Ford has rehired human engineers after admitting its AI-driven quality checks failed to meet expectations. The company brought back veteran “quality” staff—over 300 inspectors and hundreds of experienced engineers in recent years—after automated systems were not able to replicate the judgment and expertise of long-tenured engineers across product cycles.

Ford’s leadership framed the issue as a training-data problem rather than an AI system being inherently “broken.” Charles Poon, vice president of vehicle hardware engineering, said the company initially overestimated what AI could achieve by ingesting design requirements, noting that AI is only as good as the information used to train it. He added that the organization had not paid enough attention to preserving the knowledge held by its most experienced engineers, many of whom had left before their expertise could be used to improve the systems.

According to reporting Ford adopted AI in parts of its operations—including quality checks on the production side—aiming to cut costs and increase productivity. The company also pointed to AI-powered equipment deployed in plants, such as cameras intended to detect quality issues early and reduce supply disruptions. But Poon said the automated checks did not live up to expectations, and Ford concluded it needed the experience of veteran technicians/engineers to retrain and refine its tools.

Ford said it reintroduced these experienced workers not only to perform quality reviews, but also to help mentor younger staff and directly train the automation and machine-learning/artificial-intelligence systems. The company described this as a “talent refresh” tied to improving overall vehicle quality.

The effort has coincided with Ford returning to the top of an industry quality benchmark: it said it is number one among mainstream automakers in the US JD Power Initial Quality Study—a position it has not held since 2010—crediting the renewed focus on best-in-class quality and the return of veteran expertise.