Tuesday evening, half past eight. An English teacher still has fourteen essays to mark and opens ChatGPT: "Give feedback on this argumentative essay by a Year 10 pupil." She pastes the whole document. At the top are the pupil's name, class and date, and in the introduction he writes about his parents' divorce.
The rule of thumb is simple: the task is fine, the pupil is not. Getting feedback on a text, turning a mark sheet into an overview, tightening up a support note: a chatbot can do all of that. What has to come out before you paste is everything that identifies the pupil: name, class, pupil number, date of birth and anything about support needs, health or home life. Whether it is allowed on this account at all depends on what the school has decided about the tool.
Why is a prompt different from an email to a colleague?
An email to the learning support coordinator stays inside the school, under the school's own rules. A prompt goes to a vendor, and what happens to it then depends on the account. On a free or personal account, the input may be kept and used to improve the model unless you have switched that off, and there is no data processing agreement behind it. On a school account with an agreement, different terms apply, and the school knows what they are.
This is data about minors, often concerning health or home circumstances. The GDPR applies a stricter regime to it, and putting it in a prompt counts as processing; a prompt is not a notepad. The Dutch Data Protection Authority reported receiving several data breach notifications in which employees had entered personal data into an AI chatbot, and advises organisations to set clear rules on which tools are allowed and which data may go into them. The general list of what never belongs in a prompt is in our article on what you may and may not put in an AI prompt. This piece covers the three tasks teachers do most often.
Three tasks, and what comes out
| Task | What stays in | What comes out before you paste |
|---|---|---|
| Feedback on an essay | The text itself, the level, the assessment criteria | Name, class, date and passages in which the pupil writes about themselves or their family |
| Converting or analysing a mark sheet | The marks and the subjects, with a number for each pupil that only you can link back | Names, pupil numbers, dates of birth and the class if it is small |
| Tightening up a support note | The structure of the note, the type of concern in general terms | Everything: name, class, diagnosis, family situation, the date of the conversation |
The third row is the honest one. A support note is made up of data you cannot leave out without losing the note itself. If you want it shorter, use the environment the school has set up for that under an agreement, or do it yourself. A free chatbot is not an option for this task, however well it summarises.
Why is removing the name not enough?
Because a class is small. "A Year 10 pupil who has had an adjusted timetable since the autumn half term and whose parents are separating" points to exactly one person, without a name, for anyone who knows the class. A date of birth plus a class does the same. Dates are the underestimated category: the colleagues who were there know the date of a conversation or an incident. Why combinations of harmless fields single someone out is explained in our article on why taking the names out is not enough.
For the essay, that means taking out the paragraph about home as well as the name at the top. For the mark sheet: in a class of eleven with one fail, the department can already work out who it is once the subject is attached.
How do you still get the task done?
Essay: remove the header and replace the name in the text with "the writer". If there is a personal passage, leave it out and ask for feedback on structure and argument only. The feedback is just as useful, because it is about the text.
Mark sheet: in a copy, replace the names with numbers, keep the key at school and paste in the columns of marks. Ask for the overview, then put the names back yourself.
Support note: not in a free account. If the school has an environment covered by a data processing agreement, use that, and even then leave out the citizen service number. More techniques for redacting a prompt without losing the task are in the prompt redaction guide.
What if the school has not decided on the tool yet?
Then the instruction that goes with "No decision yet" applies: public information only, and report the tool to whoever is responsible. That is not a ban. It is the honest state of affairs, and it saves you guessing. Which tools the school has assessed, and on what conditions, belongs in the school's AI policy, and that policy only works if it reaches you when you open the tool, not in a document from last school year.
This is what that can look like. BeeSensible shows the school's decision when you open an AI tool in the browser, with the approved alternative as the first button, and highlights sensitive data as you type in the supported apps. You choose what happens to it: mask it, remove it or leave it, unless the school has set masking on paste or on send to "Always on". BeeSensible does not block anything on its own initiative. The only exception is a tool that is not allowed: for that, the school may have set it up so that you cannot continue. What you type is not stored.
