What happens when stories themselves — when the biological imperative that helps us process randomness and chaos as a means of "existential problem-solving" — are not only beyond our control, but the product of randomness and chaos? Not, of course, in the most literal sense — since algorithms are nothing but pattern, calculated variation, and predictability. They are, however, random in the sense that they have no understanding of the experiences, emotions, moralities, and conflicts they attempt to replicate, and to which they (appear to) speak. They can't tell the truth from a lie, and they can't (for now, and the foreseeable future) tell us anything we don't already know — or, more disturbingly, that we don't already think we know.
What becomes, then, of the problems humanity has yet to solve? A world taught by machines that can learn "new" things from us, but from which we can learn nothing new, is a world doomed to repeat past failures with even greater uniformity and accuracy than it already does. In that world, Freddy wins — because he has to. He wins because if he doesn't, he can't come back for another repetitive, predictable sequel. His defeats are only temporary because they're the result of temporary solutions (solutions that have already been tried, and that have failed again, and again), and his victory is greater because it isn't merely the "murder of innocence," but the murder of imagination — that is, the complete and total death of our ability to fathom what has never been fathomed before.
And if we lose that, we lose the very thing that, for all these centuries, has kept our self-destructive species alive: our ability not simply to create, but to interpret, deeply feel, and learn from the stories we tell.