Beyond Policing: AI Writing Detection Tools, Trust, Academic Integrity, and Their Implications for College Writing

Current AI detectors demonstrate significant technical limitations, including high false-positive and false-negative rates, and show bias against non-native English speakers. We argue that AI detection-focused approaches in writing foster an atmosphere of distrust and anxiety, while collaborative AI integration enhances learning outcomes and student engagement. Successful AI integration in writing requires clear institutional guidelines on AI and AI detection, training for educators and students, and a focus on process over product in writing assessment. By shifting from a policing mindset to a trust-based educational framework, this method aims to promote academic integrity, engagement, and responsible AI use in college writing.