Prompt engineering is communication. The skill of crafting prompts that work with AI mirrors how you'd speak to different people, and those skills transfer across systems. But mastery with one AI doesn't mean you'll master another. Organizations should recognize that people skilled at reading audiences and adapting their message will get the most from AI tools.
Prompt engineering is a real skill. And I think the best way to think about it is that being able to speak to a given audience is a real skill. Being able to communicate with a certain person or certain class of person can be a real skill. And similarly, being able to use these prompts well to cajole the good outputs out of them is going to be a real skill. I think it's just going to continue to be because natural language processes make machines act more like humans in terms of the way they work with language. And the skills that make you a good communicator with a certain type of person are also going to make you a good communicator with a certain type of AI. That also means that being good at one AI doesn't necessarily mean you're going to be good at all AIs, but you can be thoughtful and more helpful across more AIs. Thinking about it as a matter of communication tells you what kind of skill sets you should bring to the party. It tells you maybe a little bit unexpectedly which people in your organization are going to add the greatest value in this new paradigm.
