I wonder if we'll get to the point where we train different personalities into an image model that we can bring out in the prompt and these personalities have distinct art/picture styles they produce.