New in Journal of Pragmatics: how ChatGPT handles metaphors in health talk.
1. It recognises & reuses metaphors.
2. It also creates new ones.
3. Replies sound supportive.
4. Yet they remain predictable.
Sounding human ≠ understanding human experience.
authors.elsevier.com/a/1nfkC1L-nhf1…
1. It recognises & reuses metaphors.
2. It also creates new ones.
3. Replies sound supportive.
4. Yet they remain predictable.
Sounding human ≠ understanding human experience.
authors.elsevier.com/a/1nfkC1L-nhf1…