When I needed a super simple “hello world” introduction to AI apps, I went for a simple brainstorming app. Start with a word or phrase describing a topic, and the app comes up with a number of title suggestions for a blog post. (And they’re pretty decent.)
Is this a good introduction? For purposes of going over the absolute basics of the anatomy of an AI app, sure. But is this sending the wrong message about AI?
In other words: Is brainstorming, just you plus an AI, the kind of activity we ought to be encouraging people to internalize?
It gets to the heart of why brainstorming works. It’s all about the ideas. There are (at least) three forces at play in ordinary day-to-day work that hamper ideas.
- Our own limited experience
- Our own limited focus context
- Our own internal filter
So it’s good to have others around for brainstorming. #1 because someone else might have an idea we never would have thought of. #2 because there might be ideas we could’ve thought of, but didn’t because we’re so focused on the more immediate aspects of the problem. Lastly, #3 is to a large extent environmental, thus typical brainstorming sessions cultivate an attitude of explicitly avoiding self-censorship.
Today’s LLMs are more well-read than you. 🙂 But not all their knowledge exists at the surface, especially for simple prompts. So for #1 they can be helpful, but it takes some skill to craft a prompt that stirs up the needed latent knowledge. Potential bias remains an issue, though.
For point #2 LLMs are extremely well-suited.
Point #3 is more difficult to map to AI interaction, though I have a few theories I’m working on.
The net: If you have a chance to brainstorm with a group rather than alone, go for it. But when that’s not an option, you’ll be much better off with an AI tool than without.