The personal pronouns used by AI and AI gender jointly affect AI trust of users, and this interaction varies by interaction context. Study 1 (N = 161) explored the emotional communication context in which the AI responded to the user’s emotion regulation questions. Findings suggest that when a female AI used the pronoun “we,” it facilitated users’ perceived autonomy and subsequently fostered higher AI trust. Study 2 (N = 158) examined a persuasive communication context in which the AI disputed the user’s opinion about the legal drinking age, revealing that, for the female AI, the pronoun “we” was associated with lower agency perception, while for male AI, the pronoun “we” was associated with higher agency perception. This agency perception subsequently correlated positively with AI trust. In summary, to effectively build AI trust, AI natural language generation should be dynamically calibrated to both the conversational context and the programmed gender persona of the agent.
Wang et al. (Thu,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: