While Github Copilot is generally good at taking some of the pain out of writing Unit/Integration tests*, this is a fairly good example of why you need to be able to review what it's doing... I think I can probably come up with a slightly more time efficient approach.
*At least when it has a manually written success path to extrapolate from