The old way of getting a business loan is dead. In the past, a business owner could walk into a community bank. You could hand over a business plan and talk to a manager who knew your town. I served as a county commission president and a regional council president. I know firsthand that local economic health relies on these real human relationships. When local leaders try to attract new companies or help family businesses grow, they need regional banks to look at the big picture. Today, those human choices are gone. Automated computer programs make the final lending decisions instead.
Under this new automated system, getting a loan comes down to a rigid math formula. These computer programs judge small businesses by tracking digital metrics. They scan items such as daily cash-flow spikes and website activity. Banks claim this makes lending much faster. In reality, it creates a hidden trap for local business owners. The formulas heavily favor giant corporate chains with massive cash reserves. A new medical clinic or a neighborhood retail shop might have great local demand and huge long-term potential. But if the owners lack the deep cash cushions the software looks for, the system instantly rejects them. This creates a severe asset bias that locks out independent newcomers.
The biggest flaw in these programs is that they cannot measure human capability. They completely ignore local community context. An independent operator might focus heavily on face-to-face service instead of an optimized website. These formulas automatically penalize them. The software mistakes a quiet online presence for a dying business model. Younger entrepreneurs and regional businesses face the hardest path. They often have excellent credentials but shorter financial histories. Today, human judgment is gone. It is replaced by a digital screen that you cannot argue with or question.
State governments are starting to see the economic damage caused by these invisible screeners. Eighteen states have now passed automated decision-making laws. These states include California, Maryland, and Virginia. The new rules regulate how personal and commercial data are processed for financial services. Furthermore, frameworks such as the updated Colorado AI Act explicitly draw banks into scope. The law requires strict impact audits and disclosures for high-risk automated tools. These states realize that asset bias hits household economics directly. It forces turned-down business owners to turn to high-interest online cash advances and predatory alternative lenders.
Beyond basic access barriers, these automated underwriting systems create severe cybersecurity dangers. Regional banks routinely connect their legacy portals to private tech networks. This expands their digital attack surface significantly. A major ransomware attack or data breach at a single dominant tech vendor can freeze entire local systems. It locks business owners out of their accounts, halts cash flow, and leaves highly sensitive financial histories exposed to international bad actors.
The integration of artificial intelligence tools introduces an even quieter threat. Banks use automated software to scrape web content, read business plans, and summarize ledgers. The machine handles the entire economic narrative. These models are prone to regular errors. They frequently misinterpret complex tax returns or miss simple seasonal cash-flow context. Because these software programs operate as proprietary corporate secrets, local business owners cannot audit the data that rejected them. This allows systematic underwriting errors to persist without any form of appeal.
EUROPE WROTE THE AI RULEBOOK WASHINGTON IS ARGUING ABOUT. THEN IT HIT PAUSE
When these automated filters deny credit to viable local operators, the harm ripples through the entire community. Small firms operate on tight weekly margins and cannot wait months for an algorithm to fix itself. Driven away from traditional regional banks, entrepreneurs find themselves pushed toward predatory lenders. This rapid inflation of borrowing costs directly drains family finances. It stalls local job creation and strips neighborhoods of vital retail stability. By allowing software to dictate credit terms, the financial sector is quietly choking out local commerce.
Federal rules must catch up to these state-level movements to save independent commerce. Lawmakers in Congress have pushed for federal accountability metrics through proposals such as the Algorithmic Accountability Act. National financial regulators need to mandate complete transparency in commercial lending algorithms now. Under existing federal mandates, the Consumer Financial Protection Bureau enforces strict guidelines regarding complex black-box underwriting algorithms. Banks must be required by law to give applicants the exact metrics used to score them. Most importantly, federal law must ensure that every independent business owner has a clear right to appeal a computer’s rejection to a living, breathing human underwriter.
Eric Wargotz is a political and policy commentator, a former U.S. Senate nominee, former elected president of the Queen Anne’s County Board of Commissioners in Maryland, clinical professor emeritus at George Washington University, and has held numerous leadership roles across government and medicine. The views expressed here are entirely his own.
Continue reading...
[ H/T Washington Examiner ]
