Loss of Skill Acquisition Opportunities for Junior Engineers and Challenges in Maintaining Expertise via AI

Drawing on his experience designing control systems for nuclear power plants, Mitchell revealed that design decisions were made to intentionally leave manual operation steps within the system. He stated that this was a deliberate inefficiency intended to prevent "skill degradation," where operators lose the ability to understand the actual situation on the ground as automation progresses.

The proliferation of AI is creating similar concerns. According to research from Harvard University, companies that introduced generative AI saw a decrease in junior-level employment of approximately 9% within six quarters of implementation compared to companies that did not. Meanwhile, the results show that senior-level employment continues to grow.

An analysis by Stanford University also indicated that in occupations heavily impacted by AI, employment of young workers has decreased since the end of 2022. This trend is concentrated in scenarios where AI replaces tasks; in scenarios where AI merely assists with tasks, junior-level employment remains flat or shows an increasing trend.

Expertise is acquired through the accumulation of practical experience, such as failure and debugging. It has been pointed out that as AI takes over these processes, there is a risk that the next generation of senior professionals—those capable of intuitively detecting errors in models—will fail to develop.