The future of prompting is multi-modal

It's time to move beyond chat and text input

I've been asking our Designing with AI Discussion series guests what they think the future interface of design looks like, because I'm convinced our current approach isn’t going to scale. I hope the future interface of design looks like a seamless blend of human and machine, and that we'll be able to interact with digital systems in ways that are more natural and intuitive.

That's not to say that prompting isn't already an incredibly effective tool. In fact, with Midjourney, I can confidently generate almost any type of image in any medium, and I have a reliable method for doing so. Even if I don't land on the first prompt I come up with, I have a method for creating complex prompts that will get me really close to what I'm looking for. The suite of tools will help me refine my ideas until I get to the exact image I want.

But, even at this stage, prompting still feels incredibly primitive. When I’m deep in the creative process within any AI tool, I want more ways to communicate my ideas, just like I would with my human creatives.