Because providing a template requires someone with knowledge of that particular problem space (in this case "writing letters to government agencies to inquire about disability benefits") to put in the work of designing the template, coming up with likely placeholders and predicting how people might need to use it.

GPT-3 can do a good job of this all today, without any additional training.

An expert could help out by providing a few example prompts, but the result would be a lot more powerful and require a lot less development work to put together than trying to build some kind of placeholder wizard from scratch.

A madlibs is more predictable, requires only one example, easy to understand for people with language issues or disabilities, more regulation-safe as it doesn't spit out unpredictable things.

And the madlibs is free to use, for everyone.

All things considered, GPT-3 is a more of a shibboleth of AI revolution than an actual one. Much better to give models like GPT-NeoX to talented development teams who can really optimize and fine tune it, attach great sampling strategies to it, and create great products from there. The API is just too limited for me to want to build a company atop it.

Right, but someone still has to build it.

If no-one has built the madlib template for "a letter to the council asking about my disability benefits" then madlibs being more predictable isn't going to help you.

Because you can tell GPT3 what the concept is — ie, vaguely what you are going for and it will produce. Waaay more flexible than a template.

Still requires you to have an understanding of the concept, an understanding of how to prompt GPT-3, and how to verify the accuracy/veracity of the claims the AI model makes.