I'm assuming that - by design - it has a feed-back-loop that allows it to tweak, learn and improve itself by feeding back the choices people make vs its own rec

I'm assuming that - by design - it has a feed-back-loop that allows it to tweak, learn and improve itself by feeding back the choices people make vs its own recommendations.