Abstract
In this short paper, we employ a month-long autoethnography to investigate the utilization of ChatGPT through metaphor analysis. We conceptualize three metaphors—unreliable narrator, court jester, and sounding board—that possess the most explanatory capabilities in describing what ChatGPT is, when it can be used, and how it can be helpful. We posit that grounding the use of ChatGPT in metaphors could facilitate discussions and streamline the intricate mechanism of Large Language Models (LLMs). Our study indicates that by proffering playful metaphors as substitutes to apocalyptic and arcane ones, we can enhance the accessibility and comprehensibility of ChatGPT for non-experts and policymakers, thereby potentially contributing to more informed and productive dialogues about the role and potential of LLMs in everyday life.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 565-569 |
| Number of pages | 5 |
| Journal | Proceedings of the Association for Information Science and Technology |
| Volume | 60 |
| Issue number | 1 |
| DOIs | |
| State | Published - Oct 2023 |
Keywords
- Autoethnography
- ChatGPT
- Large Language Models (LLMs)
- Metaphors
ASJC Scopus subject areas
- General Computer Science
- Library and Information Sciences