.CONVENTIONAL SOFTWARE reacts predictably to instructions. "Generative" artificial-intelligence (AI) models, such as that made use of by ChatGPT, are actually different: they react to asks for recorded everyday language, and also may produce surprising results. On the face of it, creating successful triggers for AI is actually a lot simpler than, for instance, understanding a programming foreign language. Yet as AI designs have become more capable, taking advantage of the protocols within these dark cartons has ended up being harder. "Prompt engineering", as this skill-set is actually known, has been parallelled to helping a dance companion or even peeking a monster to see how it will definitely react. What does it involve?