If a CRISPR change doesn't kill you, it seems to me unlikely to kill your offspring. This may be naive, I don't know to what extent a gene variant would be fine

If a CRISPR change doesn't kill you, it seems to me unlikely to kill your offspring. This may be naive, I don't know to what extent a gene variant would be fine in an adult but not fine in a child. Most self gene edits would not be random, but would be more along the lines of "in studies x% has variant A;A and y% has variant A;G or G;G. A;A is "better" in some sense I care about, and I have A;G, so I'm going to edit that G to an A." Maybe there are dependent genes that need editing as well that we don't know about because otherwise you're no longer a viable lifeform, but it seems that any cascading effects would come up before you had a child. It's like editing a character in the string constant pool of a program. Maybe it'll be fine, or maybe you need to edit some other part whose logic depended on the original character to match the new character and avoid a crash or misbehavior. Getting my genes sequenced for $100 revealed I'm a carrier of https://en.wikipedia.org/wiki/Tay%E2%80%93Sachs_disease https://en.wikipedia.org/wiki/Tay%E2%80%93Sachs_disease It would be highly irresponsible of me to breed with another carrier due to the 25% chance of an offspring with the disease, which all but guarantees early death. If one of us parents has the gene edited away, or it is edited away in the womb, this is a good thing.