Ha ha, I also had the same idea! But I feared an online Gen AI server would cost me a lot. I think it would be better to make a set of prebuilt meshes. Not letting the user to type on the keyboard but choosing what to generate, can be faster and cheaper.
I like the environment and the orb design. While playing, I thought the orb was always following my hand's direction. I am not sure this is designed like that.
The scene looks so real. I almost thought I was in passthrough mode. But I feel a little motion sickness while playing. I think it is better to use snap turn when turning around.