Peer review has a mixed reputation among teachers because students often give vague or overly kind feedback, such as this is good, I liked it. Without structure, the session can feel like a wasted class period. When peer review is paired with AI-generated feedback and a clear protocol, however, students receive better guidance and learn to evaluate writing more critically, which improves both their drafts and their own analytical skills.

The first step is to give peer reviewers specific questions tied to the rubric. Instead of asking whether the essay is good, ask whether the thesis makes a debatable claim about Into the Woods and whether each paragraph supports it with a specific scene. These focused questions make it easier for students to spot problems, since they know exactly what to look for and can point to particular sentences in the draft.

AI feedback can serve as a starting point for the discussion rather than a replacement for it. A student might receive automated comments on their draft highlighting an unclear thesis and a missing example from the second act, then bring those comments to a peer conference. The peer can react to the feedback, suggest how to address it, and offer a reader's perspective that a tool cannot provide.

Structure the workshop in clear stages

A well-run workshop moves through distinct phases so that students know what to do at each point. Begin with a few minutes of individual reading and annotation, followed by structured feedback exchange, and end with a written revision plan. The revision plan is the most important step, because it converts conversation into action and gives you a simple way to check that each student actually used the session productively.

• Students read their partner's draft and mark the thesis and strongest piece of evidence
• Reviewers answer three rubric-based questions in writing before discussing
• Writers compare peer comments with any AI-generated feedback on their draft
• Partners discuss points of agreement and disagreement between the two sources
• Each writer submits a short revision plan listing three specific changes

Teach students to evaluate automated feedback critically

Students should understand that AI comments are suggestions, not commands. A tool might flag a paragraph as lacking analysis when the student believes the analysis is present but subtle, and that disagreement is an opportunity for discussion. Encouraging students to explain why they accept or reject a piece of feedback builds metacognition and prevents the passive acceptance of every suggestion that can weaken a writer's voice.

Set clear expectations about when automated feedback is part of the process and when it is not. For example, you might allow it during drafting and peer review while requiring that the final submission reflect the student's own thinking and revision decisions. Communicating these boundaries in advance prevents confusion and keeps the focus on learning rather than on policing how each comment was generated.

Measure whether the workshop improved the drafts

A simple way to see whether the session worked is to compare a student's first draft with the revised version using the same rubric. Look for movement in specific criteria, such as a sharper thesis or better explained evidence. This comparison gives you concrete data on which parts of the workshop were most effective and helps you adjust the protocol the next time you run it.

It also gives students a tangible record of their growth. Asking them to write a brief reflection on what changed between drafts and why helps them internalize the lessons. Over several assignments, students begin to anticipate the feedback they would receive and address weaknesses before anyone points them out, which is the real goal of any revision process.