There are 23 million, so that means 2.3 million errors. That would be 2.3 million erroneous cards. That is not, I would submit, something that says this can be regarded as an accurate and reliable piece of information.
Finally, and you don't provide reports on this, I would suggest that the inaccuracy is not spread evenly across the country. It is worse in ridings where there is high movement of people. People change their residences. I can't tell you where this is. I can guess it's in urban areas and that, therefore, there are some ridings, like my own, in which I suspect it is far below 10%, and others in which I suspect it is upwards of 20%, perhaps 25%.
I submit that makes this highly unreliable and facilitates the possibility of widespread voter fraud as one alternative. I'll leave it there. It just seems to me that if this is used as a form of identification, we've opened up the avenue in certain urban ridings for there to be potentially widespread voter fraud and there would be no capacity for you to police it.