A Definition and a Test for Human-Level Artificial Intelligence
This paper addresses the fundamental problem of defining and testing human-level AI for the entire AI research community.
This paper proposes that human-level artificial intelligence (HLAI) can be defined by its ability to learn from others' experiences through language. It suggests that language acquisition without explicit rewards could serve as a sufficient test for HLAI.
Despite recent advances of AI research in many application-specific domains, we do not know how to build a human-level artificial intelligence (HLAI). We conjecture that learning from others' experience with the language is the essential characteristic that distinguishes human intelligence from the rest. Humans can update the action-value function with the verbal description as if they experience states, actions, and corresponding rewards sequences firsthand. In this paper, we present a classification of intelligence according to how individual agents learn and propose a definition and a test for HLAI. The main idea is that language acquisition without explicit rewards can be a sufficient test for HLAI.