Why did artificial intelligence researchers shift toward engineering problems in the 1980s and 1990s?

During the 1980s and 1990s, artificial intelligence researchers increasingly shifted toward engineering-focused problems, primarily through the development and commercialization of expert systems.1

This shift was driven by the practical success of programs like DENDRAL and MYCIN, which demonstrated that AI could solve complex, specialized problems in fields such as medical diagnosis and chemical analysis by codifying human expertise into "if-then" rules. This process, known as knowledge engineering, allowed researchers to move beyond abstract laboratory experiments and create commercially viable "smart" systems.123

While early AI research often focused on the theoretical goal of artificial general intelligence, the focus on engineering enabled AI to emerge as a functional tool for industry and the military. This transition reflected a broader move toward applied AI, where the priority shifted from replicating human cognition in its entirety to building systems that could effectively perform specific, high-value tasks.312