The app model can't do speculative assistance.

Speculative assistance is necessary to do anything exploratory, where you don't know what the answer of the service will be before you do it.

But in the same-origin paradigm, once you reach out to the 3P service, that service could do whatever they want, on a technical level, with that data.

That data could leak out in unexpected ways.

For example, "What would a family trip to Tokyo in the summer look like?" could turn into getting unexpected emails from Delta encouraging you to try their co-marketing deal at sushi restaurants.

This means the app model is an awkward fit for an AI-native speculative assistance.