Ford has reportedly rehired more than 350 experienced engineers after AI-led quality initiatives failed to deliver the improvements the automaker expected. According to a Bloomberg report, Ford’s chief operating officer Kumar Galhotra said the company had leaned heavily on automated quality systems, with disappointing results. The returning specialists now work to catch potential defects long before a vehicle component reaches the assembly line.
Charles Poon, Ford’s vice president of vehicle hardware engineering, was open about the miscalculation: “Mistakenly we thought that by just introducing artificial intelligence and ingesting the design requirements that we had, that that would produce a high-quality product.” The US automotive industry leader had underestimated the value of human experience, as the AI was evidently incapable of filling the necessary knowledge gap to effectively execute the designs.
Human Experience is Still a Strategic Asset
AI can automate processes, but it cannot automatically inherit institutional knowledge. Experienced employees build judgement, pattern recognition and the ability to handle exceptions over years on the job, none of which transfers to a model simply by feeding it design requirements or policy documents. When organisations let experienced people go before that knowledge has been properly transferred, AI will be missing this potentially crucial information.
Ford’s rehired “gray beard” engineers illustrate the point well. They aren’t being brought back to permanently replace AI; they’re being used to train younger staff and reprogram the AI tools themselves. The acknowledgement is simply that its AI has more to learn and it’s not clear how long these veteran engineers will be needed.
The Lesson for Customer Experience
The same dynamic is playing out across contact centres. Many organisations are deploying AI agents, automating customer interactions and reducing reliance on experienced advisors. Yet those advisors often hold knowledge that never makes it into a knowledge base, including how to handle complex or emotionally charged scenarios, when to apply policy with discretion rather than by the letter, how to navigate undocumented workarounds, and how to resolve unusual cases through accumulated experience rather than a flowchart.
Without capturing that expertise first, AI risks repeating some of the biggest mistakes companies make with AI in customer experience, delivering interactions that are efficient on paper but poorer in practice.
Research Reinforces Ford’s Experience
Recent academic research backs this up. A 2026 paper by researchers from the University of St. Gallen found that organisational knowledge is fragmented across systems, documents and individual employees’ tacit expertise. As AI takes on more decision-making responsibility, the researchers argue, organisations must ensure that knowledge is genuinely accessible to both humans and AI systems. The challenge, they suggest, is no longer simply deploying AI, but working out how knowledge should be maintained and when humans need to stay in the loop. The paper draws on manufacturing quality inspection to make its case, making it especially relevant to Ford’s situation.
The Way Forward
Ford isn’t abandoning AI. It’s pairing it with experienced engineers to improve quality and strengthen the knowledge base its AI systems depend on, which is worth factoring into effective workflow designs for human and AI collaboration in CX. There is a direct translation for CX leaders that before replacing people with AI, organisations should first capture, preserve and structure the expertise that makes great customer experiences possible. More generally, this ties into a prevalent theme right now in CX AI of adoption being the beginning, not the end. Deployment is just the first step on a long and evolving path to positive outcomes. In order to get there, for the time-being at least, human-AI collaboration is vital.