A false negative would, as I’m understanding the goal here, be a case where the AI missed an existing problem.
It wouldn’t change the current state so it wouldn’t actively hurt anything though, and of course it’s plenty likely a human checker would have overlooked those misses and more.
A false negative would, as I’m understanding the goal here, be a case where the AI missed an existing problem.
It wouldn’t change the current state so it wouldn’t actively hurt anything though, and of course it’s plenty likely a human checker would have overlooked those misses and more.