Constructing Ethical AI Based on the “Human-in-the-Loop” System

Abstract

The Human-in-the-Loop (HITL) system was first proposed by Robert Monarch, a machine learning expert. It adopted a “hybrid” strategy combining human intelligence and machine intelligence, aiming to improve the accuracy of machine learning models and assist human learning. At present, there have been a number ethical design attempts based on the HITL system, and some progress has been made in the ethical choices of disaster rescue robots and nursing robots. However, there is no analysis of why the HITL system can serve as an effective path in constructing ethical AI and how it can implement the efficiency of AI in ethical scenarios. This paper draws on the feasibility of the HITL system and analyzes how ethical AIs are possible when using the HITL system. We advocate for its application to the entire process of ethical AI design.