Innovation in the equipment finance industry is poised for a quantum leap as artificial intelligence advances, but risk management will be crucial as lenders implement rapidly evolving technologies.
Roughly 72% of lending institutions implement AI in 2024, up from 55% in 2023, according to software provider JumpGrowth. About 65% of companies had started implementing generative AI, which uses machine-learning algorithms to create new content, up from 33% in 2023.
Of the lenders that have adopted AI, 57% have allotted 5% or less of their digital budgets for AI purposes, according to JumpGrowth. Meanwhile, 13% of lenders have devoted 11% or more of their budgets toward AI, while the remaining 30% fall in the 6% to 10% range.
While still in the early stages, AI is becoming integrated into an industry that’s seen very little innovation during the past half-century, Bill Verhelle, chief executive at Fairport, N.Y.-based equipment finance service provider QuickFi, told Equipment Finance News.
“The business model in the equipment finance industry has generally been the same for the last 50 years,” Verhelle said. “And what I see coming is a new business model, where the whole model will be replaced with something entirely new. That new business model is what we are developing.”
Raj Tulshan, founder of Charleston, S.C.-based financial advisory firm Loan Mantra, shared a similar sentiment during an Aug. 21 webinar on AI-driven finance.
“We are at the 25-year anniversary of the dot-com boom and 20-year anniversary of Google IPO, and we haven’t seen any major innovation in the financial sector,” Tulshan said. “It’s all been derived.
“It’s all about efficiency, cost reduction, etc. And I think that’s where the direct benefit of AI in the financial sector will be.”
— Raj Tushan, Founder, Loan Mantra
Managing risk for lenders
Equipment financiers are increasing their use of AI for:
- Fraud detection;
- Data processing;
- Credit scoring;
- Loan approvals; and
- Developing new financing models.
But as they do so, lenders must manage associated risks, such as algorithmic discrimination, which could exacerbate or perpetuate existing biases in the data that AI systems are trained on, Patrick Hoiby, chief executive at Fort Worth, Texas-based Equify Financial, told EFN.
“This can lead to unfair lending practices if not properly managed. To mitigate this, it is crucial to be diligent during the implementation phase and ensure that governance frameworks are put in place that include audits of AI models to ensure they are transparent, fair and are adhering to ethical lending practices.”
— Patrick Hoiby, CEO, Equify Financial
The number of financial services firms that adopted AI for risk-management purposes increased 11% in 2023, according to Stanford University’s 2024 AI Index Report. About 35% of surveyed financial services firms have implemented at least half of five possible AI-related fairness measures, and 93% have implemented at least one measure, according to the report.
These measures included the involvement of diverse stakeholders in model development and use of technical bias mitigation techniques during model development.
Human oversight
Human oversight will also play a key role in risk management as AI advances, Chris Grivas, president of Chadds Ford, Pa.-based CAG Truck Capital, told EFN.
“I don’t think I could sleep at night if every credit decision at the company was without some kind of oversight,” Grivas said. “I think a story credit is where it will really require the human component; it’s a riskier deal where AI is going to be able to get all the facts, but can it really make the right decision? It’ll be a great tool, but keep your hands on the wheel.”
A story credit is a term used for borrowers who struggle to obtain traditional financing due to challenges such as a subpar credit score.
QuickFi’s Verhelle said non-adoption or slow adoption poses the greatest risk equipment lenders face amid the AI revolution.
“I think the next two to three years, we’re going to see dramatic changes and capabilities of AI. I actually think the risk might be more in not doing enough,” he said. “Today, there are hallucinations, and there are cases where the calculation may not be correct, but that’s pretty minor … I think as the models develop, that’s going to quickly be squeezed out.”
The lending industry is known for being hierarchical and departmentalized, Verhelle said. Those factors, plus regulation, contribute to the difficulty many lenders face in adopting any new technology, including AI.
Fraud prevention, smarter loans
Nearly 75% of financial institutions have begun using AI for fraud detection, according to a 2024 survey conducted by BioCatch. The identity verification provider uses machine learning and biometrics to fend off cybersecurity threats. Of the company’s 600 survey respondents, 87% said AI has increased the speed with which they can identify potential threats.
AI solutions may also help combat fraud and identity theft in the agriculture equipment finance sector, Jay Darden, regional sales manager at Mount Joy, Pa.-based Farm Credit Express, told EFN.
“We are seeing an uptick in fraudulent [loan] applications,” Darden said. “So, we are looking at technology like Truepic, which is an app that geo-stamps a person’s location and is a potential means for verifying identity, because we are definitely seeing more sophisticated identity theft.”
Truepic uses an AI model backed by the Coalition for Content Provenance and Authenticity, a joint development foundation project founded by Adobe and Microsoft, to capture images and trace the origins of the photo’s source, according to its website.
Lenders can lean on tech not just for fraud detection but to avoid other types of risk. For instance, in the trucking industry, AI could provide a comprehensive driving history report to avoid risk, CAG’s Grivas said.
“Imagine signing a credit application and, all of a sudden, the AI is searching for that [driving history] data as well,” Grivas said. “And then you say, ‘Do I really want to lend money to an owner-operator that has six violations?’ You might not, because that guy’s highly likely to lose his license.”
Mitigating borrower risk
AI could also help equipment financiers identify borrowers in different credit tiers. Equify’s Hoiby said AI can “provide a more nuanced understanding of a borrower’s creditworthiness” due to the vast amount of financial and operational data now easily accessible.
“An example in our industry is that an AI tool could assess the borrower’s financial history, industry trends and even include macroeconomic factors to predict cash flow stability over the term of the loan,” he said. “For lenders like Equify Financial, this would enable us to tailor financing packages that align more closely with the borrower’s actual needs and repayment capabilities.”
The technology also has the potential to manage risks associated with fluctuating asset values and other unpredictable factors by providing “real-time insights into market trends, asset performance and thousands of other data points,” he said.
AI’s ability to analyze broader industry and economic trends bodes well for heavy-equipment borrowers as they look to dodge risk, Hoiby said.
Loan Mantra’s Tulshan similarly said geographical and macroeconomic factors will help machines and thus help borrowers learn and predict outcomes better than they can now.
Widespread regulation inevitable?
U.S. lawmakers enacted 25 AI-related regulations in 2023, up 56% from 2022, according to the Stanford AI report. The number of U.S. agencies issuing AI regulations rose to 21 in 2023 from 17 the previous year.
Regulating AI might be a necessity given how fast it’s evolving and being implemented in equipment finance, Farm Credit Express’ Darden said.
“As a general rule, I’m a limited regulation kind of guy,” he said. “However, I think AI is probably off the charts as far as the good and bad it could do. So I would certainly think there’d need to be some parameters to make sure it doesn’t get out of hand.”
