"I don't think that people normally acquire a mutation except at conception."
The hypothesis is, I believe, that a virus can insert itself into the cells that produce the gametes and transfer the mutation to offspring. So you are correct, but this is a new mechanism they are investigating in the study of the mechanisms of evolution.
You understand that a mutation doesn't necessarily cause an immediate change in phenotype. The mutation could be in the junk DNA and not activated.
Retroviruses. Nasty critters. Aids itself is a retrovirus.
The virus enters the cell, then splices it's DNA into the cellular DNA. Then the virus dies.
So time goes by, and the cell divides, then divides again, on and on. Something then happens that trips the infected cells to start producing new viruses. So you might have hundreds or thousands of tiny virus factories in your body, all chugging out new killer viruses.
Retroviruses are also able to hide from the immune system. If it gets into your cell and does the splicing, it might have been able to evade any of the t-cell or lymphocytes that normally detect foreign invaders.
So you can be infected and not show any antibodies for the disease.