On Messages and AI…

I did a webinar the other day with my friend Jill about the usage of AI. (Recording available sometime soon, I hope.)

A point that occurred to me –

AI can wraps a message in context. AI can also unwrap that context and reveal the original message.

So, what are we really communicating?

I saw a cartoon where someone was telling a colleague, “AI can take four bullet points and turn it into a complete email!” The next pane had someone saying, “AI can take this complete email and distill it down to four bullet points!”

Clearly, the implied idea is: why not just send the four bullet points?

Will that be the future mark of clarity? When AI can’t enhance your message anymore? If AI can’t make your prompt any better, does that mean it’s your core message? Can we use AI as a testing tool to make sure we’re being as clear as possible?

I’m reminded of Antoine de Saint-Exupéry, the French writer who famously said, “Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away.”

When should the prompt just be the message?

Sometimes, you visit a website and the CSS doesn’t load. And sometimes, let’s face it, that’s a better experience. A website is a message wrapped in design context (no doubt, someday AI will do this design for us). But sometimes we just want the message, not the the rest of it.

I’m looking for an AI that will read a prompt and say, “You know what? This is pretty good. Just send this.”

(I’m second-guessing my usage of the word “context” here, but I can’t think of a better one. Let me know if you come up with something.)