Two questions I get asked a lot by educational practitioners about AI are firstly which AI is the best and secondly how do I write prompts. The first I will come back to in a future article (hint – there is no simple answer).
Going back a year ago prompting was a bit of a mystery and the only real way to find out was by trial and error. It wasn’t helped by the Generative AI providers, who themselves do not fully understand how the AIs work. But of course there has been a lot of exchange of experience for using AI in education. AI has seen a boom in blogging, albeit in the form of newsletters. There are a lot – for example Ethan Mollick in his weekly One Useful Thing newsletter often reports on how he is developing more effective prompts.
Now Open AI – the people who developed Chat GPT – have published a Prompt Engineering Guide (see diagram above. They say:
This guide shares strategies and tactics for getting better results from large language models (sometimes referred to as GPT models) like GPT-4. The methods described here can sometimes be deployed in combination for greater effect. We encourage experimentation to find the methods that work best for you.
I’ve been trying it out and it seems pretty good.