The Generative AI Future for Equipment Finance: It’s Time to Think About How to Make it Happen

by Valerie L. Gerard and Denis Stypulkoski Jan/Feb 2024
What are the practical applications of AI in equipment finance? What can it do today and what can we train it to do tomorrow? Valerie Gerard and Denis Stypulkoski explore possibilities for streamlining sales, documentation, underwriting, asset valuation and more.

Valerie L. Gerard,
The Alta Group

Denis Stypulkoski,
Founder and Principal,
Reimagine Advisors

While AI in equipment finance is still in its infancy, the technology has been used for some time in areas such as managed services. As AI matures, its greatest promise lies in generative AI applications where deep-learning models create rich, original content derived from enterprise data.

No concrete models exist for how generative AI will achieve this. Plotting this future will require an ambitious reimagining of how people who perform basic equipment finance tasks can turn AI-based insights drawn from intuitive data analysis into effective, high-level decisions through which they, their employers and customers mutually prosper. It will require developing a radical, disruptive business case in which generative AI moves beyond incremental tinkering with conventional equipment finance business processes and into a world in which our workers can shed non-value-added tasks in favor of creative and relationship-building work that helps the business grow and evolve.

Turning this hard-to-envision goal into hard-and-fast reality demands a digital ecosystem that supports a co-piloting partnership between the generative AI engines that interpret the data and the people who will utilize those enlightened analyses to foster more productive relationships with customers.


First, we must clarify what AI is, and what it can do, by defining what AI’s related components are and how they differ. After that — in discussing a collections use case — we can do higher-level thinking about what a generative AI-based, disruptive business program might look like.

Predictive analytics, machine learning (ML) and predictive AI are interconnected within AI but are nevertheless distinct.

Predictive analytics ties historical patterns, data and statistics into its forecasts of future events, and while it has evolved to sometimes use ML, predictive analytics is an older, more procedural approach that typically is expert-driven on a periodic basis.

ML is just that — machine algorithms learn from data and past events to detect patterns and make predictions.

Things get a little confusing with predictive AI, because some zealous advertising markets predictive analytics as predictive AI. Predictive AI uses ML algorithms derived from historical data to continuously recognize patterns, identify developing trends and make predictions.

Now that we’ve defined terms, it’s important to state the biggest difference between AI and predictive analytics: AI can be autonomous and learn by itself, but predictive analytics needs help from people to examine data, identify trends and investigate assumptions (although it can sometimes use ML).

And, if we’re going to reach the top of the AI pyramid — generative AI — we must know how it differs from predictive AI.

Simply put, predictive AI uses ML and statistical algorithms to evaluate data and predict future developments. Generative AI takes the training data initially provided from advanced algorithms and deep learning methods and creates powerful new content with attributes related to that data, and generative AI learns from its own prior outcomes.

Imagine this future: conceptualizing and making tangible a built-out constellation of generative AI applications for every stage of the equipment finance lifecycle.

The positive exploitation of data within the industry has been happening for some time. ML algorithms are automating core equipment finance functions, such as asset valuation. While this activity has created beneficial effects, it’s conceivable that generative AI can have a much more transformative effect on the entire equipment finance leasing cycle — if the industry’s stakeholders contribute their ingenuity and vision to that metamorphosis.

What AI is doing now only hints at what might be possible later.


A generative AI application can become an extension of the sales staff. It could aggregate sales leads into groups with shared interests who might react favorably to particular product and service marketing promotions and score those leads to boost close rates. This could simplify matching lenders with equipment financing prospects according to specific prospect needs.

Additionally, AI could help equipment leasing and finance firms spot market trends and prospect behavior patterns that could enhance their product marketing. Behavior-oriented sales support could also make it possible to predict when end users may need to renew a lease or upgrade equipment.

The sales angles don’t stop there.

AI could benefit pay-per-use products because of their intrinsic predictive usage. AI predictive analytics could drive new pay-peruse product applications for a broader selection of equipment types and target those opportunities more precisely by industry type and location.


Envision a future in which AI co-pilots sales associates’ discussions with prospects about their business and equipment financing requirements. This sales AI co-pilot, working alongside a risk-assessment AI co-pilot, could create optimized combinations of deal attributes and values to produce multiple proposals for the client, while capturing that data for posting into a digital ledger. The generative-AI documentation co-pilot interprets that data and drafts the lease and loan contracts alongside each proposal, immediately satisfying the needs of the legal, credit and risk requirements without the need for human intervention.

Third-party-originated documentation bought and sold in the capital markets syndication process can be automatically reviewed by AI for comparison to the attributes and data values being presented by the seller to determine any misstatements, inaccuracies or misrepresentations.


AI’s data-analysis proficiency is invaluable to credit underwriting. The enormous amount of data that AI tools can process in real time, such as an applicant’s income, employment and credit history, enables automated retraining of credit scoring models as well as expedited underwriting decisions that are more likely to be correct and avoid defaults. Integrating AI with equipment finance and leasing systems in this way eliminates bias and, consequently, makes these systems more reliable than traditional credit underwriting systems. Utilizing AI to automate routine analysis and provide deeper insights into potential unforeseen risks will enable underwriters to focus on gathering and evaluating additional data from borrowers outside of AI’s reach.

The industry generally benefits from the labor saving, mistake reducing effects of automation on such repetitive, time-consuming functions as risk assessment, credit scoring, document processing and contract management. The resulting efficiencies produce a healthier bottom line and better customer experiences. This offers a glimpse of how equipment finance professionals might operate on an intuitive, instinctive, decision-making level in a generative-AI future.

The credit-related AI picture also concerns anti-money laundering (AML) rules such as know your customer (KYC) verification. While there aren’t any specific regulations that govern AI for KYC, regulators look favorably on firms that take innovative AI steps to strengthen their compliance efforts, such as spending less time processing false positives so they can concentrate on genuine cases.

AI is also pertinent to sanctions compliance measures needed to minimize risks to instant-payment systems involved in cross-border transactions. Depending upon how an institution assesses its level of risk, the Office of Foreign Assets Controls, which has jurisdiction in this area, endorses using AI tools as innovative compliance solutions to mitigate those risks.


AI-enabled billing and invoicing software has helped businesses significantly reduce calculations and manual operations associated with these processes. AI-powered accounting solutions support features such as invoice segregation, data extraction and invoice generation.

Because an AI system contains detailed reports of a company’s total accounting history, it will generate actionable data on how quickly customers pay, their payment methods and if the payor is a person, group or institution, which makes invoicing more effective. Since AI-based billing systems can access historical data, they can tell how payments have been coordinated through a transactions system and, subsequently, predict how to better match future invoices.


AI data mining also facilitates service that is custom-tailored to an individual applicant’s needs and, hence, creates a marketplace advantage. It enables lenders to offer distinctive business solutions that align with customer desires. That service ethic gets additional support from AI-generated chatbots, which can promptly answer frequently asked questions about subjects such as financing options and help resolve lease-related concerns.

A corollary to this involves predictive maintenance, where algorithms can assess sensor-based and other data to determine when an asset will need repairs or other servicing. The upshot is less asset downtime and maintenance expense, higher asset performance, less equipment failure — and more satisfied customers. AI’s ability to integrate with IoT devices to track equipment health can provide data to guide maintenance financing options.


AI predictive analytics can examine a leased asset’s history and anticipate its future residual value and end-of-lease condition. Lessors also could identify factors that dictate a change in scope for scheduled equipment inspections to ensure the equipment’s long-term value. They could get timely alerts about deteriorating asset conditions that might prompt a mid-term buyout to avoid absorbing a forthcoming change in the asset’s value. This intelligence could even guide end-of-lease term negotiations strategies, such as whether to sell or release the equipment to the current end-user/lessee.

This knowledge can inform investment decisions and portfolio risks and enable the pinpoint transaction pricing that can give lessors a big competitive edge.


The pillar-to-post pace and magnitude of the refinements in operational processes pale in comparison to the speed-of-light advancements that await the industry when/if it is transformed by generative AI. Nobody knows what that looks like, nobody has conceived what seems inconceivable: a system where a perfect mesh of people, process and technology create a workforce and business cycle that function at peak levels. Nobody has conceived any generative AI use cases that would manifest that future.

…Well, almost nobody.

We hope we can kick off a serious discussion about this subject by submitting for everyone’s consideration what such a use case could be for small-ticket, late-stage collections. It’s a logical extension of what began decades ago, when small-ticket leaders embraced predictive analytics for credit scoring, then directed a collections experience evolution through machine-learning-driven predictive analytics and predictive AI.

During the predictive analytics phase, models were designed, re-designed and back-tested to identify transaction attributes that helped separate good deals from bad ones. After predictive analytics proved itself on the front end, lenders began considering it for back-end operations through the design and implementation of behavioral models.

As the behavioral models imported updated credit data about portfolio assets and borrowers and combined it with macro-economic data from federal government agencies, ML was applied to collections activity through self-adjusting, self-updating behavioral scoring mechanisms that made sense of the data. This let lenders segregate customers into categories — which ones to call, how often to call them and who is at greatest risk — to achieve the highest possible collection success rates.

From here, leveraging predictive AI was a natural evolution. The difference is the speed at which the ML predictive AI engine was ingesting data and the capability of the algorithms to predict customer delinquency and ultimate loss performance in real time.

From there, how big a stretch is it to contemplate a generative-AI upgrade? For us, it involved drawing upon recent conversations with industry leaders to optimize the knowledge of retiring workers who had proven effective in collections interactions with customers. Our hypothetical use-case retiree is Harry, a hypothetical company’s best post-60 collector.

An AI bot could conceivably listen in on all of Harry’s post-60 calls and emails, with the goal of capturing every attribute of the delinquent customer and the decisions and actions Harry takes to nurture that customer toward payment.

To replicate Harry’s collections success, we must think like Harry thinks. We must capture every element of every piece of data about that customer — industry data, news, credit data and more — that Harry would be harvesting at the 60-day window, to come up with a persuasive game plan to get this customer to pay. It’s nearly impossible for human managers to accomplish this alone, but AI provides a powerful tool to help future employees repeat his success.

Specifically, a generative-AI engine would create the script that teaches Harry’s successor collectors to behave like Harry, because it’s adjusting all the data Harry was seeing and creating the same insights that Harry would be thinking up — not only sharing Harry’s insights with the new collectors but also sharing his thought processes around them, so they’re learning from a generative-AI tool that’s helping them be more successful in late-stage collections.

This ‘auto-Harry’ becomes a real-time co-piloting tool to coach agents in human-to-human engagements, based on the time-tested methods of their predecessors.


How do you think big in a generative AI sense, so that the industry moves well beyond incremental process improvements? One approach, discussed here, is to start in collections, by more fully automating early-stage interactions, where conversational natural language models and bots could free up employees to partner with their collections copilot to handle the most challenging late-stage collections problems. While this is speculative at this point, and might look highly improbable to some, if we don’t start somewhere, how will we ever get to the next level?

There’s the rub. Since our industry tends to adopt change in baby steps, what’s necessary is a paradigmatic shift to bigger-picture strategic thinking. Companies and individuals that attempt to collaboratively brainstorm generative-AI solutions should monitor parallel industries to learn from their major disruptions — so they can predict our industry’s future and meet it with opportunistic business agility.

The biggest imperative is creating and committing to a foundational digital ecosystem containing a single version of the truth of every deal. From there, AI can reliably make informed decisions based on the deal’s true attributes.

The Alta Group and Reimagine Advisors can do this higher-level thinking to help equipment finance lenders, but this assignment belongs to the whole industry, with everybody contributing to an overarching vision of a generative AI future. •

ABOUT THE AUTHORS: Valerie L. Gerard is co-chief executive officer of The Alta Group and leads the Strategy & Competitive Alignment practice. She helps companies design and implement value-creating solutions, and partners with leadership teams on both strategic and tactical issues ranging from growth strategies and business-model optimization to multi-vendor customer financing programs and long-term capitalization. She has been a member of the Board of Trustees for the Equipment Leasing and Finance Foundation since 2018 and is the 2023 recipient of the Foundation’s Steven R. LeBarron Award for Principled Research.

Denis Stypulkoski has 30 years of experience bridging the domains of the equipment finance business and technology. He developed Reimagine Advisors during the COVID-19 pandemic to give back to his colleagues and innovate the industry. His forward-thinking business model coaches clients to break down the hierarchy of traditional departmental “boundaries” and replace it with a sustainable, agile framework focused on the integrated customer and employee experience, empowering employees to continuously learn and contribute new ideas.

Leave a comment

No categories available

No tags available