A new platform enters the AI underwriting race, but is it the tool your shop actually needs, or just another vendor selling the future?

AdvanceIQ.ai launched the AIQ Platform this month, a unified AI-powered risk and portfolio intelligence system built specifically for MCA and revenue-based financing lenders.^{1 3} The platform combines three modules: PortIQ for portfolio analytics, SMB RiskIQ (SRI) for deal scoring, and ARIA, a natural language AI agent that flags portfolio trends in plain English.³

The timing is deliberate. AI adoption among lenders jumped from 15% in 2023 to 38% in 2024, and Fannie Mae projects 55% will pilot or expand AI tools by end of 2025.⁷ The US MCA market hit $19.65 billion in 2024 and is on track for $32.7 billion by 2032.⁸ AdvanceIQ is betting that MCA shops are ready to stop cobbling together spreadsheets and start paying for integrated intelligence. The question is whether they are right, and whether their product actually delivers what a $50M-$200M originator needs.

Key Developments:

1 AdvanceIQ.ai | AIQ Platform Overview
2 Funder Intel | AdvanceIQ.ai Blog Coverage
3 Funder Intel | AdvanceIQ.ai Launches AIQ Platform for SMB Lending Analysis
4 Funder Intel | AdvanceIQ.ai Tag
5 deBanked | LendSaaS Embeds AdvanceIQ.ai's SRI
6 LendSaaS | The Leaders in MCA Servicing
7 Certified Credit / Fannie Mae | AI Adoption in Lending 2025
8 Verified Market Research | US Merchant Cash Advance Market
9 Market.us | AI in Lending Market Report
10 Heron Data / Insight Partners | $16M Series A Raise
11 Underwrite.ai | AI Underwriting Case Studies
12 Mass.gov | $2.5M Settlement Over AI Lending Bias
13 CFPB | Guidance on AI Credit Denials
14 ByzFunder | 40% Growth in 2025 Driven by AI Underwriting
15 Zest AI | AI-Powered Credit Underwriting

Because the numbers justify the hype, at least in theory. The AI in lending market was valued at $7 billion in 2023 and is projected to hit $58.1 billion by 2033, a 23.5% compound annual growth rate.⁹ Lenders implementing AI tools report 30-50% reductions in operational expenses and 2.5x faster loan closures.⁷ ByzFunder, an MCA-specific shop, reported 40% year-over-year growth in 2025, crediting AI underwriting as a primary driver.¹⁴

The pitch from every vendor sounds the same: faster decisions, better pricing, fewer defaults. But the reality varies wildly depending on what "AI" actually means under the hood. Some platforms use genuine machine learning models trained on MCA-specific data. Others slap a ChatGPT wrapper on a basic rules engine and call it intelligence. The difference matters when you are making $50K-$500K funding decisions based on the output.

AdvanceIQ.ai's entry is notable because it claims to be purpose-built for the MCA and RBF ecosystem, not a consumer lending model repurposed for commercial use. That distinction is worth examining.

Three things, each targeting a different pain point.

PortIQ is the portfolio analytics layer. It segments performance data by originator, ISO, account executive, underwriter, and industry vertical. If you are running a $100M book and want to know which ISOs are sending you deals that default at 2x your portfolio average, this is supposed to surface that in real time instead of waiting for your monthly Excel reconciliation.^{1 3} It also generates investor-ready reports, including what they call "Syndicator Portfolio Pulse Reports," which matters if your capital stack depends on keeping syndicators happy with transparent performance data.

SMB RiskIQ (SRI) is a proprietary 10-point scoring system with A-F letter grades, designed specifically for MCA and RBF deal flow. The idea is to filter low-potential deals before your underwriters spend 20 minutes reviewing bank statements that were never going to fund.^{1 3} It already integrates with LendSaaS, a platform that has facilitated $6 billion in funded volume and processes $16 million in daily ACH volume.^{5 6}

ARIA is the AI agent layer. It uses natural language processing to proactively flag trends: a regional revenue drop, a vertical showing unusual delinquency patterns, a segment where your pricing is leaving money on the table.³ Think of it as an analyst who reads your entire portfolio every morning and sends you a briefing before your first coffee.

Most MCA shops above $25M in annual originations have built internal scorecards. They know their data. They have years of performance history tuned to their specific risk appetite. A new vendor saying "our 10-point score is better" needs to prove it against your existing model, not against a generic benchmark.

For context on what AI scoring can deliver when it works: one online lender using Underwrite.ai saw first-payment default rates drop from 32.8% to 8.5%.¹¹ Zest AI reports an average 25% increase in approvals with no additional risk across nearly 300 lender clients.¹⁵ LendingClub's AI model produced 40-50% fewer 30-day delinquencies compared to peers.¹¹

Those are impressive numbers. But they come from platforms with years of training data and validation. AdvanceIQ.ai is new. The SRI score may be excellent, but we have not seen independent performance data, backtesting results, or head-to-head comparisons against established internal models. Lendr's CEO said it provides "clearer portfolio views and faster decisions."³ That is encouraging but vague. Clearer than what? Faster by how much?

If you are evaluating SRI, the right question is not "is AI better than manual?" The answer to that is obviously yes. The right question is "is this particular AI model better than the scorecard my team already built, and by enough margin to justify the integration cost and vendor dependency?"

AdvanceIQ.ai is not entering an empty market. The competitive landscape is crowded and well-funded.

Heron Data raised $16.5 million in Series A funding in 2025 from Insight Partners, processes over 500,000 files per week, and serves 150+ customers. Their focus is bank statement analysis and automated underwriting intake, and they recently launched a Broker Suite for deal flow automation. Customers report processing cost reductions of up to 80%.¹⁰

Ocrolus raised $80 million in Series C and also integrates with LendSaaS, meaning it competes directly with AdvanceIQ.ai on the same platform. Their focus is document automation: extracting and analyzing bank statements, tax documents, and pay stubs using AI.

Zest AI serves nearly 300 lenders (primarily credit unions and community banks) with ML-based credit models. They are more focused on traditional lending than MCA, but their technology sets the benchmark for what AI underwriting should deliver: explainable models, measurable lift, and regulatory defensibility.¹⁵

Bectran handles B2B trade credit automation with implementation timelines under four weeks and claims 90% reduction in credit application processing time.

The differentiation question for AdvanceIQ.ai is whether "purpose-built for MCA" is a real technical advantage or a marketing claim. If SRI was trained on MCA-specific performance data (factor rates, split percentages, ACH payment patterns, ISO quality signals), that is genuinely different from a consumer credit model adapted for commercial use. If it is using the same underlying approaches with MCA-flavored labels, the distinction is cosmetic.

Industry data suggests focused AI implementations (one loan product, pre-built components) take 60-90 days. Comprehensive end-to-end deployments run 6-9 months, and legacy system integrations can stretch to 12 months.⁹

For AdvanceIQ.ai specifically, the LendSaaS integration is already live, which removes the biggest friction point for shops already on that platform.⁵ If you are not on LendSaaS, the implementation timeline is unclear. The company's website does not publish pricing, which in this industry usually means custom enterprise contracts, likely six-figure annual commitments based on comparable platforms like Zest AI.

For a mid-size MCA shop doing $50M-$100M annually, the math needs to work. If a platform costs $100K-$200K per year, it needs to either reduce defaults by enough basis points to cover that cost, or increase throughput enough to fund more deals with the same team. Back-of-envelope: if your portfolio is $75M at a 1.35 average factor rate, a 2% reduction in defaults saves roughly $1.5M annually. A $150K platform fee is cheap insurance at that scale. But if you are a smaller operator doing $10M-$20M, the same math does not work unless the pricing scales down proportionally.

This is the part of the conversation that vendors consistently underplay.

In July 2025, the Massachusetts Attorney General secured a $2.5 million settlement against Earnest Operations for AI underwriting models that produced disparate impact against Black, Hispanic, and non-citizen borrowers. This was the first state AG enforcement action specifically targeting AI underwriting bias.¹²

The CFPB has been equally direct. Their guidance states that lenders cannot use boilerplate adverse action notices when AI denies credit. You must provide specific, accurate reasons for the denial, which means you need to understand what your AI model is actually doing.¹³ Courts have held that choosing to use an AI tool can itself be a "policy" that produces bias under disparate impact theory.

For MCA lenders, the regulatory exposure is different than for consumer lenders (MCA is technically a purchase of future receivables, not a loan in most states). But the trend line is clear: regulators are coming for AI-driven credit decisions regardless of the product structure. Any AI scoring tool you adopt needs explainability features, not just a letter grade and a number. Ask AdvanceIQ.ai (or any vendor) how their model handles adverse action explanations and fair lending testing before you sign.

FICO's own researchers have warned that "unleashing pure machine learning models into the lending market would likely usher in systemic risk and lack of transparency."⁹ That is coming from the company that built the most widely used credit score in history. Take the warning seriously.

More than usual, actually. Tomo Matsuo's career arc tracks the evolution of the MCA industry itself. He started as Director of Underwriting at Bizfi (originally AmeriMerchant), one of the earliest MCA platforms, then moved to VP of Business Intelligence and eventually COO.^{4 5} From there he went to Paysafe Group as SVP of Lending Solutions, sitting at the intersection of payments infrastructure and lending at a Blackstone-backed company. He then co-founded Yardline Capital, an e-commerce SMB financing platform that was acquired by Thrasio.

The takeaway: Matsuo has built underwriting systems, managed MCA portfolios, and operated at institutional scale. He is not a tech founder who discovered fintech last year. That does not guarantee the product works, but it means the problem definition is grounded in operational experience, not theoretical assumptions about what lenders need.

Get instant access to comprehensive filing details in 11 major states.

Identify existing liens, verify available collateral, and lend with confidence while others lose millions.

Seize the competitive advantage of superior risk assessment.

Schedule a FREE demo call

Washington State's Department of Commerce awarded $5.6 million to 11 CDFIs through its Equitable Access to Credit program, up from $1 million to 8 CDFIs in 2024. The funds work as a revolving mechanism: loan repayments cycle back into the fund, creating what one CDFI director called "a self-replenishing source of financing." At least 65% of grants support rural counties and Native CDFIs. For alternative lenders, the signal is directional: state governments are building parallel lending infrastructure specifically for the communities traditional underwriting models screen out.

UK-based FundingSearch.com launched a matching platform that connects lenders with pre-qualified SME loan applications backed by live financial data pulled directly from Xero, Sage, and Companies House. Over half of SME loan applications currently fail, consuming broker and underwriting resources. Founder Phillip Evans says placing complex deals "can take weeks of manual research." The platform covers asset finance, invoice finance, working capital, trade finance, MCA, and property finance. For US lenders watching the open-finance trend: this is what happens when verified accounting data replaces self-reported financials in the origination pipeline.

New Orleans-based Bonita Payments launched QuarterMaster, a SaaS platform built to manage merchant onboarding, agent and channel performance, and lifecycle execution from a single operating framework. The tool targets the execution gap that causes preventable churn: inconsistent follow-up, low visibility, and fragmented tools across partner channels. CIO Ramon Maldonado led the build over the past year. Bonita is positioning as a platform company through its broader QuarterSuite product line, connecting processing to capital pathways and merchant enablement.