
Most school leaders now have a general AI assistant within reach. Staff draft letters in ChatGPT, summarise meeting notes in Copilot and plan lessons with Gemini. So when a leader has a question about their own school, such as why Year/Grade 9 attendance has dipped this term, the obvious move is to open the same tool and ask.
That can work. Give ChatGPT or Copilot a clean export and a clear question, and it will find patterns, draw a chart and suggest explanations. These tools are capable and improving fast, and for plenty of jobs they’re the right choice.
What matters is how much work sits around the answer, and who does it. That’s where EduIntelligence, an AI decision intelligence tool built for schools, takes a different approach.
Watch: EduIntelligence in under two minutes
Consider what has to happen before ChatGPT can answer a question about attendance.
Someone exports attendance from the MIS or SIS. If behaviour might be part of the story, they export that too, then assessment data, and perhaps the timetable. Each file has its own column names and its own way of identifying a student. Someone decides what counts as persistent absence this term. And if school policy allows student data near a public tool at all, names and identifying details have to come out first.
Then comes the prompt. The assistant doesn’t know the school’s curriculum, which year groups matter most this term, or that the bus timetable changed in September. Every new conversation starts from nothing, so the context gets typed in again.
The answer can still be good. But it depends on what was exported, how it was cleaned and how the question was phrased. Ask again next month with a slightly different export, and the definition of persistent absence may quietly change with it.
EI begins where the general assistant leaves off. It connects to the systems your school already uses, so the starting point is the question rather than the export.
Ask “What’s driving the change in Year/Grade 9 attendance this term?” and EI brings the relevant information into the analysis, looks for patterns across it and comes back with an evidence-backed answer: what changed, the possible drivers, and next steps worth considering. Answers come back in minutes, because EI is doing the data work a person would otherwise do by hand.
Three things separate that from pasting files into a chat window.
Context it keeps. EI holds your school’s context, including its curriculum, location and priorities, and the role of the person asking. A head of year and an admissions director asking about the same cohort get answers framed for their jobs. Policies you add become a primary source for recommendations, so suggestions fit how your school actually works.
The same definitions every time. EI works from a governed data layer, so a measure such as retention or attendance is defined once and calculated the same way whoever asks and whenever they ask. The wording of an answer can vary. The numbers underneath it don’t.
Answers built for the next decision. EI anticipates the follow-up questions a leader would ask and answers them in the first response, then suggests where to look next. The point is to get from “what happened” to “what do we do about it” while the question is still in front of you.
The most common worry about general AI in schools is staff pasting student information into a public tool. Many school AI policies rule it out, which leaves leaders with a capable assistant they can’t use for their most important questions.
EI handles this before the AI model is involved. When a request touches student data, names and identifying details are stripped out and replaced with an anonymous identifier. The model has no route from that identifier back to the student. Names are put back only when the answer is shown to you. And no customer data is used to train third-party AI models.
General AI is the better choice for a lot of school work: drafting and rewriting, summarising a document you already have, brainstorming, or a one-off look at a small dataset with nothing identifiable in it. If a question needs one spreadsheet and you’re happy to set the context yourself, a general assistant will handle it well.
EI earns its place when questions recur, cross more than one system, involve student-level information or need to give the same answer to different people. That describes most of the questions that end up in leadership meetings, board reports and inspection preparation.
Many schools will use both. A useful test is to take a real question your leadership team asked last term and check three things: how long it took to answer, how many systems it touched, and whether you’d have trusted the figure if someone else had produced it. If the honest answers are days, several and not entirely, the slow part is the work around the AI, and that’s the work EI takes on.
General AI gives you answers. EI understands your school.
See EduIntelligence Answer Your School’s Questions
Book a demo and we’ll show you how EduIntelligence handles the questions your leadership team asks every term.