Abstract
This study explores the effectiveness of AI chatbots in public safety advertising campaigns against marijuana-impaired driving, specifically focusing on the role of hypocrisy induction, gender, and linguistic styles. In our online experiment, participants engaged with a chatbot delivering either hypocrisy-inducing or neutral messages about driving under marijuana influence. The chatbot, characterized as male or female, utilized either formal or casual language. Results revealed that hypocrisy induction is most effective when delivered by a casually speaking female chatbot or a formally speaking male chatbot. The heightened guilt among participants explains the effectiveness of this approach. The Language Expectancy Theory (LET) serves as a theoretical framework, elucidating how the alignment of chatbot gender and linguistic style enhances the hypocrisy induction effect. This research highlights the potency of AI chatbots in leveraging public safety advertising campaigns and offers guidance on optimizing their design by aligning gender and language style with campaign goals.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call