Ford Brings Back Its 'Greybeard' Engineers After AI Misses on Quality

The automaker wagered that automated inspection could stand in for decades of human expertise, then quietly reversed course.

Ford has spent the past three years rehiring the veteran engineers it once expected artificial intelligence to replace, after the automated quality systems it rolled out across its factories failed to match what experienced inspectors could catch by eye.

The company brought back roughly 350 senior engineers, the kind it refers to internally as "greybeards," recruiting them from its own retirees and from the ranks of its suppliers. Their job is part repair crew and part teaching faculty. They catch design and manufacturing flaws before a part ever reaches the assembly line. They also mentor younger staff and retrain the AI tools that had been missing defects on their own.

Charles Poon, Ford's vice president of vehicle hardware engineering, was blunt about where the strategy went wrong. The technology, he told reporters, is "only as good as the information you use to train it." Ford, he admitted, had assumed that feeding its design requirements into an AI system would be enough to guarantee a quality car. It was not.

That admission is striking for a company that, only a year ago, was describing AI in sweeping terms.

A very human fix for a software problem

Chief operating officer Kumar Galhotra told investors last autumn that Ford was deploying AI across its entire industrial system, including some 900 AI-equipped cameras installed on plant floors to spot quality issues at their source and head off supply disruptions. The pitch matched the wider mood on Wall Street, where investors have rewarded almost any promise that automation will widen margins.

The cameras and the algorithms could see. What they could not do was judge.

Experienced inspectors carry a kind of knowledge that does not live in any manual. They know which odd reading signals a real defect and which is harmless noise, because they have watched the same failure modes play out across many product cycles. An AI model trained on documented requirements never absorbs that instinct, because the instinct was never written down. Worse, Ford discovered that many of its most knowledgeable engineers had already left the company before their expertise could be captured, taking decades of pattern recognition out the door with them.

The damage ran in two directions. The automated tools missed defects, and the junior engineers who might have learned the craft from veterans had been handed software instead of mentors. The apprenticeship pipeline that turns a new hire into a seasoned troubleshooter had quietly broken.

What the turnaround looked like on the floor

Rather than retreat from AI, Ford repositioned the veterans as its trainers. The greybeards now feed their judgment into the machine-learning systems while coaching the next generation, an approach the company frames as pairing seasoned engineering with advanced technology.

The results showed up in the metric the industry treats as its scorecard.

Ford finished first among mainstream brands in the 2026 J.D. Power Initial Quality Study, a position it had not held since 2010. The climb was steep. The brand had sat near the bottom of the mainstream pack only three years earlier, and Ford says its infotainment systems, long a sore point for owners, posted the largest single improvement in the ranking. The company also layered on extra scrutiny where it counted, adding more than a thousand new inspections and dozens of new tests for a single 2026 model built at its Kentucky Truck Plant, and assigning scores of additional inspectors to the line.

Chief executive Jim Farley said the renewed focus on quality had trimmed Ford's warranty and recall bills by hundreds of millions of dollars. The savings matter, because Ford has been carrying the highest recall volume in the industry, a figure executives now expect to decline as the upfront fixes work their way into newer vehicles.

Ford isn't alone in walking it back

The reversal slots into a pattern that has been building across corporate America for more than a year.

Klarna offered the loudest example. The Swedish payments firm boasted in 2024 that its OpenAI-powered chatbot was doing the work of 700 customer service agents. The company cut its headcount by roughly 40% and froze hiring, while the bot handled two-thirds of inquiries and slashed resolution times from minutes to seconds. Then customer satisfaction sank on the harder cases. By early 2026, chief executive Sebastian Siemiatkowski was hiring people back, conceding that an obsession with cost had produced one plain result. "What you end up having is lower quality," he told Bloomberg. Klarna now runs a hybrid operation, letting the AI field routine questions while humans take the interactions that call for judgment.

Others tell similar stories. McDonald's pulled an automated drive-through ordering trial after a run of viral mistakes and put human staff back on the speaker. IBM shifted a large slice of its HR work onto an AI system, then found the software could not handle the off-script cases that needed discretion.

The research has caught up to the anecdotes. Gartner has forecast that by 2027, half of the companies that cut customer service jobs because of AI will end up rehiring, and that more than 40% of agentic AI projects will be scrapped before they reach completion. A separate figure circulating among executives is starker: 55% of firms that carried out AI-driven layoffs now say they regret the decision. An IBM survey cited by Fortune put the success rate lower still, with only one in four AI projects delivering the return it promised.

A rough consensus has formed in the wreckage. The companies seeing durable gains are the ones using AI to make their people faster rather than to clear them out, pairing automated speed on routine work with human oversight on anything that carries risk.

The common thread is not that the technology fails to work. It is that firms treated automation as a clean substitution for people, then discovered they had pulled out the human judgment the entire system quietly depended on.

The lesson buried in the buzz

Ford's experience cuts against the confidence its own leadership projected at the peak of the hype.

"AI will leave a lot of white collar people behind," Farley said in an interview last June. A year later, the people his company was working hardest to recruit were the seasoned engineers it had allowed to leave, brought back to teach the technology the judgment it lacked.