Fighting the Future of Small Business Fraud: How Lenders Can Balance Speed, Risk, and Trust

Highlights: 

• Fraud is evolving with AI, making synthetic identity fraud and fake businesses harder to detect, which requires lenders to move beyond traditional, checklist-based verification methods.
• Effective fraud prevention requires a multi-layered strategy that integrates AI-driven pattern recognition with critical human judgment to better assess risk and context.

Small business lending continues to evolve as financial institutions look for ways to make more efficient and more seamless decisions. The same technology helping lenders approve small business loans faster is also giving fraudsters new tools to create more convincing schemes. Today’s challenge isn’t choosing between speed and security; it’s finding new ways to deliver both.

In a recent episode of the Market Pulse podcast, I spoke with Jill Molitor, Director of Fraud and Credit Administration at Stearns Bank, about how small business fraud is changing and what lenders can do to adapt.

One of the biggest challenges facing lenders today is finding the right balance between speed and security. Just a few years ago, many conversations in lending centered around automation, efficiency and faster decisioning. While those priorities remain important, fraud prevention has become a much larger part of the discussion as bad actors become increasingly sophisticated.

During the conversation, Molitor emphasized that fraud prevention can no longer rely on a single verification step because traditional methods may no longer be enough. Fraudsters are finding ways to make fraudulent businesses and identities appear legitimate, creating new challenges for financial institutions.

One of the biggest challenges lenders face today is that fraud does not always look like fraud. Synthetic identities can be built over time using a combination of real and fabricated information, creating applicants that appear legitimate through traditional verification methods.

These identities may include names, Social Security numbers, addresses, phone numbers or other details that allow them to pass traditional verification checks. Fraudsters may then establish a credit footprint over time — applying for credit, building history and eventually gaining access to larger amounts of financing before disappearing.

The same concept can apply in the business world. Molitor explained that fraudsters may use business filings, create convincing websites or alter information from legitimate companies to make a fake business appear real.

For lenders, this means a basic checklist approach is no longer enough. A business may appear to have the right documentation, but lenders need tools and processes that help verify whether all pieces of the story make sense together.

Artificial intelligence is also changing the fraud landscape.

Many financial institutions are exploring how AI can help identify risk and improve fraud detection. However, there is also another side of the equation: fraudsters are using AI as well.

Molitor noted that while financial institutions are implementing AI carefully and strategically, they also need to prepare for bad actors who are leveraging emerging technology.

Machine learning and data-driven tools can help lenders identify suspicious patterns, such as unusual account activity, rapid changes, repeated failed attempts or activity that falls outside expected behaviors. However, those signals still require context.

According to Molitor, technology can help identify potential warning signs, but human expertise remains essential in determining whether something represents real risk. A situation that appears unusual may have a reasonable explanation, and experienced teams play an important role in evaluating those details.

One recurring theme throughout the conversation was the importance of combining technology with human insight.

AI and automation can provide powerful capabilities, but they cannot replace the experience and judgment of professionals who understand lending, businesses and customer behavior.

Lenders need to avoid looking at individual indicators in isolation, emphasized Molitor. A single data point may not tell the full story. Instead, financial institutions should consider multiple signals together and determine whether the overall picture makes sense.

This approach can help lenders reduce fraud risk while still providing the fast decisions customers expect.

Fraud prevention is not only about identifying bad actors. It is also about understanding a borrower’s complete financial picture.

It is important to evaluate both business and owner information, especially for small businesses where personal and business finances can often be closely connected.

Molitor explained that lenders may benefit from looking at a broader view that includes business credit, personal credit information when applicable, affiliated businesses and overall capacity.

For example, a business may appear to have limited history on its own, but additional context could reveal stronger financial support from related companies or experienced owners. On the other hand, reviewing connected information may help identify potential concerns such as debt stacking or financial stress.

The goal is not simply to find risk — it is to gain a more complete understanding.

Fraud will continue to evolve as technology changes, which means prevention strategies must evolve as well.

Financial institutions need to continue investing in tools, training and collaboration. Employees across organizations need to understand emerging fraud trends, recognize potential warning signs and know when to elevate concerns.

For lenders, the future of fraud prevention will likely require a combination of advanced technology, strong data insights and experienced professionals who can apply critical thinking.

As small business lending becomes faster and more digital, staying ahead of fraud will require more than automation alone. The lenders that succeed will be those that find the right balance — using technology to identify risk while relying on human expertise to make smarter, more informed decisions.

More Market Pulse Resources

To learn more, listen to other Market Pulse podcast episodes or register for the next Market Pulse webinar.