Moravec's Paradox highlights a counterintuitive aspect of AI development, where it is relatively straightforward to program computers to perform tasks that require what humans consider higher-order intelligence (such as playing chess or solving complex mathematical problems), yet remarkably challenging to replicate the sensorimotor skills and perception tasks that even young children can perform effortlessly. This paradox stems from the evolutionary observation that human high-level reasoning demands less computational power than the intricate processes governing sensory and motor functions, which have been honed over millions of years. The paradox profoundly impacts AI and robotics, indicating that substantial computational and engineering efforts are necessary when designing systems capable of interacting with the real world, such as autonomous robots and perceptual AI systems.
The term "Moravec's Paradox" was first articulated in the 1980s, gaining broader attention and popularity with the rise of robotics and AI research in the late 20th century. Its naming and exploration coincided with growing interest in replicating human-like intelligence and capabilities in machines.
Hans Moravec, a prominent figure in robotics and AI, is credited with formulating and exploring this paradox. His insights, along with contributions from researchers like Rodney Brooks and Marvin Minsky, have been integral to understanding the challenges inherent in replicating human faculties in AI systems.