Introduction
Like me, you’ve probably been dazzled by the strides recently made in the field of Generative AI. It feels like we are living in a science-fiction reality, where machines understand us.
At the heart of this revolution, OpenAI’s GPT-4 has emerged as a marvel in the realm of language models. As someone who had the privilege of playing around with it, I can testify to its astonishing capabilities.
In this article I want to share a technique that intrigued me quite a lot. In literature it is referenced as the Flipped Interaction Pattern, and I found it to be a very powerful framework to help in problem solving tasks. The distinct aspect of this prompt engineering method lies in its “reversed” (i.e. flipped) approach: instead of directly asking for answers or solutions from the AI, it primarily focuses making the AI deliver the right questions that could effectively drive us towards our desired solution.