抄録
The primary goal of this research is to gain insights into the influence of agents on human decision-making. We aim to ascertain whether individuals comprehend and subsequently modify their own opinions in response to the perspectives presented by agents. The central focus is on debating, where participants take opposing positions on a given issue. Debates entail advocating for or against a motion, with judges determining the more logical side. Current research explores the domain of artificial intelligence (AI) capable of participating in debates with humans, commonly known as 'debating AI'. Despite considerable efforts to enhance the performance of debating AI, there exists a noticeable research gap concerning how agents bring about changes in opinions through intervention. This study meticulously examines how individuals' opinions undergo transformation following agent interventions in an online debate setting. The comprehension of effective intervention methods becomes crucial for the successful implementation of agent interventions.
