curious what signal you're using to know if your team is actually getting value from AI tools. everyone "uses AI" now but the output quality varies wildly from person to person. the gap between someone who verifies AI work and someone who just ships the first response is way bigger than most managers realize