
It has the customer base, the brand recognition and the ambition. But international expansion and replicating success are two very different things.
The US is a crowded market, with established banks, digital challengers and payment companies competing for customers. Acquiring them can be expensive. Getting people to sign up is one thing; convincing them to move their money, make Nubank their primary banking provider and stay for the long term is much harder.
Nubank has already proved it can scale. What it hasn’t yet proved is that its model can travel successfully into a market as competitive and complex as the US.
That’s the real test. Cracking America isn’t simply about adding millions of customers. It’s about turning that growth into a sustainable, profitable business. And 140 million customers elsewhere won’t guarantee that happens.
Jamie Dimon: Get Out of the Way? The Battle for Agentic Payments
I am thoroughly enjoying all the chatter around AI agents and payments. The industry is imagining a world where agents can shop, compare prices, manage subscriptions and make payments on our behalf. It sounds convenient, and the commercial opportunity is enormous.
But there’s a question that deserves just as much attention as the technology itself: when an AI agent gets something wrong, who is liable?
The concern isn’t just that an agent might go off script or be manipulated into making an unauthorised transaction. It’s that responsibility could become increasingly difficult to pin down when the customer, bank, payment provider and company building the agent are all involved.
The US has spent years building safeguards around customer data, privacy, fraud prevention and access to financial accounts. Banks invest heavily in identity verification, transaction monitoring and compliance. Now we’re talking about giving AI agents the ability to act on a customer’s behalf.
The potential is hard to ignore. An agent could compare financial products, negotiate bills, manage subscriptions and find the best way to pay for something. Customers might rarely need to open a banking app.
But convenience comes with a catch. If an agent can access your account and initiate payments, whose interests is it serving? The customer, the bank, the technology company or the merchant trying to make a sale? And if the agent makes a mistake, who picks up the bill?
Banks cannot simply blame the algorithm. Equally, technology companies cannot expect banks to absorb every loss while they control the customer interface.
Banks Are Building Their Own. Is That the Right Move?
Banks are understandably keen to develop their own AI agents. They already have the customer relationships, financial data, regulatory infrastructure and, crucially, access to the money.
Building their own agents gives them greater control over customer information and what those agents can do. It also reduces the risk of technology companies inserting themselves between banks and their customers.
But is building everything in-house really the right strategy?
Banks understand risk, regulation and financial services. That doesn’t automatically make them the best at building consumer AI products or creating the agent ecosystems customers will want to use.
There’s also the risk of fragmentation. If every bank develops its own closed system, we could end up with dozens of competing agents that struggle to work across banks, merchants and payment networks. That hardly sounds like the seamless experience the industry is promising.
And let’s be honest, banks have spent years trying to make their apps easier to use. Now they’re being asked to build systems that understand what customers want, make decisions and potentially move money with limited human intervention.
That’s quite a leap.
The alternative isn’t necessarily better. If independent technology companies become the main interface between customers and their banks, financial institutions risk being pushed into the background. They would still provide the accounts and process the transactions, but someone else could control the customer relationship.
For banks, that’s a serious concern.
The Real Battle Is Over Control
Imagine asking an AI assistant to find the best credit card, compare savings accounts or manage your bills. Rather than opening several banking apps, you tell the agent what you want and let it do the work.
The agent becomes the starting point for the decision. It could influence which financial product you choose, which bank you use and how you pay. The bank might still provide the underlying service, but it could have far less influence over the customer’s choice.
That’s a significant shift in the balance of power.
Payment networks, merchants and technology companies also have a stake in this new layer of commerce. Whoever controls the agent could influence where transactions happen, how payments are routed and which providers are selected.
For years, banks and fintechs have competed to own the customer experience. Agentic AI could put that relationship in the hands of a completely different set of players.
This is no longer just a conversation about chatbots. It’s about who gets to sit between the customer and their money, and who captures the value that comes with it.
The Liability Question
Before agentic payments become mainstream, the industry needs clearer answers on responsibility and control.
Who authorises a payment? Can an agent be restricted to a particular amount, merchant or purpose? What happens if it is compromised? And who is responsible when a transaction goes wrong?
These questions become more complicated when an agent operates across several providers, each with its own systems and obligations.
Existing payment protections still matter, but their application to agent-driven transactions needs careful consideration. Giving an agent permission to make payments doesn’t necessarily mean the customer intended to authorise every transaction it might initiate.
The industry needs to establish clear rules around consent, permissions and responsibility before customers can be expected to trust agents with increasingly consequential financial decisions.
Because if consumers don’t know who is responsible when something goes wrong, convenience will only take the industry so far.
The Next Battle in Payments
I can see why banks want to build their own agents. They have too much at stake to let someone else take control of the customer relationship without a fight.
But building everything in-house may not be the answer either. Agentic commerce will need to work across banks, merchants, payment networks and technology providers. Some level of cooperation will be unavoidable.
The challenge is finding the balance between openness and control: giving agents enough access to be useful without compromising customer security or leaving responsibility unclear.
Financial services have spent years trying to make payments faster, simpler and more secure. Agentic AI could take convenience to another level, but it also raises new questions about consent, accountability and who ultimately controls the transaction.
Banks shouldn’t ignore the opportunity. But neither should they assume that building their own agents will automatically protect their position.
Because the next battle in payments may not be about who moves the money.
It may be about who decides where the money goes.
The post Can Nubank Really Crack the US Market? appeared first on The Fintech Times.
