Paper still matters: where AI fits with printed work

Paper still matters, and AI works best when it fits around it. Students often have exercise books and textbooks, so AI does not need to produce more printed worksheets. Where paper matters most is in tests and in work you collect back as a set. Use AI in the steps before and after, such as drafting and checking, and keep the students’ own thinking on the page. When paper work goes back into a tool, for example by scanning, treat it with the same care as anything else you type in.

Why paper still has a place

The founder, a teacher, described how printing works in their classes: “Students often have books and textbooks so printing is not necessary until it comes to tests, and the collecting back of things as a collection is often important too.”

A maths exercise book holds a term of working, mistakes and corrections in one place. A printed poem with a student’s annotations shows how their reading developed across a lesson. A test on paper gives every student the same conditions, with no notifications, no flat batteries and no other tabs open.

This is a view from teaching practice, not a finding from a study. Some tasks are better on a screen, and a class can use both. The point is that paper should be a choice you make for each task, not something that disappears because a new tool assumes everything happens online.

Where AI fits before the paper

Most of the useful AI work for printed tasks happens before anything reaches a student. You might ask an AI tool for a few candidate questions on a skill, a simpler version of a text for students who need it, or a larger-print layout for a student with a vision adjustment.

In each case you check and own what comes out. You choose the questions, fix the errors and make sure the answers are right. For maths, the answer key should be computed exactly, not written by a language model; our article on maths tests explains why. None of this needs any student information, so leave names and results out.

Keep the thinking on the page

We hold one principle across this site: AI helps learning when the design keeps the thinking with the student. Paper is one of the simplest ways to do that.

When a Year 9 student solves a problem in their book or on a test paper, the working is theirs, and you can see it. You can see where they started, where they crossed something out and where the reasoning broke down. That is the evidence you need to teach the next lesson. A printed task does not guarantee good thinking, but it makes the student’s own thinking easier to see.

Collecting work back as a set

Collecting work as a set matters for reasons that are easy to overlook. With a full class set in front of you, you can sort scripts into piles, compare answers to the same question across the class, moderate with a colleague and hand everything back with feedback in one go. It also leaves a clear record of what each student could do on the day.

Some AI tools expect every task to be completed and submitted on screen. Before you adopt one, ask whether it fits the way you collect and return work. A good tool should support your routine rather than replace it.

When paper goes into a tool

Scanning a class set into marking or analysis software is the point where paper meets AI, and it deserves care. A scanned test carries each student’s handwriting, their answers and often their name at the top of the page.

In Victorian government schools, the department’s generative AI policy (read 25 September 2026) requires schools to direct staff and students to keep student names and other personal information, and sensitive school information such as student assessment data, out of generative AI tools. Our article on report comments quotes the wording. Other schools: check your sector’s or state’s rules.

A student number still identifies a student to the school, so it does not make a scan anonymous. In Victorian government schools, test scripts count as student assessment data, which staff must not put into generative AI tools. If the scanning or marking software uses generative AI, ask your school’s leaders before you upload anything. Other schools: check your sector’s or state’s rules, and find out where the scans are stored and for how long.

What to do on Monday

Keep using books and textbooks for everyday work. Print when it counts: for tests, and for tasks you want to collect back as a set.

Use AI to help prepare printed tasks, and check every question and answer before you print.

For each new task, decide deliberately whether it belongs on paper, on screen or both, and why.

If you are thinking of scanning student work into any tool, sort out names, approval and storage first.

Things will keep moving. As the founder puts it: “Things are changing so fast even this opinion might change.” The habit worth keeping is choosing the format for the learning, then fitting the tools around it.

To see how your school’s rules and tools fit together, try the free AI readiness check for your school.