Can Machines Imitate Humans? Integrative Turing-like tests for Language and Vision Demonstrate a Narrowing Gap

TL;DR

This study benchmarks AI's ability to imitate humans across language and vision tasks, revealing that current AI systems are nearing human-like deception capabilities and emphasizing the importance of imitation as an independent evaluation metric.

Contribution

The paper introduces large-scale datasets and Turing-like tests for evaluating AI's human-likeness in language and vision, providing new benchmarks and insights into AI imitation abilities.

Findings

AI can convincingly imitate humans in language and vision tasks.

Simple AI judges outperform humans in detecting AI responses.

Imitation ability is weakly correlated with traditional AI performance metrics.

Abstract

As AI becomes increasingly embedded in daily life, ascertaining whether an agent is human is critical. We systematically benchmark AI's ability to imitate humans in three language tasks (image captioning, word association, conversation) and three vision tasks (color estimation, object detection, attention prediction), collecting data from 636 humans and 37 AI agents. Next, we conducted 72,191 Turing-like tests with 1,916 human judges and 10 AI judges. Current AIs are approaching the ability to convincingly impersonate humans and deceive human judges in both language and vision. Even simple AI judges outperformed humans in distinguishing AI from human responses. Imitation ability showed minimal correlation with conventional AI performance metrics, suggesting that passing as human is an important independent evaluation criterion. The large-scale Turing datasets and metrics introduced here offer valuable benchmarks for assessing human-likeness in AI and highlight the importance of rigorous, quantitative imitation tests for AI development.