Exploring Human-Like Behavior Explanation on AI Speaker Recognition as Communicable Partners across Age Groups

Abstract

AI speakers are not just tools for presenting information but also have the potential to function as communication partners. Pradhan et al. (2019) reported that individuals differ in their understanding of AI speakers, with some perceiving them as radio-like machines while others see them as human-like social agents.This distinction provides important insights into building trust with AI speakers. The objective of this study was to develop fundamental knowledge that enables AI speakers and humans to establish trustworthy relationships, akin to those between humans.We examined whether participants' behavior towards the AI speaker differed when they were informed that it operated as an autonomous entity, similar to an agent robot. Additionally, we investigated whether these results were independent of age. Forty younger and forty elderly participants were involved in a joint Simon task experiment. The hypotheses were as follows: (1) Participants who were informed that AI speakers "think and make decisions on their own, like humans" would strongly perceive AI speakers as communicative partners compared to participants who were informed that AI speakers were "programmed to act like machines." (2) The aforementioned differences would persist irrespective of individual characteristics, such as age or experience with robots. The experiment's findings supported both hypotheses.