Opinion
Column by Mary Wisniewski
Are 12-Year-Old Neobanks Too Old Now?

The founder of the 12-year-old Starling Bank, Anne Boden, believes it’s time to start another bank. So, she is. It’s called Lockwood and it’s a bold bet: The U.K. startup is planning a current account built with AI agents from the start and it will let agents pay and shop within the limits customers set.
She believes that the crop of neobanks of yesteryear – developed to ditch the branch for the smartphone – are now at a disadvantage in an AI agent era because of another design assumption: that humans are their principal users.
“Today's banks require customers to manually verify and authorise every individual payment initiated by an agent, defeating the purpose of autonomous financial assistants,” Boden wrote on Lockwood’s blog.
To Boden, a new generation of neobanks that uses machines as some of their builders, operators and users is required. So, Lockwood plans to get an electronic money institution license from the Financial Conduct Authority in 2027, and then launch the bank in full in 2028. So, it’s early days. Likewise, it’s early days for consumers using AI agents to pay for stuff.
Maybe you’re sitting there dismissing her view as an overreaction to something existing banks can, and already are, offering customers to some extent (ahem Chime and its AI personal assistant, Jade). And perhaps you think the idea is too early or too niche. Plenty of consumers and industry professionals do want final say on buy decisions, especially when those purchases are priced over a certain amount. I’m at least wondering about these things.
Boden anticipates these objections. “Yes, existing neobanks can add agent features to their technology, just as traditional banks eventually produced good mobile apps,” she wrote. “Supporting agents properly means more than adding another interface. It requires a bank to be able to identify each agent, understand who it represents, enforce a bounded mandate and retain evidence of every decision. Retrofitting those capabilities across existing customer journeys, while updating existing controls is possible, but it is a substantial undertaking.”
So she is building Lockwood with agents to do all of that from day one in a bid to help consumers tackle their admin tasks. She’s not the only one. Brett King, founder of mobile banking startup Moven and author of Bank 5.0, is planning to launch an AI-native bank called Baink that will offer accounts to consumers and agents. According to him, by 2030, customers of banks will be agents, not consumers. That speaks to something King told us in a recent Yours in Finance interview: That offering utility is what earns the right to be someone’s primary bank.
“That will increasingly be the primary metric for most people: How quickly can I get something done?” he said. “This is even going to accelerate with AI, where it's really an element of shifting to outcome-based financial services rather than product-based.”
It’s not just general agents like Meta’s Muse or OpenAI’s Dots or now Klarna’s new AI concierge tool versus niche agents offered by regulated banks that will compete. Boden and King are opening another door: Newcomers.
Starting from scratch, of course, are also substantial undertakings. And none of this means banks or neobanks are sitting around at a moment when the interface of banking is expected to morph or that they will lose.
Chime, a 14-year-old neobank in the U.S., sees the opportunity for the neobank and its agent to win. Not from blocking agents like some other big banks are doing right now but from helping consumers use general-purpose agents as well as Chime’s own agent.
Chime co-founder Ryan King is betting that there won’t be a single general-purpose agent that consumers trust to do everything. “An agent that books dinner reservations is VERY different from one that pays all your bills, just as a (human) travel agent is NOT your financial advisor,” Ryan King wrote on LinkedIn.
That is why he expects Jade to serve as “the front door” for any outside agents Chime customers use in the company’s race to keep winning primary accounts.
Years ago, I remember incumbent banks dismissing the neobanks, viewing their offerings as limited at best for consumers they didn’t want. The times have changed. Consider how SoFi, now a real bank, hit a milestone in cracking the top 50 on the list of U.S. banks with the most deposits in September. Or that the fastest-growing segment for Chime is people earning $75,000 a year or more.
Yet, Starling was once a newcomer and its founder is arguing the bank is too old. Chime, meanwhile, thinks 14 years of helping people overcome banking inertia is, in fact, an advantage.
It’s anyone's game and everyone playing will think it’s them.
I don’t know who wins. I wouldn’t bet against the incumbents or the neobanks. I also wouldn’t bet against the new guy or gal. And I would never bet against consumers' indifference. But it’s clear what’s required to compete in an agentic era. Offering products that improve people’s financial outcomes is the only chance to win consumers' business.
So, what do you think? Are the neobanks of yesteryear already behind or is a decade of earning trust their biggest advantage? Let me know!




