Artificially sincere? How chatbot language style shapes consumer perceptions across brand contexts

Purpose The present study aims to empirically investigate how the language styles of artificial intelligence (AI) chatbots, specifically their use of abbreviations, formality and message length, influence consumer perceptions of sincerity, effort and brand-related outcomes in service interactions. Design/methodology/approach Four experiments (n = 878) were conducted to test the effects of language style in various service and brand contexts. The studies manipulated chatbot communication features and measured their influence on perceptions of sincerity, engagement, brand attitudes and behavioral intentions. Findings The results suggest that abbreviation-heavy messages consistently reduce perceptions of sincerity and a service agent’s effort, especially when the agent is human. Informal and concise chatbot responses enhance engagement and perceived sincerity, but only when aligned with brand context. Informality strengthens brand attitudes and purchase intentions when psychological brand closeness is high (Study 3) and when the brand is positioned as premium (Study 4). Research limitations/implications Since the studies use mock interactions in online experiments, real-world generalizability may be limited. For validation, future research could explore these effects via longitudinal and field-based studies. Practical implications This study offers actionable insights for service designers and brand managers on tailoring chatbot communication to enhance consumer trust, engagement and brand alignment. Language style should be strategically aligned with brand positioning and the strength of the consumer–brand relationship. Originality/value This study advances the understanding of chatbot communication by framing language features as social signals, extending social response theory to AI-mediated service contexts and revealing boundary conditions for effective chatbot design.