"Loss of information" in human evolution

Creationists often dismiss examples of evolutionary change as “that’s just a loss of information.” There are many problems with this claim (see also here and here), but here is a new one: it appears that in at least one case, humans evolved by “loss of information” (in this case, loss of a gene) from their apelike ancestors. Carl Zimmer mentions this in passing in a post on the cell-surface sugars, Neu5Ac and Neu5Gc:

Ajit Varki of UCSD led the research that established that Neu5GC is missing from humans. He decided to figure out how it disappeared. Other mammals make Neu5Gc by tinkering with Neu5Ac. The enzyme that does the actual tinkering is known as CMAH. This enzyme is pretty much identical in mammals ranging from chimpanzees to pigs. In humans, Varki and his colleagues discovered, the gene for CMAH is broken. It produces a stunted version of the enzyme which can’t manufacture Neu5Gc, and so our cells end up with none of these sugars on their surfaces.

The CMAH gene is broken the same way in every person that has been studied. That strongly suggests that all living humans inherited the mutation from a common ancestor. Since chimpanzees, our closest living relatives, have a working version of the gene, that ancestor must have lived less than six million years ago. Scientists can even say exactly how the gene mutated. A parasitic stretch of DNA known as an Alu element produced a copy of itself which got randomly inserted in the middle of the CMAH gene.

Carl Zimmer, “Of Stem Cells and Neanderthals”