Sure, no algorithm is able to extract any more information from a single photo. But how about combining detail caught in multiple frames of video? Some phones already do this kind of thing, getting multiple samples for highly zoomed photos thanks to camera shake.
Still, the problem remains that the results from a cherry-picked algorithm or outright hand-crafted pics may be presented.