How can AI resolve one of the biggest pain points in the used car market ?
A study released this month projects that the global automotive AI market will experience exponential growth over the next decade, expanding from $2.5 billion today to roughly $75 billion by 2030. AI and especially computer vision are well known for disrupting the automotive industry by the development of self driving cars (it is estimated that 1 out of 10 cars will be self driving by 2030). Though, there is another market branch of the automotive industry that could be disrupted in its entirety by AI : the used car market.
Indeed, the used-car market is often used to explain the principle and implications of the information asymmetry. This cycle leads to market inefficiency because transactions that would have benefited both buyers and sellers fail to take place. That is why, although sellers are willing to sell high-quality used cars to buyers at a fair price, the transactions do not occur because the buyers are unsure about the condition of the cars.
Of course, there are ways to reassure car buyers, like certifications by third parties, extensive research, third party databases, going to an auto mechanic, running a diagnostic test or checking the VIN report. But closing this asymmetric information gap has a cost: obtaining information can be costly both in terms of time and money (fees of third parties, time of research etc)
This is where the potential of AI comes in. In this $1,300Bn market (expected to grow at a 5,5% CAGR from 2020 to 2027), it could actually close the asymmetry and resolve this market failure at a low cost both in terms of time and money. But one question remains: How can AI bring full transparency in the used car selling process?
May it be implemented in marketplaces, auction companies, recyclers or DMS, AI can soften processes and bring the transparency that this market has been missing for decades.
Here are a few examples :
Audio Motor Profile & Virtual Lift
The first thing you might want to know when buying a car is what’s under the hood (and is also the most difficult thing to know). In this spirit, ACV Auctions, a tech driven US auction company, is developing 2 main inspection tools. With the Virtual Lift, they enable buyers to have an unbiased view of a vehicle’s undercarriage. Paired with the Audio Motor Profile, a feature that allows buyers to hear high-definition engine sounds of all listed vehicles, buyers can have a complete overview of what’s under the hood of the coveted car. Those tools are bringing trust through a digitized process, the data collected through images and sounds leaves room for future AI developments.
Visual Intelligence
The second step, that is crucial as well, is to inspect the said car and detect its flaws and dysfunctions. The classic process is an inspection made by the seller, describing, of its goodwill, the condition of a car. If the buyer wants to be reassured and have impartial information, he must pay for an inspection made by a third party (car expert), which takes time and costs money. This process has been improved over the years, in parallel with the improvement of technologies, including pictures, videos that will enable the buyer to have a better understanding of the car condition.
With AI, and especially computer vision, this process can be consequently softened and improved: Monk.ai provides a solution for the used car industry players (auction companies, marketplaces, DMS, recyclers) that mimics the human eyes with deep learning and computer vision algorithms. Working only from a few pictures, it instantly detects and classifies the condition of the exterior of a car. The algorithm is able to classify the type of damage, its location and severity and to generate a price range or a grade for the inspected car. Afterwards, the full report is sent in real time to the seller, who can then dispose of an unbiased, fully transparent report of the state of the car and remove the potential mistrust of the buyer.
By setting new auto remarketing industry inspection standards, those AI tools remove the human biais in the process and converge towards the end of this long lasting asymmetry.
Reliability Check
One of the other key points when buying a car is to be assured that it wasn’t subject to fraud. CarDr.com, well known in the USA for providing transparent and comprehensive used car inspections has recently raised $5 millions to reinforce its inspection processes with an AI powered vehicle health assessment tool. The AI analyzes vast amounts of car diagnostic data to more accurately assess vehicle health, prevent fraud, and determine a car’s true value. Following the inspection, the AI processes & compares the OBD (On Board Diagnostics) parameters & inspection results of the inspected vehicle to hundreds of vehicles to arrive at the “Reliability score” of the vehicle.
Coupled with CarDr’s vehicle inspection app, the AI accuracy is impressive: it detects odometer fraud at 99%, VIN fraud at 100% and generates a reliability rating and a fair market valuation within minutes.
Price Setting Process
The last step where asymmetry can interfere is the price setting process. The tools used today allow the regular processing of several million price statements and automate all the calculations of a quote. Nevertheless, disparities may still exist from one supplier to another. The level of quality of the estimates depends mainly on the coverage of the market, the intelligence of the algorithms developed and the repository used to identify and record all the finishes of used cars launched each year. This is why Autobiz, the European leader in car quotation, maintains its technological lead by devoting 15% of its turnover each year to R&D in order to continue improving its statistical models and has come up with intelligent algorithms able to set up the most accurate price in a few seconds. Thanks to these performant, unbiased algorithms, buyers do not need to spend hours comparing similar cars anymore to be sure they were offered a good price.
Conclusion
Those AI solutions are gradually being implemented by players in the used car market, boosting transparency and giving confidence to those who are buying or who are investing in it, as well as distributors, thereby alleviating the long lasting market asymmetry. But in this market, AI is not only a motor of transparency but is also used for optimizing the market, for example with online behavior analysis (Vroom acquiring Carstory to assemble predictive market data) or inventory management and thereby optimizing processes, boosting profitability and improving the customer experience.
