Big ideas often start small, but Masahiro Mori’s concept of the “uncanny valley” had a particularly low-profile debut. In 1970, Mori, then a 43-year-old robotics researcher at the Tokyo Institute of Technology, published his now famous work not in a science journal but in an obscure Japanese periodical called Energy, owned by the oil company Esso. “It was an advertorial magazine,” he says. In that article, Mori envisioned a time when robots would become so sophisticated that they would look almost exactly like humans. But that “almost” was a problem. Whatever else those future bots might prove to be, he said, one thing was certain: They would strike us as monstrous. Mori created a chart describing how our degree of identification and empathy with inanimate objects increases as their appearance approaches our own—we relate more to stuffed animals, for instance, than to industrial robots. But at a certain level of near-humanness, our affinity falls off a cliff. Mori dubbed this the uncanny valley. Today, that insight is a pressing concern for designers of virtual humans—not just in robotics but also in CG animation for film and videogames. via Wired
Wired talked to Mori, now 84, about the origins of his idea. Continue reading.


