Could you apply this to anime next? Pretty please?

You can use the encoder with any StyleGAN model, so, totally possible to do!

Wow. Adding to my list to check out! My first silly idea is: if you use a non-human StyleGAN model, but ask it to find the latent representation of a picture of Donald Trump, would you get the piece of furniture (or whatever model) that most resembles Donald Trump?

There's a good paper about this; you can find decent approximations in the latent space, but it will be meaningless to the model, i.e. interpolation will be garbage

The "nodal point somewhere inside the bridge of the nose" aspect of current GAN animations is ubiquitous (and of course an unfortunate necessity). It will be good when we can animate these faces more naturally (pivoting from the base of the neck etc) 👍🏼

I think this is because of how StyleGAN is trained, all the faces are aligned to the same general areas, so it's not relative to the rest of the body. Meanwhile, it's only seen good examples of faces, eyes, etc., in specific areas of the image.

I wonder how this would compare to traditional image morphing (e.g. psychomorph)

It does a great job of image morphing too!