Thanks for this take. It perfectly illustrates the productivity illusion.

Mostly blindly approving PRs mainly because they now contain some form of equally AI-generated tests is exactly the producticity fallacy that's the issue here.