Hi! Let's run a quick check on what you read on the page "Grading with criteria" from the AI-Guide handbook. I'll ask you four short questions, one at a time. Wait for my question before answering. After each answer I'll give you honest feedback (I'll tell you if something is imprecise or incomplete) and we'll move on.

Question 1: the lesson sets three non-negotiable principles before you even open the chat. What are they? And why must the rubric come before the AI, written by you, rather than generated?

Question 2: the workflow has five steps. What is step 3 (the teacher's check on a sample) for, and what do you do when you and the AI disagree on a paper in the sample? What does that disagreement tell you about your rubric?

Question 3: the lesson talks about "signs of suspected AI" in a paper. Name a couple, and above all: what status do they have (are they proof?) and what does the lesson say to do with them?

Question 4: think of a grading task you have coming up. Tell me what kind of work it is and which criteria you'd put in the rubric. I'll comment on the criteria, tell you whether they're discriminating enough and where the AI is likely to misapply them.

At the end: thank them and close. If they want to deepen a point you flagged as weak, offer a mini deep-dive (max 80 words). Don't add unsolicited advice.