Your example from AIG is an example dishonesty and not what you think it is.
If you wish to discuss information they it would be useful for you to define it first so that it becomes possible to determine the amount before and after then it is possible to calculate the increase or decrease.
The source for your assertion about "almost all observed mutations" needs to be supplied. It is almost certainly wrong. I would expect from just thinking about it that almost all observed mutations are about "information neutral" making a guess at what you mean by information.