I would guess build a health prediction model. Instead of next text token or next frame in a video, how about next 12 months or years of body health? Hopefully

I would guess build a health prediction model. Instead of next text token or next frame in a video, how about next 12 months or years of body health? Hopefully it doesn’t become Gattaca.