video1.0<iframe src="https://www.loom.com/embed/eda3ae4652b84bc0a731c5fbc0ee6ca6" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/eda3ae4652b84bc0a731c5fbc0ee6ca6-5aab3bab482d98ca.gif3142.233I shared why AI prompt engineering fails in education most often as a clarity and generic output problem, not a language problem. I emphasized that ethics is not optional, covering transparency, data stewardship, and the risk of fluent but inaccurate outputs, and I cited the UNESCO 2024 AI teacher framework and a 2025 survey on teacher workload. I taught five prompting methods, including zero shot, role assignment, chain of thought, critic prompting, and constraints, with a suggested stacking order and live demos for persuasive writing. I requested that you try one method this week. I also invited you to join the AI literacy lab circle or book a consultation for the May cohort.