Ford's decision to bring back more than 350 experienced engineers illustrates an increasingly important principle of industrial artificial intelligence. Competitive advantage no longer depends solely on deploying advanced AI models but on combining automation with high-quality proprietary data and decades of engineering expertise.

For much of the past decade, artificial intelligence has been presented as the inevitable replacement for repetitive engineering work. Automotive manufacturers have invested billions in automation, predictive analytics and machine learning, expecting software to shorten development cycles, eliminate human error and improve manufacturing quality simultaneously. The underlying assumption appeared logical: once algorithms become sufficiently capable and are trained on enough historical information, engineering knowledge itself becomes scalable.

Ford's latest decision suggests the transition is proving considerably more complicated.

The company has acknowledged that excessive reliance on automated engineering tools contributed to quality problems significant enough to require the return, promotion or hiring of more than 350 experienced engineers. Rather than abandoning artificial intelligence, Ford is attempting to rebuild something that modern technology unexpectedly exposed as scarce: institutional knowledge accumulated over decades of vehicle development.

The story is therefore less about the limitations of AI itself than about an economic misconception that has spread well beyond the automotive industry. Automation can process information at extraordinary speed, yet it cannot compensate for knowledge that has never been properly captured in the first place.

Automation Solved One Problem While Creating Another

Manufacturers increasingly view automation as a solution to rising development costs, growing software complexity and competitive pressure. Vehicle platforms now contain millions of lines of software code alongside increasingly sophisticated electronic systems, making traditional quality assurance both expensive and time-consuming.

Ford expected artificial intelligence to assume a growing share of routine engineering verification while improved design specifications would naturally increase production quality. According to Charles Poon, Vice President of Vehicle Hardware Engineering, the company believed that combining AI with more precise engineering requirements would allow many traditional inspection processes to become significantly more efficient.

Instead, the opposite occurred.
The company discovered that automation performed well only when supported by high-quality engineering data and accumulated practical expertise. Once experienced specialists left the organisation before their knowledge had been transferred into digital systems, artificial intelligence began operating with an incomplete understanding of real-world manufacturing challenges.
This distinction is critical because machine learning models do not create engineering expertise independently. They identify patterns within existing information. If the underlying knowledge base contains gaps, the algorithm simply reproduces those gaps at greater speed.

The Cost of Losing Experience

Ford's response illustrates how industrial companies are reassessing the value of experienced personnel.
More than 350 engineers have now been recruited, promoted or brought back to strengthen engineering teams. Their responsibilities extend beyond solving current technical issues. They are expected to mentor younger engineers while simultaneously improving the quality of data used to train AI systems.
This reflects an increasingly important shift in industrial strategy.

For several years, many organisations viewed experienced engineers primarily as labour costs. Today they are increasingly recognised as repositories of tacit knowledge that cannot easily be documented in manuals or databases.
Experienced engineers frequently recognise production anomalies before measurable indicators appear. They understand how suppliers behave under stress, how seemingly insignificant design modifications affect downstream manufacturing, and how previous product generations failed under particular operating conditions. Much of this expertise exists as practical judgement rather than structured documentation.
Artificial intelligence cannot replicate knowledge that was never formalised.

Ford's Quality Problems Did Not Have a Single Cause

The company does not attribute declining vehicle quality solely to automation.
Ford identifies several contributing factors, including difficult launches of the Explorer and Aviator models, supply chain disruptions during the COVID-19 pandemic and insufficient coordination between internal divisions.

Chief Operating Officer Kumar Galhotra acknowledged that the company relied for too long on identifying and correcting defects after they appeared rather than preventing them during earlier development stages.
That admission reflects a broader transformation occurring throughout advanced manufacturing.

Increasingly, manufacturers recognise that quality cannot simply be inspected into finished products. It must be designed, tested and validated throughout the entire development process.
Artificial intelligence can accelerate many of these activities, but it cannot replace engineering judgement where uncertainty remains high.

AI Is Becoming a Tool Rather Than a Substitute

Ford's revised strategy does not represent a retreat from artificial intelligence.

Instead, the company is integrating software engineers, manufacturing teams and suppliers far more closely throughout product development. A dedicated software validation team of approximately 40 specialists now focuses specifically on testing vehicle software before customer delivery.
The company has also expanded automated testing dramatically by introducing more than 100,000 AI-driven software tests designed to identify rare failures and improve system resilience before vehicles reach consumers.

This represents a more mature understanding of automation.
Rather than expecting artificial intelligence to replace engineers, Ford increasingly treats AI as an amplifier of engineering capability. Algorithms perform repetitive validation at enormous scale, while experienced specialists interpret results, identify anomalies and continuously improve the underlying datasets that make automation effective in the first place.
The distinction may appear subtle, but it fundamentally changes the role of artificial intelligence inside complex industrial organisations.

During the first wave of industrial artificial intelligence, competitive advantage was widely believed to depend on acquiring increasingly sophisticated algorithms. Today, many companies are reaching a different conclusion. Algorithms are becoming more accessible, while high-quality proprietary engineering data has emerged as the true strategic asset.

Artificial intelligence learns from examples rather than experience. If historical engineering decisions, production failures and design compromises have not been systematically documented, even the most advanced model operates with an incomplete understanding of reality. 

Another lesson emerging from Ford's strategy concerns the timing of quality assurance.
Traditional manufacturing often relied on identifying defects once production had already begun. Problems were corrected after discovery, recalls were issued when necessary, and quality departments focused on detecting failures before products reached customers.
As electronic systems become more complex, identifying problems after production becomes increasingly expensive. Software errors may interact with hardware, suppliers and manufacturing processes in ways that are difficult to isolate once thousands of vehicles have already been delivered.

Ford is therefore shifting quality control much earlier in the development process. Software engineers now work more closely with production teams and suppliers, while automated validation is performed before vehicles reach customers rather than after defects begin appearing in the field.
Ford's experience challenges one of the most persistent assumptions surrounding artificial intelligence: that automation inevitably reduces the need for experienced professionals.

By Claire Whitmore 
June 26, 2026

Join us. Our Telegram: @forexturnkey
All to the point, no ads. A channel that doesn't tire you out, but pumps you up.