There are many safe choices of targets, simply reverting a known mutation that causes a well-studies genetic illness is such an obvious choice. Mutating the wild-type to a non-functional version that has known benefits but also drawbacks, and almost certainly has unknown drawbacks as well is such an insane choice for the first experiment of this type.
Not that making a better choice there would have saved this experiment. The lack of informed consent, the highly experimental nature of the work and going ahead even when you know that you did not achieve exactly the mutations you intended damn this experiment in any case.