Random mutations much more easily debilitate genes than improve them, and that this is true even of the helpful mutations. Let me emphasize, our experience with malaria’s effects on humans (arguably our most highly studied genetic system) shows that most helpful mutations degrade genes. What’s more, as a group the mutations are incoherent, meaning that they are not adding up to some new system. They are just small changes - mostly degradative - in pre-existing, unrelated genes. The take-home lesson is that this is certainly not the kind of process we would expect to build the astonishingly elegant machinery of the cell. If random mutation plus selective pressure substantially trashes the human genome, why should we think that it would be a constructive force in the long term? There is no reason to think so.