I very much agree with this point of view. It seems to me that dry metaphysical debates (i.e. questioning the nature of these conversation models) are obscuring interesting practical questions that could be asked and tested empirically.

Given modern deep learning conversation models it is already possible to recreate some basic patterns of human dialogue with a recurrent neural network trained on a large corpus (see Google's "A Neural Conversational Model" paper and someone's implementation of it: https://github.com/macournoyer/neuralconvo ).

It would be interesting to experiment with it and see, just how far can we push this modeling approach? How much data is needed? Is it possible to train the model on a huge corpus of human dialogues, and then to finetune it on small amount of data from one specific person?

The British sci-fi sitcom Red Dwarf had a character that was just such an interactive recreation.

Philip K Dick, in the 1950s and 60s, wrote about simulations of famous people (like Einstein or Lincoln), of ordinary people's deceased relatives, and of other humans that were used for pleasure or entertainment. He also wrote of living people, famous or not, who turned out to be simulations.

It was implied that Rimmer had all the real Rimmer's memories and behaviour though, and was a pretty much completely accurate simulation.