My point is that somewhere there must be measurable differences in genetic information.
There are lots, principally because there are lots of different metrics by which 'genetic information' can be measured. The point is that to discuss a specific claim like 'mutations can only decrease genetic information', we have to be using the same metric or the discussion is meaningless.
We could measure the size of the genome, the number of genes, the proportion of coding to non-coding DNA or any number of things that might be colloquially considered measures of genetic information and that is before we get onto the information theoretic measures like Shannon information, Shannon entropy, Kolmogorov complexity, etc...
As far as I know though absolutely none of these metrics would be compatible with the idea that point mutations must lead to a decrease in information.
I don't know why mutations couldn't accomplish that, because corrupted alleles would be nulled, as you dsay.
I'm not sure what these things have to do with each other. Can mutations effect changes in the amount of genetic information? Certainly in fact they are the obvious mechanism which does, but that change can be of varying magnitudes and in either direction or even laterally. But this has nothing to do with the fact that some mutations can create null alleles, that may be one example of a change in genetic information but without an agreed metric and knowing exactly what the mutation was we can't say if it is or what sort of change.
TTFN,
WK