How Upstart's AI Lending Trilemma Exposes the Bank Verification Gap for MCA Funders

Upstart's "Oldest Truism" and Why MCA Funders Should Pay Attention

During Upstart's Q2 2026 earnings call, CEO Paul Gu made a bold declaration. Lending's oldest truism, the idea that you cannot simultaneously achieve growth, strong credit performance, and profitability, is "not true for us." The claim rests on a specific foundation: AI models that improve continuously because every loan originated feeds data back into the system. For a platform lender processing billions in consumer credit, that feedback loop is powerful. For the average MCA funder reviewing deals from a network of brokers, it is a luxury that simply does not exist.

This gap matters more than most funders realize. The MCA industry is scaling rapidly. QuickBooks Capital alone originated $1.9 billion in business loans last quarter, and brokerages like Fidelity Funding Group are posting $24 million months. Volume is not the problem. The problem is that independent funders lack the closed-loop infrastructure to verify merchant cash flow with the same confidence that platform lenders enjoy. Bank verification software for funders exists to close exactly this gap, giving underwriters visual, timestamped evidence of live banking sessions without requiring a proprietary data ecosystem.

This article breaks down what Upstart's trilemma claim actually means for MCA operations, where the data advantage breaks down for independent funders, and how asynchronous bank verification technology fills the void.

The AI Trilemma and the Platform Lender's Structural Advantage

What Upstart Actually Means

The lending trilemma is not new. For decades, lenders accepted that pushing for growth meant loosening credit standards, which eroded profitability. Tightening credit improved loss rates but slowed origination. Optimizing for profit meant sacrificing one of the other two. Upstart's argument is that AI models trained on sufficient data can price risk precisely enough to approve more borrowers without increasing defaults, and do so profitably.

The key phrase is "sufficient data." Upstart controls the entire origination pipeline. Every application, every approval, every repayment outcome feeds back into model training. The system learns which borrowers perform and adjusts in near real-time. This is the closed-loop advantage: the lender sees the full lifecycle of every dollar it deploys.

Why MCA Funders Cannot Replicate This Loop

Independent MCA funders operate in a fundamentally different environment. Deals arrive through brokers who may submit the same merchant to multiple funders simultaneously. The funder sees a bank statement, maybe a voided check, and a signed application. They do not see the merchant's full financial history, competitive funding positions, or real-time cash flow trajectory.

This is the structural disadvantage. Without a closed data loop, even the most sophisticated AI model starves for reliable inputs. A machine learning credit risk model is only as good as the data it trains on, and when that data comes from static PDFs or broker-submitted documents, the risk of manipulation is significant. As we explored in our analysis of how SMB lending fraud is concentrating in the MCA sector, document-based verification alone leaves funders exposed to edited statements, synthetic portals, and fabricated transaction histories.

Bank Verification as the Data Quality Layer

If platform lenders solve the trilemma through data volume and feedback loops, independent funders need a different approach to data quality. The answer is not more data points. It is higher-confidence data points. A single verified screen recording of a merchant navigating their live banking portal, captured in real time with timestamps and activity logging, provides more underwriting confidence than a stack of static bank statements.

This is where bank verification software for funders becomes a strategic asset rather than a compliance checkbox. Exact Balance's asynchronous recording workflow lets merchants capture their banking session at their convenience, with an AI-guided coach walking them through each required screen. The underwriter reviews the recording on demand, checking for visual consistency, URL authenticity, and transaction detail. No scheduling calls. No timezone coordination. Just verified evidence that the numbers are real.

How Async Bank Verification Closes the Trilemma Gap for MCA Funders

Growth Without Loosening Standards

The growth side of the trilemma is straightforward for MCA. Brokers are generating more deal flow than ever. The bottleneck is not lead volume; it is verification throughput. Every deal that sits in a queue waiting for a live verification call is a deal that might close with a competitor first. Asynchronous verification removes the scheduling friction entirely. Merchants record when they are available, and underwriters review when they are ready. This alone can compress funding timelines from days to hours.

Fidelity Funding Group's $24 million month illustrates the scale of opportunity. When a brokerage moves that kind of volume, the funders on the other side need infrastructure that keeps pace. Manual call-based verification cannot scale to match. Async recording can.

Stronger Credit Performance Through Visual Evidence

The credit performance side of the trilemma depends on making better approval decisions. Better decisions require better information. A screen recording of a live banking session is nearly impossible to fake convincingly. The underwriter can see the bank's URL in the browser bar, watch the merchant scroll through transaction history, and verify that balances and dates are internally consistent. This is a fundamentally different quality of evidence than a PDF that could have been edited in minutes.

AI-powered analysis adds another layer. Exact Balance's platform uses AI vision to validate that recordings show authentic banking portals, detect anomalies in page rendering, and verify that the guided steps were completed in the correct sequence. These are not hypothetical capabilities. They are the same class of techniques that the Federal Reserve's own research on financial data integrity identifies as critical for responsible lending decisions.

Profitability Through Operational Efficiency

The profitability angle is often the most overlooked. Every hour an underwriter spends scheduling and conducting live verification calls is an hour not spent reviewing deals or making funding decisions. The operational cost of synchronous verification is hidden but substantial: phone tag with merchants across time zones, repeated calls when someone misses an appointment, and the cognitive load of managing a calendar alongside a deal pipeline.

Async verification collapses this cost structure. One underwriter on Exact Balance's dashboard can manage dozens of verification requests simultaneously, reviewing recordings in sequence and marking each as verified with a single click. The full audit trail, including when links were opened, recordings started, and submissions completed, provides compliance documentation without any additional administrative effort. As we discussed in our piece on how MCA audit season exposes bank verification documentation gaps, this kind of automated record-keeping is becoming essential as regulatory scrutiny increases.

Real-World Scenarios: What This Looks Like in Practice

Consider a mid-size Canadian MCA funder processing 150 deals per month through a mix of direct applications and broker submissions. Under a traditional workflow, their two-person underwriting team spends roughly 40% of their time coordinating live verification calls. That is the equivalent of losing one full-time underwriter to scheduling logistics.

Switching to async verification reclaims that capacity immediately. The funder sends a verification request through Exact Balance, the merchant receives a secure email link with custom instructions specifying which accounts, date ranges, and transaction details to show, and the merchant records their session in their browser without installing any software. The recording arrives in the funder's dashboard alongside an activity log showing every step the merchant completed.

Now multiply this across the industry. Brokerages like Fidelity Funding Group are pushing record monthly volumes. Platform lenders like QuickBooks Capital are originating at a $7.6 billion annual run rate. The funders who cannot verify at this pace will lose deals to those who can. The trilemma Upstart describes is real, but the solution for independent funders is not building a billion-dollar AI platform. It is deploying verification infrastructure that produces high-confidence data at scale.

The parallel to collections is instructive as well. Erica Gilerman's recent profile in deBanked highlights how collections professionals in MCA need airtight documentation: signed stipulations, verified account details, and clear audit trails. The verification standards that matter at the collections stage should be established at origination. Funders who capture visual proof of banking activity at the point of underwriting build a stronger position if a deal ever reaches dispute or default.

Frequently Asked Questions

What is bank verification software for funders?

Bank verification software for funders is a category of tools that allow MCA lenders and alternative finance companies to confirm the authenticity of a merchant's banking activity before funding. Unlike open banking APIs that pull transaction data programmatically, verification software like Exact Balance captures visual proof through browser-based screen recordings of live banking sessions. This produces timestamped, reviewable evidence that transactions and balances are genuine, not screenshots or static PDFs that can be easily manipulated.

How does async bank verification prevent MCA fraud?

Asynchronous bank verification prevents fraud by requiring merchants to record their live banking portal in real time, guided by AI-powered step detection that ensures they show the correct screens. Because the recording captures the actual browser session, including URLs, page transitions, and scrolling behavior, it is far more difficult to fake than a submitted bank statement. AI analysis can flag recordings that show inconsistent page rendering, suspicious URL patterns, or skipped verification steps, alerting underwriters before a funding decision is made.

Can AI fully replace manual bank verification for MCA lenders?

AI can automate significant portions of bank verification, including document classification, transaction categorization, and anomaly detection. However, the MCA industry's broker-driven deal flow and the prevalence of document manipulation mean that fully automated verification without human review still carries unacceptable risk in 2026. The most effective approach combines AI-guided recording and analysis with human underwriter review. Exact Balance uses this hybrid model: AI coaches the merchant through the recording process and flags potential issues, while the underwriter makes the final verification decision.

Why do MCA funders need verification beyond open banking APIs?

Open banking APIs provide structured transaction data, but they do not prove that a merchant actually owns the account or that the data has not been intercepted and modified before reaching the funder. API-based connections can also be spoofed using synthetic bank portals or compromised credentials. Screen-recorded verification adds a visual evidence layer that APIs cannot replicate. The underwriter watches the merchant navigate their own banking portal, confirming identity, account ownership, and transaction authenticity in a single workflow.

Conclusion

Upstart's claim that AI breaks lending's oldest trilemma is compelling, but it depends on a closed-loop data infrastructure that most MCA funders do not have. Independent funders can close the gap not by building their own AI platforms, but by deploying bank verification software that produces high-confidence, tamper-resistant evidence at the speed their deal flow demands. Asynchronous screen-recorded verification eliminates the scheduling bottleneck, AI-guided recording ensures data completeness, and full audit trails satisfy growing compliance requirements.

Exact Balance was built for this exact workflow. Visit exactbalance.ca to see how async bank verification fits into your underwriting process and start verifying at the pace your pipeline requires.