That runs counter to some very well established UI principles. People get confused when their interface changes except as a result of direct interaction. Open up a menu in response to a click, yes; reorganize menus to "optimize" them based on what a model predicts a person is going to do, no.

The killer is being able to tell a program what you want it to do, then not having to fuddle with buttons or menus at all (unless you want to tweak things).

While I agree in principle, AI-infusion in UIs does not need to break convention. For example, AI suggestions could subtly highlight menu options that would be most useful in an auto-detected workflow. We could also create a persistent area on the UI for the AI to "speak up".