Ford has been forced to rehire hundreds of veteran workers after its aggressive bet on artificial intelligence to replace its most experienced employees cost the company billions of dollars in production errors.
The American carmaker has hired over the last three years more than 350 veteran engineers —whom it internally calls “grey beards”— to fix the faults generated by its automated quality-control systems. According to Bloomberg, these professionals now lead quality reviews and, in many cases, help train and improve the very artificial intelligence systems that had previously left them out.
“We had increasingly relied on the automated quality systems and weren’t getting the results we wanted,” acknowledged Kumar Galhotra, Ford’s chief operating officer. “We brought back technical specialists and they look for failure points before a part even reaches the production plant.”
The problem, according to the company’s executives, is that artificial intelligence lacks the nuanced judgment needed to solve complex problems. AI-based inspection systems can detect patterns and anomalies, but they don’t replace the experience accumulated over decades by an engineer who has seen hundreds of product cycles.
The lesson has been expensive, but the results are already showing. According to J.D. Power’s latest initial quality study —the sector’s annual benchmark— Ford has reached first place among mainstream brands for the first time in 16 years. That said, the company is still the one that accumulates the most recalls in the United States, although executives attribute these problems to past errors related to automation, not to the return of humans.
“Artificial intelligence is a fantastic tool, but it’s only as good as the information you train it with,” noted Charles Poon, Ford’s vice president of vehicle engineering. “In recent years, we didn’t pay as much attention as we should have to the experience of our most veteran engineers.”
The company assures it will not abandon the use of artificial intelligence, but plans to combine it with human oversight and expertise. A lesson many companies would do well to remember: AI is a powerful tool, but it doesn’t replace the judgment that years of experience bring.
— Marta, for inteligencia intermitente.
