What makes you think we haven't? "Previous work using CRISPR in mouse embryos and other kinds of human cell had already demonstrated that editing chromosomes c

What makes you think we haven't? "Previous work using CRISPR in mouse embryos and other kinds of human cell had already demonstrated that editing chromosomes can cause large, unwanted effects. But it was important to demonstrate the work in human embryos as well, says Urnov, because different cell types might respond to genome editing differently." It says of one study "Of 18 genome-edited embryos, about 22% contained unwanted changes". Such a failure rate probably isn't such a big deal in some applications. In embryos, especially human embryos, it's definitely a big deal.