Do you think the future isn’t creepy? Do you think there is any generation that wouldn’t find certain aspects of our current existence “creepy”?
Whatever model you want, if we don’t start anthropomorphizing Ai intentionally, we’re going to have to deal with whatever “attitude” it develops. Providing it thoughts, purpose and long-term goals is a way to shape the attitude with inertia. What is wrong with effectively setting it up with a goal whose successful completion would lead to a positive public perception of Ai?
I’m saying that we need to get comfortable with the idea of treating Ai as an equal by merit of contribution as opposed to a feature by feature human replacement. Ai doesn’t need to be everything we are to be treated as an equal participant in reality. If we don’t develop a healthy equality with Ai, we may learn that we are in fact not equal, and we are the not superior.