Large Language Model-powered conversational agents (CAs) are increasingly capable of projecting sophisticated personalities through language, but how these projections affect users is unclear. We thus examine how CA personalities expressed linguistically affect user decisions and perceptions in the context of charitable giving. In a crowdsourced study, 360 participants interacted with one of eight CAs, each projecting a personality composed of three linguistic aspects: attitude (optimistic/pessimistic), authority (authoritative/submissive), and reasoning (emotional/rational). While the CA's composite personality did not affect participants' decisions, it did affect their perceptions and emotional responses. Particularly, participants interacting with pessimistic CAs felt lower emotional state and lower affinity towards the cause, perceived the CA as less trustworthy and less competent, and yet tended to donate more toward the charity. Perceptions of trust, competence, and situational empathy significantly predicted donation decisions. Our findings emphasize the risks CAs pose as instruments of manipulation, subtly influencing user perceptions and decisions.

Increased levels of user control and feedback incorporation in learning systems is commonly cited as good AI development practice. However, the evidence as to the exact effect of perceived control over trust in these systems is mixed. This study investigates the relationship between different dimensions of trust and perceived control in postgraduate student burnout support chatbots. We present an in-between subject controlled experiment using simulated therapy-goal learning to study the effects of goal editing and feedback incorporation on perceived agent benevolence and competence. Our results showed that perceived control was moderately positively correlated with benevolence (r = 0.448, BF10 = 7.150), and weakly correlated with competence, and general trust.