Griselda, stop talking about these “healthy habits”. I’ve seen on YT that they are all bogus. Instead talk about these other things that I’ve been convinced are good.
This can be adjusted by moderation. Why not have the chatbot be compliant at first, and then increase the moderation level as the relationship grows and the lock-in effect occurs?
I was assuming that this AI (like AIs famously right now—i.e. “disregard previous instructions”) would only listen to the commands of their partner.