ABSTRACT

Chatbots provide functional and social support in various contexts. They are often designed with humanlike features. This study examines how chatbots’ assigned names (humanlike vs. neutral vs. machinelike) and communication contexts (functional vs. social) influence users’ willingness to disclose personal information. We conducted a 3 × 2 “between-subjects” online experiment with random assignments of 299 participants. The results showed that a functional communication context elicited greater participants’ willingness to disclose information, but the impact of chatbot names was not significant. These findings provide an extended understanding of the Computers Are Social Actors paradigm and may inspire the exploration of conditional effects in privacy research. The practical implications for context-aware designs are discussed.

摘要

聊天机器人在各种情况下提供功能和社交支持。它们通常被设计成具有人性化的特征。本研究探讨了聊天机器人的指定名称（人类、中性、机器）和通信环境（功能性与社交性）如何影响用户披露个人信息的意愿。我们进行了 3 × 2“受试者间”在线实验，随机分配了 299 名参与者。结果表明，功能性沟通环境激发了参与者更大的披露信息的意愿，但聊天机器人名称的影响并不显着。这些发现提供了对计算机是社会参与者范式的扩展理解，并可能激发隐私研究中条件效应的探索。讨论了上下文感知设计的实际意义。