Find the AI value in your school, and govern it while you do
AI Value Creation for Schools, free for independent schools.
Principals tell us the AI conversation at their school has been about threat for two years, and that the board has started to ask the other question: where is it helping? Heads of digital learning tell us they have a dozen ideas and no governed way to try one. Business managers tell us they would back a trial tomorrow if someone could say what success looks like and what data it needs.
Finding AI value and governing it are one motion, not two. The four parts below are how a school does both: an opportunity register, a use-case canvas that makes every idea state its purpose, success criteria, data, controls and owner before it starts, a prioritisation matrix that scores value against effort and risk, and a one-page board proposal for the idea that wins. The AI Maturity Assessment, taken with the value lens, tells you where to start; the Strategic Plan download gives the idea that wins an objective to sit under.
What the kit covers
Value and governance are the same page here. An AI policy is a lens over duties a school already carries, and an AI opportunity is only real when the duties it touches are named alongside the benefit. PolicyAI has tagged the obligations a school already carries that AI activity now touches; the canvas asks which ones each use case touches, so the idea arrives at the board with its controls already attached.
The AI opportunity register
A register of AI opportunities across teaching, learning support, wellbeing, administration, finance, facilities and communications, seeded with twelve use cases written for an independent school. Each row records the problem, the beneficiary, the value type, the data it would need, the duties it touches, the value, effort and risk scores, the owner role and the status. Sorting and a summary tab come with it.
The use-case canvas
A canvas for one opportunity: purpose, the people it serves, what success looks like and how it is measured, the data it needs and the data class, the duties it touches and the controls that follow, the tool and its vendor assessment status, the owner and the RASCI, the trial plan, and the decision to proceed, park or stop. Written in the purpose and success-criteria language educators already use for a programme.
The prioritisation matrix
A matrix that scores each opportunity on value (workload, learning, wellbeing, financial), effort (cost, time, change) and risk (data class, duties touched, residual rating from the AI risk register), and plots them on a two-by-two so the executive can pick two to trial this term and be able to say why.
The one-page board proposal
A one-page proposal for taking one opportunity to the board: the problem, the proposal, what success looks like, the data and the controls, the cost, the owner, the trial period and the decision requested. A worked example is filled in.
How schools use it
Principals tell us they run the register through the executive as a one-hour workshop: add the school's own ideas to the twelve, score them in the matrix, and pick two. Heads of digital learning tell us the canvas is the part that changes the conversation, because an idea with success criteria, a data class and an owner is a proposal rather than an enthusiasm. Chairs tell us the one-page proposal is the first AI paper they have received that asked for a decision rather than reassurance.
In EthosOne the opportunities become typed initiatives in strategic planning, with owner, problem, outcomes, activities with RASCI, KPI or OKR measures and linked risks, so the board sees AI value on the same Gantt as the rest of the strategy. Trials that grow into projects run as schedules with groups, tasks, milestones and a baseline, and every approval is a resolution with exact text. The embedded helper drafts purpose, success criteria and work breakdown from your context, and the school applies what it agrees with. EthosOne's own AI is Frontier AI, hosted via Amazon Bedrock in AWS Sydney; the model is stateless and no school data trains any model.
- Stored in AustraliaAWS Sydney and Melbourne, one database per school
- AI stays in the country tooAmazon Bedrock; never trained on your content
- Published, not describedTrust Centre · ISMS policy · How we handle your data
The two we are asked every time.
Still weighing something up? Thirty minutes with us, on your own registers, answers the rest.
Take the AI maturity assessment01Which AI use cases are in the register?
Twelve, written for an independent school: lesson and resource differentiation, feedback drafting for teacher review, individual learning plan drafting, reading support, timetable and cover optimisation, parent communication drafting, enrolment enquiry triage, policy and procedure drafting with review, board pack narrative drafting, facilities and maintenance triage, finance reconciliation checks, and a staff knowledge assistant over approved documents. Each is a starting point to edit or delete.
02How does this connect to the AI risk register?
The canvas asks which duties each use case touches and carries the residual rating of the related risk from the AI Risk Register into the matrix. An opportunity with a High residual risk is not blocked; it arrives with its treatment and its owner, which is what the board needs to say yes.
Finding the AI value starts with aread on where the school stands.
Where AI creates value in a school and how to govern it while you do: an opportunity register, a use-case canvas and a board proposal, with the AI maturity assessment (value lens) as the next step.