The global used car financing sector was worth around US $460 billion in 2024. According to recent projections, it could be worth nearly US $590 billion by 2030, a CAGR of 2.45%. Finance will be the lifeblood of this growth. But lenders are navigating turbulent times. Among the headwinds buffeting the industry, three major ones stand out: macroeconomic volatility, opaque vehicle valuations and one-size-fits-all credit risk assessments.
Predicting the cost of funds is a familiar headache for any lender, but, for those specialising in used vehicles, recent swings in interest rates and global trade tensions have made the task particularly fraught. Lenders also wrestle with the perennial problem of valuing ageing vehicles. Predicting the value of a five-year-old hatchback with 70,000 kilometres on the clock is far harder than forecasting the fate of a gleaming new model. Lastly, customer risk profiles remain blunt instruments in Europe, where pricing rarely reflects an individual’s likelihood of default.
AI could change that. Where lenders previously set prices on a car-by-car basis, prescriptive models will direct them towards tailor-made deals for the customer behind the wheel. Many platforms and lenders are already rolling it out. Laggards are risking irrelevance in an increasingly automated and competitive market.
From prediction to prescription
Predictive AI has already made inroads forecasting residual values and borrower behaviour. But a new generation of AI will recommend action as well as anticipating outcomes. Today, lenders are beginning to deploy prescriptive AI to sharpen pricing strategies.
One major use case is based on estimating current market value of used vehicles more precisely. Instead of relying on patchy guidebooks or gut feel, lenders can draw on AI models trained on vast data sources like auctions, dealer listings and even vehicle histories scraped from the internet. Similarly, lenders can predict future values with unprecedented granularity, ensuring contractual minutiae like lease-end buyout options are set on firmer financial ground.
Customer-level insights are another key use-case for prescriptive AI. By integrating behavioural data, lenders can better predict customers most likely to default or be lured away by marginally better competitor deals. Prescriptive models nudge lenders towards offering discounts to loyal, low-risk customers, while keeping rates firm for flightier buyers. Over time, this approach will wean the sector off a dominant blunt-edge pricing mentality, and equip lenders with unprecedented responsiveness to borrower demands.
The future promises even more disruption. Some large dealer networks are already experimenting with conversational agents that guide customers to the right car – and the financing needed to use it – based on lifestyle needs, driving habits and budget. And some buyers are already using AI concierges that scour national and even cross-border inventories, pinpointing the perfect vehicle and arranging financing and shipping in a single smooth transaction.
Finance operations will also get a rethink. Rather than exchanging payslips and bank statements over email, customers could simply grant lenders temporary access to their open banking data, allowing AI agents to assess income, spending habits, and risk profiles instantly. What now takes days of paperwork could be condensed into seamless, ten-minute journeys.
On the frontiers of an AI-assisted future
Far from the stuff of science fiction, many leaders in the vehicle financing space are embedding prescriptive AI for immediate customer use. Earnix, a longtime leader in banking and insurance analytics, is rolling out a “co-pilot” AI agent embedded within its pricing and personalisation platforms. “This is an AI agent which enables users to do more within the Earnix platform in a much more efficient way,” says Giovanni Oppenheim, Director of Banking Solutions at Earnix. “We are drawing on the experience of implementing this platform over 20 years in many different countries and continents and for different use cases.” It means lenders will be better able to automate complex pricing decisions, optimise loan structures and accelerate customer service.
The current version of the co-pilot acts much like a sophisticated internal assistant, streamlining tasks and spotlighting insights. But Earnix is setting the bar high in the industry by promising to go further. Over the next few years, it will integrate generative AI capabilities directly into predictive models, enabling relationship managers to query real-time customer profiles, default probabilities and personalised discount strategies at the point of sale.
The ambition is clear: insulate lenders against today’s market uncertainties and futureproof them against the evolving customer expectations of tomorrow. In a world where instant, tailored service is becoming the norm, those who rely on yesterday’s tools will find themselves left behind.
With prescriptive AI turbocharging an unrecognisable future for used car finance, lenders need to reorient their strategies today. Earnix can help them do just this. Fill in your details on this page to learn more.
