Protein folding as a tiny example is something we have trouble simulating, expanding to multi cellular mechanisms across a lifetime is basically impossible without directly testing it. And those tests will be done on people. Prions are just an example of the kind of issues people will be dealing with. It’s really not just about inserting some code from somewhere else it’s a question of how physical structures interact and degrade.
Long term I have no problem with it. But, right now we could give children blue hair for example but we have no idea what else that might cause.
PS: That said, copying things from other humans is probably significantly safer and this editing will take place. I just don't think we should introduce it at the society level without at least a lifetime of test data and significant sample sizes.
There will always be the risk of the unknown, there is still the risk of the unknown with seemingly every day things like cars. The utility of cars and gene therapies is so large that we really can't afford for all the data to come in.