Draft feedback is often the most valuable form of marking, yet it is the first thing to be cut when time runs short. Students who receive comments only on a final essay cannot act on them, while those who get feedback at the draft stage can revise with purpose. AI tools make it realistic to offer this stage to a whole class. For a text like Two Brothers, where arguments develop over several attempts, the benefit is considerable.

The most effective approach is to give the tool the assignment prompt, the rubric and any specific focus for that draft. A first draft might be assessed only on contention and structure, while a later one targets evidence and expression. Narrowing the focus prevents students from receiving an overwhelming list of problems. It also keeps feedback aligned with what you have taught.

Draft feedback should be framed as guidance rather than a verdict. Comments that point out what is working and name a next step encourage revision. Scores can be withheld at this stage so that students concentrate on improving rather than on a number. Teachers can decide when to reveal grades.

Keeping the Teacher in the Loop

AI feedback works best when a teacher reviews it before students see it. A quick read lets you correct comments that misread a student's meaning or miss the subtleties of a scene in the play. It also gives you a snapshot of where the whole class stands. The teacher's role shifts from writing every comment to guiding and refining.

• Provide the exact prompt and rubric with every batch
• Limit each draft round to one or two feedback priorities
• Review generated comments before releasing them to students
• Ask students to respond to feedback with a short revision note
• Compare drafts to confirm that comments led to real changes

Building Revision Into the Schedule

Plan class time for revision rather than assuming it will happen at home. A lesson in which students read their feedback, choose one comment and rewrite a paragraph is far more effective than simply handing back drafts. Circulating during this time lets you answer questions and catch misunderstandings. The routine builds the habit of treating feedback as the start of work.

Consider asking students to submit a short note explaining what they changed and why. This encourages reflection and makes it easy to see whether feedback was understood. Notes also help you spot comments that were unclear. Over time, you can refine how you phrase suggestions.

Avoiding Over-Reliance

There is a risk that students treat AI comments as instructions to follow mechanically rather than as input to weigh. Remind them that feedback is advice and that the final choices are theirs. Encourage them to ask why a suggestion might help. This builds the judgment that good writers need.

Be transparent about how feedback is generated and what its limits are. Students who understand that a tool is applying the rubric, not reading with human insight, can use it more wisely. Pair AI comments with teacher conferences for complex issues. The mix keeps the feedback human where it matters most.

Evaluating Whether It Works

Track whether students who receive draft feedback make measurable gains between drafts. Look at contention clarity, explanation of evidence and overall rubric scores. If improvement is small, the feedback may be too generic or the revision time too short. Adjust and test again in the next unit.

Gather student opinions as well. They can tell you whether comments were clear, whether they felt overwhelmed and what they would change. Their insight often reveals easy improvements. Treat the process as ongoing refinement.