It has been argued for a very long time, lines of code is largely meaningless as a metric. But now that AI is writing those lines... it seems to be meaningful again? I continue to be optimistically skeptical.