Will AI Replace Bank Tellers? The Branch Banking Transformation

If you are a bank teller, the numbers are stark: AI exposure at 76% and automation risk at 72% in 2025, up from 55% and 52% just two years ago. These are among the highest figures we track across all financial occupations, and they reflect a transformation that has been underway for decades but is now accelerating dramatically. [Fact] No other front-line financial role faces a steeper exposure curve in our dataset.

The ATM started this story in the 1970s, online banking continued it in the 2000s, mobile banking pushed it further in the 2010s, and AI is now writing what may be the final chapter of traditional teller work. Each wave reduced the count of tasks that required a human at a counter, and each wave was followed by branch consolidation. AI is not a different kind of wave — it is the largest one yet, because it removes not just the routine transaction but the routine question, the routine identification, and the routine recommendation that used to anchor the teller role.

The Accelerating Decline of Traditional Teller Tasks

Cash handling and basic transactions — deposits, withdrawals, check cashing — have been migrating to self-service channels for years. AI has accelerated this by making mobile check deposit more reliable, enabling conversational AI banking assistants that guide customers through transactions, and powering smart ATMs that can handle increasingly complex operations. Modern self-service kiosks dispense and accept dozens of denominations, recycle cash, perform identity verification with biometric checks, and handle deposits, withdrawals, transfers, currency exchange, and even loan applications. [Claim] In many high-traffic branches, the kiosk has replaced four or five of the five traditional teller windows, leaving one universal banker as the human touchpoint.

Account inquiries are now handled overwhelmingly through digital channels. AI chatbots can answer questions about balances, recent transactions, fees, and account features with accuracy that matches or exceeds what most tellers can provide. Banks report that 70-80% of routine customer inquiries are now resolved through AI-powered digital channels. The chatbot of 2026 is not the menu-driven assistant of 2020 — it can pull live account data, explain a pending charge, dispute a transaction, schedule a transfer, and escalate to a human only when the conversation enters genuinely novel territory. Each successful chatbot resolution is one fewer reason for a customer to walk into a branch.

Identity verification, once a core teller skill, is increasingly automated through biometric authentication, device recognition, and behavioral analysis. AI systems can verify customer identity more quickly and often more accurately than visual ID checking by a teller. Liveness detection on a phone camera, voice-print analysis on a phone call, and continuous behavioral biometrics across web sessions mean the bank often knows it is dealing with the right person before any human even joins the conversation. The teller's traditional role as the gatekeeper against impersonation is becoming a back-up rather than a primary control.

Fraud detection at the transaction level has moved from teller judgment to algorithmic monitoring. AI systems that analyze transaction patterns in real-time catch suspicious activity that individual tellers would never spot. A teller might recognize an unusually large cash withdrawal, but the model has already compared that withdrawal against the customer's two-year history, similar customers in the same ZIP code, recent fraud trends in the region, and the network of beneficiaries the customer has interacted with. Tellers still flag in-person red flags — apparent coercion, signs of elder financial abuse, scripted answers from someone reading a fraud playbook — but the routine fraud-spotting that filled training manuals a decade ago is now machine work.

Branch traffic patterns reflect all of this. [Estimate] U.S. branch visits per customer have declined roughly 5-7% per year for most of the past decade, while mobile banking sessions per customer have increased by double-digit rates. The transaction mix in branches has shifted from majority deposits and withdrawals to majority advisory, account opening, complex services, and loan inquiries — work that one universal banker can often handle better than three traditional tellers.

What Keeps Some Tellers Employed

Complex service situations still benefit from human interaction. When a customer needs to resolve a dispute, set up a trust account, handle an estate matter, or navigate a complex wire transfer to an international destination, the experienced teller or banker provides guidance that self-service channels cannot match. Estate accounts involve death certificates, probate paperwork, beneficiary designations, and emotional customers — a combination that no chatbot handles gracefully. International wires invite questions about correspondent banks, sanctions screening, currency conversion timing, and tax reporting that benefit from someone who has seen these transactions before.

Relationship building and sales referrals represent the branch's primary value proposition for banks today. The teller who recognizes that a customer with growing deposits might benefit from an investment referral, or that a small business customer might need a line of credit, creates value that AI cannot easily replicate. Many banks have repositioned tellers as "universal bankers" who combine transaction processing with advisory and sales functions, and the most successful universal bankers generate measurable revenue from product cross-sell rather than just processing transactions. [Claim] Branch networks survive at all in 2026 mainly because of the deposit-gathering and small-business relationships they enable, not because of the transaction counts they process.

Elderly and underbanked customers often prefer in-person banking. Digital adoption varies dramatically by demographic, and branches continue to serve customers who are unable or unwilling to use digital channels. Serving these customers with patience and dignity is human work. Elder financial abuse — increasingly perpetrated through romance scams, grandparent scams, and tech-support scams — often surfaces at the teller window when a confused customer requests an unusual cash withdrawal or wire transfer. Trained tellers who can recognize the patterns of coercion and gently intervene have prevented enormous losses that no algorithm could stop, because the algorithm sees a legitimate customer authorizing a legitimate transaction.

Cash-heavy small businesses, immigrant communities sending remittances, and customers with complex multi-account households all continue to use branches at higher rates than the average. Banks that abandon these segments lose deposit market share, which is why even aggressive digital-first banks maintain at least some branch footprint in priority markets. The teller who serves these communities with linguistic skill, cultural familiarity, and personal trust is genuinely hard to replace.

But these remaining functions cannot sustain the current number of teller positions. The Bureau of Labor Statistics projects an -11% decline in teller employment through 2034, which is actually conservative given the pace of AI advancement. Major banks have publicly announced branch closure programs measured in the thousands of locations. [Fact] Several U.S. regional banks have closed 15-25% of their branches in the past five years, and the largest banks have telegraphed continued consolidation.

The 2028 Outlook

Projections show AI exposure reaching approximately 82% by 2028, with automation risk at 78%. Branch networks will continue to shrink, and remaining branches will operate with fewer tellers handling more advisory and complex service work. The pure transaction-processing teller role is ending. The "universal banker" model — where a single person handles transactions, account opening, basic lending, and product referrals — will be the dominant survival path inside branches, and the count of those positions will still be well below the count of traditional teller positions today.

Outside branches, demand will grow for fraud operations specialists, contact center bankers, financial wellness coaches, and digital banking support staff. These roles use many of the same skills tellers develop — customer service, transaction knowledge, regulatory awareness, judgment — in different settings. The teller career path of 2028 looks less like "branch teller for ten years" and more like "two years at the branch, then either move into universal banking, contact center, or operations, or leave the bank for an adjacent industry."

Career Advice for Bank Tellers

Act now. The transition is not theoretical — it is happening. Develop advisory skills that let you transition to banker, financial advisor, or relationship manager roles. Read your bank's product catalog cover to cover so you can talk credibly about mortgages, credit cards, small business lending, certificates of deposit, and basic investment products. Ask to shadow a personal banker for a week. Volunteer for cross-sell campaigns. The transition from teller to universal banker is the single most achievable career move inside the bank, and it is the path most likely to keep you employed past 2030.

Get licensed for investment products (Series 6 or 7) or insurance. Even a Series 6, which covers mutual funds and variable annuities, opens the door to financial advisor career paths and signals to your employer that you are serious about advancing. Many banks will sponsor study materials and exam fees because they need licensed staff in branches. Insurance licensure broadens the product set further, and it travels well if you ever leave banking for an independent advisory career.

Build your technology skills. Learn the bank's CRM, the digital banking platform from the customer's perspective, and any AI-driven sales recommendation tools your bank has deployed. Tellers who can confidently coach customers through mobile banking setup, troubleshoot biometric login, and demonstrate the bank's investment app become more valuable than tellers who can only run a teller terminal. [Estimate] Banks consistently report that "digital coaching" by branch staff increases mobile banking adoption among older customers by 20-40%, which directly improves the branch's profitability metrics.

Consider adjacent careers in fintech, financial operations, or customer success. Many fintech startups specifically value former bank staff because they understand regulatory realities, customer behavior, and operational risk in ways that pure technologists do not. Compliance, AML investigations, fraud operations, and customer success at fintech firms are all natural landing spots. Outside finance, customer success roles at SaaS companies and credit-related companies (credit unions, buy-now-pay-later providers, payments firms) frequently hire from the teller pool.

Finally, consider continuing education while you still have a stable paycheck. Community college courses in business, accounting, or data analytics; online certifications in customer relationship management; or even an associate's degree in business administration all expand your options. [Claim] The teller who views this moment as an opportunity to grow into a higher-value role will be fine. The one who waits for the branch to close will have fewer options, and far less time to plan.

For detailed automation data, see the Bank Tellers page.

---

_This analysis is AI-assisted, based on data from Anthropic's 2026 labor market report and related research._

Update History

• 2026-03-25: Initial publication with 2025 baseline data.
• 2026-05-13: Expanded with branch traffic data, universal banker model, elder financial abuse considerations, BLS projection context, and detailed transition paths into banker/fintech/operations roles.

Related: What About Other Jobs?

AI is reshaping many professions:

• Will AI Replace Accountants?
• Will AI Replace Financial analysts?
• Will AI Replace Truck Drivers?
• Will AI Replace Graphic Designers?

_Explore all 1,016 occupation analyses on our blog._