I think more of the danger with Copilot is that managers will believe that it makes programming easy and so they will think programmers should be compensated less or they need less of them.
I think it's fairly safe to assume that organizations with managers like that will end up producing code that's defunct enough to sink the organization, given the bar of talent they'd manage to attract. A bit like expecting stellar journalism out of a newsroom that downsizes and hires high school kids to feed headlines into GPT-3 expecting fully-formed news articles out the other end.
Managers of programmers should be previous (or current) programmers generally anyway, so they should know better. Non-programmers managing programmers usually has a bunch of other issues already.
If it’s not the case (which I don’t think it is) then the reality should quickly prove them wrong so what’s to worry about?