Teaching AI Ethics in the Classroom

AI ethics becomes more concrete when students examine a specific use and its effects. Useful questions include: What purpose does the system serve? What data does it collect or infer? Who benefits? Who could be excluded or harmed? Who makes the final decision, and how can someone contest an error? The same tool can have different risks depending on whether it suggests a low-stakes practice question or influences access to an important opportunity. Start with a scenario students can understand, such as an AI writing assistant, a school help chatbot or an automated attendance tool. Ask learners to map stakeholders: students, families, teachers, vendors and people indirectly represented in data. Identify likely benefits and harms, including privacy, accessibility, bias, misinformation, intellectual property and over-reliance. Ask what evidence is missing and what additional information would be necessary before using the system. Move from critique to safeguards. Students might propose collecting less data, limiting a tool to a specific task, disclosing AI use, checking outputs, providing human review or keeping a non-AI alternative. UNESCO’s guidance for education emphasizes human-centered, age-appropriate and privacy-conscious use. The NIST AI Risk Management Framework offers a voluntary way to think about context, measurement and management; classroom use should adapt these ideas, not imply that a simple exercise certifies a system as ethical. Use discussion norms that welcome disagreement and affected perspectives. Avoid asking students to disclose sensitive personal experiences to make a point. Assess how well they support a claim, consider tradeoffs and propose a workable safeguard. End with a revised decision: use, limit, redesign, delay or reject the system, and explain why. Ethical judgment is a process that can change when new evidence or affected voices become visible.