Money 20/20 USA 2026: Ten Lessons To Shape Bank Strategy

Money20/20 USA wrapped up in Las Vegas on October 21, and if there was a single theme, it was the fact that financial services industry is being rebuilt around artificial intelligence, digital money, and new business models. In this article, we explore these trends and explain how it might impact banking soon.

The technology part is interesting. However, the potential shift in banking economics is far more important. Money 20/20 brought ideas that are harbingers of business model change.

Usually, Money 20/20 highlights how fintech companies are disrupting banks. This year, the conversation increasingly centered on how technology might change the fundamental roles banks play in moving money, managing customer relationships, and creating value. It was not the technology that was most interesting at the conference, it was the potential shift in economics that caused pause.

Here are 10 lessons community and regional bankers should take from Money20/20 2026.

10. Blocking Agents

There was lots of talk about how banks can block agents like Muse and others on their website. The answer is that there are about a dozen products, the most popular being Lexis Nexis ThreatMatrix, that track customer behavior. It stores a profile of the speed and pattern of clicks, scrolls, mouse movements and typing. Whenever the score gets more on the “automated” side of the spectrum, a bank can then kill the browser instances forcing customers to log back in without their agent.

The banking implication: This type of fraud control takes some testing and usability review as it requires the bank to manage the level of score and customer experience. Banks that are interested in block agents should look at several different products and gain experience with how to implement the tool.

9. Forget the Human-in-the-Loop AI Construct

If you are still designing agents to keep a “human-in-the-loop,” you are doing it wrong. The goal of AI is to automate processes at the enterprise level. In production, when you keep a human in the loop, you allow AI to process more signals so more humans can process at human speed. By design, you end up with a flawed process that moves outcomes only incrementally faster.

Take cybersecurity as an example. The bad actors don’t have a human in the loop. By keeping a human in the loop, you will lose that battle every time.

The better design is to have a “human-on-the-loop” to monitor the output, while other agents monitor quality control. Removing the human requirement as a design principle forces better process design and coding to reduce risk. If you keep the human in the loop, you will assume the human will still handle edge cases instead of programming to reduce edge-case risk.

The banking implications: Set out to design processes with a human in the loop only when you must have human judgement that cannot be automated.

8. Trust, Identity, and Security Will Become More Important in the Future

While trust and security are important now, in this age where every process is AI-enable and many processes are agentic-driven, new forms of trust, security and identity are required, and banks are in the perfect position to play a central role.

When AI moves from answering questions to moving money, accessing customer data, and directing other agents, banks must verify not just who the customer is, but which agent represents them, what authority it has, and whether each action remains consistent with the customer’s intent. NIST’s current work emphasizes that model guardrails alone are insufficient: agents need distinct identities, narrowly scoped permissions, and accountable access rather than shared customer credentials.

Meanwhile, Mastercard’s agentic trust services and Visa’s business-payment pilot demonstrate how identity, intent, transaction controls, and real-time risk signals are becoming part of payment infrastructure itself. For banks, the opportunity is to become the trusted infrastructure for delegated financial activity. Banks can play a role in connecting verified customers to authorized agents, enforcing limits, protecting data, enabling rapid revocation, and documenting responsibility when something goes wrong. That position must be earned through security and governance that operate at machine speed. The competitive advantage will belong less to banks with the most autonomous agents than to those customers who can confidently authorize them to act on their behalf.

7. Agentic AI Is Moving From Experimentation to Execution

The industry has spent two years discussing generative AI. Money20/20 demonstrated that the conversation has advanced toward agentic AI, systems that can take action rather than simply generate information.

Summarizing a credit memorandum saves an employee time. An AI agent can collect borrower information, analyze financial statements, request missing documents, and prepare a credit recommendation changes the bank’s operating model. Money 20/20 drove home the fact that many banks are thinking too small in their application of AI.

Huntington Bank offered an especially interesting example. On October 19, it announced The 1866 Studio, developed with Google Cloud, to help small businesses manage financial and operational activities through AI-enabled workflows. The application goes well beyond putting a chatbot on a banking website and helps small business elevate their business using AI from a trusted source.

The banking implication: Banks should stop measuring AI success primarily by employee productivity or the number of pilots launched. The more important metrics are customer acquisition cost, operating expense per account, credit decision cycle time, and incremental revenue.

A bank that saves employees three hours a week has improved productivity. A bank that uses AI to make its customers more successful may create an entirely new competitive advantage.

6. Tokenized Deposits and Stablecoins Are The Foundation of Every Bank’s Digital Asset Strategy

Stablecoins and tokenized deposits were prominent throughout the conference. Most of the major bank consortiums were present to include the Cari Network.

The official agenda included sessions on stablecoin infrastructure, programmable money, digital-dollar control, and selecting appropriate token structures. The message centered less on customers holding cryptocurrency as it was in year’s past, and more on the impact of commercial payment infrastructure.

A variety of use cases were highlighted where traction is already present that largely settled around cross-border payments, store of value, escrowed funds and payment automation.

For bankers, the important issue is deposit ownership. Consider a commercial customer moving $10 million into a payment platform that uses stablecoins to settle supplier invoices. If the bank simply facilitates the initial movement of funds, it may lose a valuable operating balance, associated payment fees, and visibility into the customer’s financial activity.

Alternatively, a bank might offer tokenized deposits, custody and settlement services, or interoperable payment capabilities that help preserve the relationship. These are different business models, and banks need to figure out what digital asset strategy to pursue.

The banking implication: Every bank should understand how digital-money infrastructure could affect deposit retention, cost of funds, correspondent relationships, and treasury management revenues.

The question is no longer whether stablecoins will replace bank deposits. It is where customer balances and transaction economics will reside as digital-money adoption evolves and more real-world assets become tokenized. The more assets become tokenized, the more of a need for banks to have a digital wallet for themselves and their customers and the more they need access to both a tokenized network and the instant payment rails to allow 24/7/365 settlement.

5. The Next Battle in Banking Will Be Identity, Not Payments

One of the more consequential developments at Money20/20 concerned how financial institutions authenticate AI agents.

On October 18, Proof and Circle presented technology designed to let agents establish whom they represent and what they are authorized to do when making USDC payments. Conference sessions also examined trust and accountability when AI agents initiate purchases.

This creates a problem that traditional payment authentication was not designed to solve.

Today, banks generally authenticate people and businesses. Now, banks need to authenticate agents acting on behalf of those customers.

Suppose a business authorizes an AI agent to manage accounts payable. Can the agent initiate a $75,000 payment? Can it change beneficiaries? Who is liable if its instructions are compromised? How quickly can its authority be revoked?

A payment might be cryptographically valid and still be fraudulent or unauthorized.

The banking implication: Banks need an authorization architecture for delegated digital identities, including agent credentials, transaction limits, permitted counterparties, revocation, monitoring, and auditable consent.

The future of payment security is not simply confirming that a transaction is authentic. It is confirming that the transaction was authorized by the right party, for the right purpose, within the right boundaries.

4. Payment Hubs Are Driving Faster Payment Adoption

Money 20/20 highlighted the frustration many banks are having with instant payment adoption. The difference between a successful bank and a bank that is not reaching their instant payment goals revolves around if a bank has a payment hub and how they treat instant payments. Those banks with a payments hub and pricing to underscore that instant payments should be the primary payment channel, are the ones that are gaining traction.

Another important theme was the growing complexity of the payment ecosystem. Banks are already managing ACH, wires, cards, RTP, and FedNow. Stablecoins and tokenized deposits introduce additional ways to move value. The mistake has been assuming that one rail will replace the others with out intent on the bank’s part.

Standard Charter’s observation that much of the industry’s progress has concentrated on improving payment rails rather than the entire payment experience. The strategic opportunity is payment orchestration: selecting the appropriate payment method based on speed, cost, liquidity, risk, and customer requirements.

The banking implication: A modern payments hub may be more strategically valuable than ownership of any particular rail. Banks should invest in intelligent routing, transparent payment status, exception management, and integrated reconciliation. These capabilities can also strengthen commercial deposits by embedding the bank more deeply into customers’ financial workflows.

Commercial customers are evolving into an always reconciled position where there is no monthly close, AI is currently making this a reality. An always reconciled company requires instant payments. Banks without instant payments and a payment hub will increasingly find their treasury management platform at a disadvantage.

3. Fraud Prevention Must Move At The Speed of Automation

AI introduces another uncomfortable reality: criminals can automate, personalize, and scale attacks using some of the same technologies financial institutions are adopting. Money 20/20 highlighted synthetic identities, deepfakes, cybersecurity, and fraud in agent-driven commerce where existing fraud programs struggle to address emerging payment models and AI-generated identities.

For community banks, this creates an economic problem as well as a security problem.

When payment execution becomes instantaneous, controls applied after settlement may have limited preventive value. And as customers increasingly delegate activities to AI agents, conventional identity and device signals may become insufficient.

Banks need to evaluate behavior, account history, counterparties, transaction context, and delegated authority before a payment is released.

The banking implication: Fraud management cannot remain a collection of disconnected tools and manual reviews. The objective should be risk-based authorization that operates within the payment workflow, supported by real-time intelligence, customer controls, and well-defined escalation procedures.

Faster payments coupled with AI without faster risk decisions simply accelerate the consequences of fraud.

2. The Bank’s Greatest Competitive Threat Is Losing The Distribution Advantage

Perhaps the most strategically important development wasn’t around tokenization or payments.

The hot debate at the conference was around what to do with third-party agents like Muse, Instinct, Dots and others. Do banks block all agents, develop their own like Huntington, JP Morgan Chase and others, or take a hybrid position?

For decades, banks have competed through branches, relationship managers, digital banking, and product pricing. Now, customers increasingly interact with agents that help decide where to borrow, how to move money, and which financial products to use?

The bank may still provide the regulated deposit account or loan. However, another platform could influence the customer relationship, determine product selection, and capture the valuable data generated by daily financial activity.

The conference highlighted the potential impact on liquidity, pricing, capital and bank performance when agents are the advisors. We highlighted our Agentic Stress Test last week where every bank can see the potential impact of this trend.

The bottom line here is that banks risks becoming an interchangeable financial utility.

This is especially relevant to community banks, whose competitive differentiation has traditionally depended on local expertise and personal relationships.

The Banking Implication: Banks need to make their products and expertise accessible through emerging AI-enabled channels while developing their own customer-facing intelligence.

That could include APIs for authorized account access, AI-enabled business financial management, contextual deposit recommendations, and proactive treasury advice.

The strategic objective should be to preserve relevance when the customer no longer starts every financial interaction inside the bank’s application.

1. Execution, Not Innovation Theater, Will Determine The Winners

Banks that went out to enable real time payments receive have now become a stark example of innovation theater. Many banks, to their detriment, checked the box and moved on. Unfortunately, so has the rest of the world and increasing speed.

While some of the agent technology remains more hype than reality, it has a large enough impact to be considered for strategy today. There is clearly a lack of specificity surrounding agentic commerce and agentic payments. However, there is a clear demand for operational efficiency and reliable payment infrastructure.

Financial services remains an unforgiving business. While a compelling demonstration does not establish a profitable business model, it does underscore the need for every bank to get educated and monitor the situation.

The banking implication: One critical discipline is for bank boards and executive team to understand what needs to be true before they take action and invest capital. This will at least help banks develop a framework for action.

If SouthState can demonstrate how you can grow deposits by 10% by using tokenized deposits would that motivate your bank to experiment? Does it require 20%? Or do you have to lose 20% of your deposits before you take action?

That discussion is critical to have now around tokenization, AI and agentic commerce.

Before investing in any new technology, management should also insist on a quantifiable business case with clear customer use cases.

Community banks shouldn’t attempt to replicate every fintech announcement. Their advantage comes from combining trusted customer relationships, access to deposits, disciplined risk management, and technology that solves specific commercial problems.

What problem does it solve? Which customers will adopt it? What revenue or deposit value will it create? What existing activity will it cannibalize? What new risks will the bank assume?

Innovation without economic discipline is simply a more expensive form of experimentation.

Putting These Lessons Into Action

Next year, Money 20/20 USA is moving locations and dates. Look for it at the larger Las Vegas Convention Center (LVCC) and Resorts World on October 19-21, 2027, which is a Monday through Thursday.

Money 20/20 2026 illustrated a financial industry increasingly focused on the practical deployment of technologies that have been discussed for years.

AI agents are moving into operational workflows. Digital money is expanding payment and settlement options. New platforms are challenging traditional distribution models. Fraud and identity management are becoming more complex.

The opportunity for banks is substantial, but so is the risk of strategic complacency.

Community banks don’t need to become technology companies. They need to become better at using technology to expand their economic advantage.

The real threat isn’t that technology will replace banks. It’s that technology may increasingly own the customer relationship while banks retain the balance sheet, compliance burden, and operational risk.

That is the lesson bankers should take home from Las Vegas.