I’m increasingly irritated by critical scholars writing terrible prompts for conversational agents and sharing these as if it proves the technology is overhyped. This isn’t the devastating critique they think it is. To critique these technologies we need to learn to engage effectively with them so we are working with a realistic appraisal of their capacities. Offering a one sentence prompt using ambiguous language without defining the context or explaining your expectation just shows you don’t know how to use the technology. I’m not offering examples because I don’t want to personalise this but I’ve seen far too many instances of this in recent months.
