The most useful feedback on live AI use comes from a trusted colleague who has watched you do it — not from general advice about AI, but from specific observations about your specific facilitation. Ask them to watch for three things, and debrief for 15 minutes after.
Why Peer Observation Works Better Than Self-Assessment
When you’re facilitating, you can’t simultaneously evaluate your own performance. You’re managing the session, processing AI outputs, watching student reactions, and tracking time. A colleague watching from the outside catches things you simply cannot see from inside the role — the pause that goes on too long, the moment you looked uncertain, the output you moved past too quickly for students to absorb.
Self-assessment of AI facilitation is also skewed by anxiety. If you’re nervous, you tend to remember every mistake and underweight everything that went well. A colleague gives you a more calibrated view: “That moment you thought was awkward? Students didn’t notice. The part where you read the output aloud was really clear.”
How to Structure the Observation
Brief your observer before the session with three specific things to watch: your pacing around AI moments (are you rushing through outputs? waiting too long?), how you handle unexpected outputs (do you acknowledge them? correct them?), and whether your narration is clear (can a student understand what you’re doing and why?). Give them a simple form with those three prompts and space for notes. Ask for written feedback — spoken feedback after a session is often vaguer than you need.
After the session, debrief for 15 minutes. Ask for one thing that worked, one thing to adjust, and one specific suggestion. Keep it focused. The goal isn’t a comprehensive critique — it’s one or two concrete improvements you can make before the next session.
What This Means for Educators
Peer observation is a professional development tool used in formal education contexts because it works. Most educators who use it for AI facilitation report that their confidence improves faster than through solo practice alone. The combination of rehearsal (preparation) and observation (feedback) is the fastest improvement loop available to you.
The Simple Rule
Ask one colleague to observe one session in the next four weeks. Give them three things to watch for. Debrief for 15 minutes. That single cycle will improve your AI facilitation more than months of solo practice.