Square's $7B Lending Surge and the AI Layoff Paradox for MCA Lenders

Square's Numbers Are a Wake-Up Call for Every MCA Lunder

When Square Loans reported $7 billion funded to merchants in 2025, the number alone was impressive. What made it remarkable was the context: parent company Block simultaneously laid off 40% of its entire workforce, explicitly citing AI as the reason those roles were no longer necessary. The implication is hard to miss. AI underwriting for merchant cash advance is not a future trend. It is the operating model behind one of the largest merchant lenders in the world, today.

For independent MCA funders, brokers, and underwriting teams, these numbers demand a serious question. If a platform lender can originate billions with fewer people, smaller margins, and nearly immaterial default rates, what does your competitive moat actually look like? The answer is not to panic. It is to understand exactly where AI creates leverage in MCA underwriting and where human judgment still matters. This article breaks down the strategic lessons from Square's AI-driven growth, examines what Stripe Capital's parallel expansion tells us about the broader market, and identifies the practical steps independent lenders should take in 2026 to stay competitive without betting everything on a platform rebuild.

What Platform Lenders Get Right About AI Underwriting

The Data Advantage That Drives Near-Zero Defaults

Square's secret is not a better algorithm in isolation. It is the feedback loop between transaction processing and lending. Because Square processes every card swipe, every invoice, and every deposit for its merchants, its underwriting model feeds on real-time revenue data rather than static bank statements or self-reported financials. The result: loan performance so strong that nonperforming loans were described as "immaterial" in Block's year-end report.

Stripe Capital followed a parallel trajectory, originating 81,000 merchant cash advances and business loans in 2025 through its payment processing subsidiary. The pattern is identical: own the payment rails, observe merchant cash flow in real time, and use machine learning models trained on millions of transactions to predict repayment probability with startling accuracy.

Independent MCA lenders do not have this embedded data advantage. That is a fact, not a criticism. But recognizing it is essential because it shapes the strategic response. You cannot out-data Square or Stripe. What you can do is layer smarter verification and analysis into the data you do collect.

AI Replaces Process, Not Judgment

Block's 40% workforce reduction is eye-catching, but it is worth examining what those roles actually did. The cuts were concentrated in operational, repetitive, and rules-based functions: processing applications, reviewing documents against checklists, scheduling and coordinating verification steps. These are exactly the tasks where AI excels, pattern matching at speed across structured data.

What AI does not replace, even at Square's scale, is the contextual judgment that experienced underwriters bring to edge cases. A merchant whose revenue dipped 30% last quarter might be a terrible risk or might be a seasonal business entering its peak cycle. An unusual deposit pattern might signal fraud or might reflect a legitimate factoring arrangement. Platform lenders handle this by building increasingly sophisticated models that encode more context. Independent lenders handle it by keeping smart people in the loop.

The winning formula for MCA shops that are not platform giants is not full automation. It is intelligent augmentation: automate the repetitive verification and data collection steps so your underwriters spend their time on decisions, not scheduling calls and chasing documents.

The Embedded Lending Threat Is Accelerating

Square and Stripe are not the only players expanding. As we explored in our analysis of how Lightspeed and Worldline are normalizing embedded MCA, payment processors across the board are building lending products directly into their merchant dashboards. For the merchant, the experience is seamless: a pre-approved offer appears, they click accept, and funds arrive the next day. No application, no bank statements, no verification call.

This convenience sets the expectation bar. When a merchant then encounters an independent MCA funder that requires them to schedule a phone call, walk through their bank portal line by line with a stranger, and wait days for a decision, the friction is obvious. The lenders who close deals in 2026 will be the ones who make verification feel nearly as effortless as the embedded lenders, even without owning the payment rails.

How Independent MCA Lenders Can Close the Gap

Async Verification as a Force Multiplier

The most direct analog to what AI does for platform lenders is removing the scheduling bottleneck from bank verification. When Square processes a loan application, there is no phone call. Their system reads the data automatically. Independent lenders can approximate this speed by shifting to asynchronous verification workflows.

With Exact Balance, the applicant receives a secure link, records their live banking session at their convenience using browser-based screen capture, and the underwriter reviews the recording whenever they are ready. No coordination overhead. No time zone headaches. The AI-guided recording coach walks the applicant through exactly what to show, and activity tracking captures the full audit trail automatically.

This is not a marginal improvement. For a team processing dozens of deals per week, eliminating live verification calls can recover hours of underwriter time daily. That recovered time goes directly into reviewing more deals or doing deeper analysis on borderline applications, exactly the kind of work that AI cannot fully automate.

AI-Assisted Fraud Detection in Bank Verification Recordings

One area where AI adds direct value to independent lenders is in spotting manipulation. As we detailed in our coverage of the FBI's carroting fraud case, sophisticated applicants are fabricating bank statements with increasing skill. Static PDF analysis catches some forgeries, but video-based verification is fundamentally harder to fake because it captures a live browser session with real-time page loads, scrolling behavior, and URL bar verification.

Exact Balance's AI vision layer analyzes recordings for step completion and session authenticity, checking whether the applicant actually navigated to their bank's portal, whether the URL matches a known banking domain, and whether the recording captures the specific data ranges requested. This is a concrete application of machine learning in fraud detection: not replacing the underwriter's judgment but surfacing anomalies that warrant closer scrutiny.

Transaction Monitoring Beyond the Snapshot

One lesson from Square's near-zero default rates is the power of ongoing cash flow observation versus point-in-time snapshots. Traditional MCA underwriting reviews three to six months of bank statements and makes a decision. Platform lenders continuously monitor merchant revenue and adjust exposure accordingly.

Independent lenders can move toward this model incrementally. Using Exact Balance, you can create follow-up verification requests at any point during the advance term, asking the merchant to record an updated view of their account activity. This is not as seamless as real-time payment rail monitoring, but it provides a structured, auditable way to check in on merchant health without the overhead of scheduling another call. Combined with the platform's full audit trail and timestamped recordings, these periodic check-ins create a compliance-friendly record that regulators like FINTRAC increasingly expect from alternative lenders operating in Canada.

Real-World Scenarios: Where This Plays Out

Consider a mid-sized Canadian MCA funder processing 150 deals per month. Under a traditional workflow, each deal requires a 20-to-30-minute live verification call, plus 10-to-15 minutes of scheduling and rescheduling. That is roughly 75 to 110 hours of underwriter time per month consumed by verification logistics alone. At that volume, adding even one more underwriter is a significant cost.

Now layer in the Square effect. Merchants are increasingly seeing instant funding offers from their payment processors. The ones who come to independent lenders tend to be merchants who either do not qualify for embedded products or need more capital than a platform will extend. These are inherently riskier deals requiring more careful verification, not less. The independent lender is caught in a squeeze: deals require more scrutiny, but the market demands faster turnaround.

Async verification breaks this tradeoff. The applicant records their banking session the same evening they apply, often within hours. The underwriter reviews it the next morning alongside five other recordings, batch-processing verification in a fraction of the time. The lender can fund by end of day while still maintaining rigorous documentation. No corners cut, no staff added.

Another scenario worth examining: the multi-position stack. A merchant applying for a second or third MCA position needs even more thorough bank verification because the underwriter must confirm existing repayment obligations. Live calls for these deals are particularly painful because the applicant often needs to navigate to multiple statement periods and show specific recurring debits. With recorded sessions, the underwriter can pause, rewind, and cross-reference at their own pace. The recording becomes a reviewable artifact rather than a one-shot, real-time conversation where critical details can be missed.

Frequently Asked Questions

What is AI underwriting for merchant cash advance?

AI underwriting for merchant cash advance refers to the use of machine learning models, automated document analysis, and algorithmic risk scoring to evaluate MCA applications with minimal manual intervention. Platform lenders like Square and Stripe use AI underwriting to process thousands of applications daily by analyzing real-time payment data. Independent MCA lenders can adopt AI-assisted tools for specific workflow steps, such as automated bank statement analysis, AI-guided verification recordings, and fraud pattern detection, without replacing their underwriting teams entirely.

How do platform lenders like Square achieve such low default rates?

Platform lenders achieve low default rates primarily through data access. Because they process merchant payments, they observe real-time revenue before and during the loan term. Their AI models are trained on millions of merchant transaction histories, allowing them to predict repayment probability with high accuracy. They also control repayment through automatic deductions from daily sales. Independent lenders without payment rail access can improve their own default rates by implementing more rigorous verification processes and ongoing merchant monitoring.

Can independent MCA lenders compete with AI-driven lending platforms?

Yes, but the competitive strategy is different. Independent lenders compete on flexibility, deal size, and willingness to serve merchants that platforms decline. The key is adopting AI-assisted tools that reduce operational overhead without requiring a full technology rebuild. Async bank verification, automated document classification, and AI-powered fraud screening allow smaller teams to process more deals with greater accuracy. The goal is not to match Square's automation level but to close the speed and efficiency gap enough that merchants do not abandon the application process.

Why is async bank verification important for MCA lenders in 2026?

Async bank verification eliminates the scheduling bottleneck that slows deal velocity for MCA lenders. In 2026, merchants expect near-instant funding decisions because embedded lenders have set that standard. By allowing applicants to record their banking sessions at their convenience and letting underwriters review on demand, async verification compresses the timeline from days to hours. Tools like Exact Balance add AI-guided recording and activity tracking to ensure recordings capture exactly what underwriters need, creating a compliant audit trail without a single phone call.

Conclusion

Square's $7 billion lending year and Block's dramatic AI-driven layoffs are not abstract headlines. They are a direct signal about where MCA underwriting is headed. Platform lenders will continue to automate, scale, and compress timelines. Independent MCA lenders do not need to match that scale, but they cannot ignore the efficiency gap.

The practical response is not to rebuild your operation from scratch. It is to identify the highest-friction points in your current workflow and apply targeted AI-assisted tools. Bank verification is the obvious starting point because it is the single most time-consuming manual step in most MCA underwriting processes.

Exact Balance was built for exactly this moment. Async verification, AI-guided recordings, and a complete audit trail let your team process more deals, catch more fraud, and fund faster. Visit exactbalance.ca to see how it fits into your workflow.