Ford brought back 350 engineers to correct the errors made by its AI systems.

      Ford reinstated 350 engineers after its AI systems fell short of meeting the expected quality levels. Charles Poon, Ford’s VP of vehicle hardware engineering, indicated that the automaker had mistakenly thought it could replace experienced personnel with AI and still maintain high product quality. This acknowledgment, initially reported by The Verge, coincides with Ford achieving the top ranking among mainstream brands in JD Power’s initial quality survey for the first time in 16 years.

      Poon clarified that the issue wasn’t that the AI was defective, but rather that seasoned workers left before transferring their crucial knowledge into the systems meant to take their place. Without the necessary engineering insights reflected in the training data, Ford’s automated tools inadvertently magnified weak information instead of identifying design issues. As a result, the company rehired, newly recruited, or promoted 350 skilled engineers to address this shortfall.

      While Poon did not elaborate on the reasons those workers departed, the overall situation is clearer. Ford has eliminated around 5,300 salaried jobs since reaching its employment peak in 2020, part of a larger trend among Detroit’s automakers that has resulted in over 20,000 white-collar job losses. CEO Jim Farley has openly stated that AI "is going to replace literally half of all white-collar workers in the US," a forecast that is now complicated by the quality challenges faced by his own company.

      The 350 returning engineers were assigned to mentor junior staff, reconstruct the data pipelines that support Ford’s AI training, and enhance the automated systems that were initially slated for replacement. Additionally, Ford established a dedicated software quality assurance team of 40 members and introduced over 100,000 AI-driven automated tests to identify edge cases and validate software changes late in the development process.

      This turnaround was sufficient for Ford to secure the top position in JD Power’s 2026 initial quality study, which evaluates problems reported by vehicle owners within the first 90 days of ownership. Ford recorded 152 problems per 100 vehicles, surpassing Nissan and Buick. The F-150, Mustang, and Super Duty models each earned the best-in-segment title for the second consecutive year.

      However, this quality victory does not negate a generally poor track record. Ford has led US automakers in recalls this year, issuing 51 so far in 2026, encompassing over 11 million vehicles—more than double that of the next closest manufacturer. It also aligns with a growing number of companies recognizing that the absence of human judgment in AI-driven processes results in challenges that the technology alone cannot resolve.

      This situation unfolds at a time when AI companies and policymakers are hurriedly trying to determine what this transition means for the workforce. OpenAI, Anthropic, Amazon, and Microsoft recently supported RAISE US, a $500 million nonprofit spearheaded by former commerce secretary Gina Raimondo, aimed at retraining American workers for the AI economy. Ford’s experience indicates that the more significant issue may not be retraining, but rather recognizing which workers are indispensable from the outset.