Summary: When interacting with generative-AI bots, users engage in six types of conversations, depending on their skill levels and their information needs. Interfaces for UI bots should support and accommodate this diversity of conversation styles.
Through analyzing 425 interactions with generative-AI bots like ChatGPT, Bing Chat, and Bard, we've discovered that conversations could involve many vague, underspecified prompts or few, razor-sharp ones. Why does this matter? First, different conversation types serve distinct information needs and demand varied UI designs . Second, there is no one optimal conversation length — both short and long conversations can be helpful, as they might support different user goals.
In May and June 2023, we conducted a 2-week diary study involving 18 participants who used ChatGPT 4.0, Bard, and Bing Chat. Participants logged a total of 425 conversations and rated each for helpfulness and trustworthiness. At the end of the study, we conducted in-depth interviews with 14 participants.
The findings of this study are reported in several articles:
