Principles for Designing an AI Competition, or Why the Turing Test Fails as an Inducement Prize

Abstract

If the artificial intelligence research community is to have a challenge problem as an incentive for research, as many have called for, it behooves us to learn the principles of past successful inducement prize competitions. Those principles argue against the Turing test proper as an appropriate task, despite its appropriateness as a criterion (perhaps the only one) for attributing intelligence to a machine.