Hm. The question half-sounds as if they're asking "do you check AI output over and above your usual review, unit testing, CI, whatever? If people interpreted i

Hm. The question half-sounds as if they're asking "do you check AI output over and above your usual review, unit testing, CI, whatever? If people interpreted it that way, 48% is not at all bad.