I recently wrote here on Medium about the AI rules at our faculty. A successful piece of paper, but honestly: it's also kind of pointless. Hardly anyone gets caught anymore. And the tools for finding deceptions are no good. We are now making do by asking specific questions about homework.