Damn, you are so right about it. A mix interaction model is the logical step… and double damn, as the draft post I was working on was to illustrate the same thoughts, scrapping it now ;) you said it very well with nice examples.

Did you notice, in Westworld, most of the times input to fold-pads is voice but output is on-screen…. Make sense, as I don’t want Siri to blurt-out things so that people in surroundings can hear …and imagine the chatter in room if everyone’s Smart device is talking back. ;)