Proposed somatic cell gene-therapies (especially those involving in utero therapies) may involve a small risk of germ-line modifications; this risk has engendered serious concern, and arguments have been made that such therapies ought not be pursued if such risks exists. We argue here that while pursuing deliberate germ-line modifications in humans would be inappropriate given the current state of the art, the risk of accidental germ-line modifications from most currently proposed in utero gene therapy is no different in kind or degree from other risks regularly taken in medical procedures. Given the possible benefits of such therapies, we argue that the risk of accidental germ-line modifications is well worth taking in these cases.