Even the scrapyard is utilizing AI now.

      Copart, curated by the author

      The most significant car news and reviews, without the fluff.

      Our complimentary daily newsletter delivers the stories that truly matter to you every weekday.

      Copart is a massive, global car salvage company, and currently, its business is thriving. The firm announced that its revenue, gross profit, and net income have increased for both the quarter and the year. During a recent earnings call, analysts and Copart CEO Jeff Liaw discussed how the auto sale, salvage, and auction company is utilizing AI.

      In Q4 of 2025, it seems that every substantial business is experimenting with large language models to some extent. For an operation like Copart, which handles a significant volume of transactions, many of which are alike, the adoption of AI is particularly logical. At this stage, it appears that instead of using tools to physically examine cars like in the rental car industry, Copart is leveraging AI to perform tasks such as pricing vehicles and determining if a car is considered a total loss based on data.

      This means they can compare the estimated repair costs to the vehicle's value much faster than before.

      What’s intriguing to me is that AI has the potential to influence insurance rates and vehicle valuations worldwide. It will also impact the global distribution of used cars to some extent.

      While the severity of car accidents seems to be decreasing, the rate at which cars are declared totaled is on the rise. Copart's team attributes this to the increasing complexity of modern vehicles, which quickly become uneconomical to repair compared to older models. For instance, a taillight for a 2024 F-150 costs over $1,000. Liaw noted, “The frequency of total losses has generally increased at a rate that far exceeds the decline in accident frequency itself. For the quarter, total loss frequency has continued its long-term upward trend, consistent with our company's history. In the U.S., for the second calendar quarter of 2025, total loss frequency was 22.2%, up from 21.5% in the same quarter of 2024.”

      Typically, totaled cars are exported overseas, where cheaper labor and a new title make restoration more financially viable. “International buyers account for about 40% of all vehicles sold at Copart’s U.S. auctions, representing nearly half of auction revenue, as these international buyers generally purchase higher-value vehicles than domestic buyers,” Liaw mentioned during another part of the call.

      Wholesale car auctions like Copart maintain a distinctive relationship with sales data and the repair feasibility of vehicles, as they both react to and establish those values. For example, Copart understands the worth of a 2005 Honda Civic LX with 150,000 miles because it has facilitated the sale of thousands of such vehicles. Buyers rely on the auction’s average transaction prices to determine what constitutes a good deal, and this same figure influences whether a car is deemed repairable or a total loss. It's a self-reinforcing cycle.

      While none of this is groundbreaking or novel, it’s fascinating to consider AI accelerating the speed and scale at which all of this is processed, ultimately affecting vehicle valuations. Given that AI can evaluate cars using more data than ever before, it will emerge as the ultimate authority on vehicle value, determining what gets totaled versus repaired and, consequently, the insurance costs for specific models.

      Before the rise of LLMs, algorithms and averages were already accomplishing this, but now everything will occur at a quicker pace.

      Here’s what Mr. Liaw had to say regarding Copart’s use of AI, prompted by a question from CJS Securities/Baird analysts Bob Labick and Chris Bottiglieri:

      “…we provide many of our sellers with tools that enable them to make instant total loss decisions informed by literally millions of similar vehicles we’ve sold over the years. Those decision support tools are greatly enhanced by current-generation large language model technologies. Furthermore, it’s evident in areas such as customer and agent support here at Copart.”

      “Even personnel who engage daily with members and buyers have access to better information backed by LLM technology. While we are still, like many companies, in the early stages, we already have various applications of that technology deployed today. We also utilize it at the auction level. When we consider the products and vehicles we recommend to our buyers, along with search results, all these different aspects are informed by AI as it stands. I have no doubt that as the tools evolve and our implementations become more sophisticated, it will enhance our operations, resulting in significant efficiency gains in service delivery. It will undoubtedly unlock future opportunities as well.”

      “I think it’s accurate to say it allows us to operate more efficiently, shortening cycle times for our clients and for us. You heard Leah explain how inventory is impacted. One reason for the reduction in inventory is our Title Express service, which we provide on behalf of our insurance clients, and we are doing so for more clients with each passing quarter. We are achieving