How AI is Transforming Vehicle Information
From Science Fiction to Game Changing Reality for Automotive Remarketing
Wednesday January 11, 2023
Artificial intelligence (AI) has been called the most transformative technology in the history of mankind. And yet some believe it’s still the stuff of science fiction, conjuring up images of human cyborgs and scenes from The Matrix. Even still, 94% of surveyed business leaders view AI as critical to their success over the next five years, and that’s particularly true for auto remarketing. Before we dive into that specific use case, let’s be sure we’re all on the same page about what AI is.
Simply put, artificial intelligence is anything that allows a computer to make a decision. This means codifying a set of rules for how decisions should be made, much like a flow chart. Historically, these rules were written by programmers; today, more often than not, they are written by the computer itself. The latter process, called machine learning, uses algorithms to create statistical models which mimic the problem-solving and decision-making capabilities of the human mind based on input data. The more data, the better the prediction.
Today, we are surrounded by examples of AI making our lives easier and more productive, from predictive text messaging and GPS to the ability to search photos on our smartphones by subject. In the automotive space, examples include assisted braking, backup camera sensors and autonomous driving vehicles.
As interest and investments in AI technology gain traction across a myriad of industries, misconceptions remain about what role humans will play. To be certain, artificial intelligence is a powerful tool, but it doesn’t exist in a vacuum.
From deciding what and how the machines will learn to giving feedback on their progress and translating their output into practical real-world decisions, people are a vital element to successfully deploying AI and machine learning to build a complete end-to-end product.
AI in Auto Remarketing
Since 1945, when a Ford dealer from Manheim, Pennsylvania began auctioning cars, condition reports have been a never-ending challenge for the used car industry. When COVID-19 led to a more digital-centric auction environment, concerns about losing the ability to view inventory in person unsettled dealers who relied on the more traditional ways of evaluating inventory.
With client confidence in the digital wholesale marketplace hinging on accurate and consistent vehicle information, this area has seen the greatest focus and investment from Manheim. From improving everything condition reports to images and beyond, we have had the singular goal of giving buyers the information they need to make confident purchase decisions—and AI is, and will continue to be, a big part of that.
AI’s data-driven approach produces more consistent results by removing the human subjectivity and catching things that the naked eye might miss. Its superpower lies in the ability to continuously improve as human feedback helps the technology learn and produce better outputs. And since Manheim and Fyusion have more data than anyone in the industry to train our AI model, clients will ultimately benefit from the most accurate automated damage detection in the industry.
While that particular functionality is still on the horizon, the good news is that AI is already behind several advancements in vehicle imaging that clients see today, from delivering truer colors and representations of the vehicle in mobile imaging applications to actively selecting the best 12 images out of thousands captured through fixed imaging tunnels.
What the Future Holds for Vehicle Information
It’s an exciting time for the evolution of AI in auto remarketing. Damage detection, in particular, is on the cusp of a dramatic change for how vehicle information is gathered and shared.
In 2022, Manheim introduced the wholesale industry’s most advanced imaging yet, reinventing how vehicles are captured and paving the way for condition reports that are more consistent than ever before. It uses Fyusion’s 3D imaging technology to look at vehicles from every angle, allowing it to assess damages more accurately than solutions that rely on outdated 2D images.
And we’ve already spent years training the next-level automated damage detection model, seeing phenomenal results in our testing. As that model continues to learn and evolve, condition reports will look much different. Instead of just still photos, interactive images will be color-coded for more exacting representations of the damage detected—and information will be curated to provide a deeper understanding of a vehicle’s condition.
When it comes to vehicle remarketing, the needs of today’s dealers are really no different than in 1945. They want transparent information about the condition of a vehicle that they can rely on to make an informed decision and achieve their profitability goals. What has changed, however, are the immense advances in technology which make it possible for us to reimagine vehicle information and deliver what the industry has been asking for: the most consistent and reliable vehicle information possible.
Brad Burns, associate vice president of vehicle information for Manheim
Stephen Miller, chief product officer and co-founder of Fyusion