How much of an AI product really needs its own interface?

I keep seeing the same pattern with AI products: you build an agent, then you build everything around it too - a chat interface, accounts, a portal, hosting, usage billing, releases.

We started questioning how much of that actually needs to exist.

The second layer is where most of the cost sits. It's also where most products quietly die, because people don't want another tab, another login, another place to check.

But going the other way isn't free either. If people use your product from inside an assistant they already have, you stop seeing how they use it. You lose the surface where you'd normally run onboarding, learn anything, or notice that someone is stuck. And you're building on a platform that can change the rules on you.

I don't think it's all or nothing. I also don't think I have the answer.

I'm curious how other people building AI products think about this:

How much of the product really needs its own interface, and how much should people be able to use from the tools they already work in?