You already know which pupils need support.
The ones you don't, are the risk.
Lumii is the pastoral and safeguarding platform for schools — the layer that hears what children don't say out loud.
Daily pupil voice, staff concerns and behaviour — one record per child, one platform for safeguarding, wellbeing and behaviour.
Built for DSLs, pastoral leads and heads of wellbeing.
A UAE company · built and ready in an hour · no IT changes
16,000+ pupil-voice conversations · 122 staff-raised concerns · 500+ behaviour flags · every flag reviewed by a trained adult
Children are already telling AI what they won’t say out loud.
72% of teenagers have used an AI companion and 52% use one regularly (Common Sense Media, 2025) — with a meaningful share treating these tools as a friend or source of emotional support. In the UK, half of vulnerable children say chatting to an AI feels like talking to a friend (Internet Matters, 2025). It is happening in every school today, on consumer tools with no safeguarding architecture, no cultural context, and no human in the loop.
Lumii is built on the opposite architecture to the consumer tools that wariness is aimed at — anonymous, walled-garden, UAE-hosted, not a general AI assistant. It does not introduce a new behaviour. It takes one that is already widespread and makes it safe, age-appropriate, and visible to the right adults.
The only question is whether the AI is safe — and whether anyone human is listening.
Three commitments behind everything Lumii does.
Real time pupil voice
Lumii is the channel that hears pupils every day — in their own words, without prompts and without pressure to disclose. Not a survey. Not a one-off. The daily picture, built from what pupils share when nothing else is asking.
Reads, reasons, surfaces what to look at
Every interaction is read in context. The system reasons about what it means and makes a flagging decision — most stay in Green; Amber routes to Reviewers; Purple and Red trigger an immediate alert. Every flag carries the reasoning behind it.
Schools decide what happens next
Every Purple, Red and Amber alert is reviewed by the school's wellbeing or safeguarding team — DSLs, Heads of Wellbeing, counsellors, pastoral leads — before any action is taken.
Lumii surfaces what to look at. Trained adults decide what to do about it.
Streams into an ocean.
Pupil voice. Staff concerns. Behaviour tracking. Each is a stream — the ocean is the whole picture of the child, held in one case record. A behaviour note alone reads as a discipline problem; next to a pupil-voice dip and a staff observation it reads as a child who needs help.
Surveys surface what was asked. The streams surface what is happening.
What comes out the other side.
In their own words.
Lumii works differently for different people. Start where you fit.
For multi-school networks
Network-level consistency, group dashboards, and structured contribution aligned with ECA's child protection and early care & education priorities.
See how it works for groups → Trust & safetyHow safeguarding actually works
The alert model, the human oversight, and the data boundaries that make Lumii safe to deploy at scale.
See the safeguarding architecture → ParentsFor families with questions
What pupils experience, what schools see, and what Lumii will and will not do with what your child shares.
Read answers to common questions →See what your school could be hearing.
A one-week trial. Setup in 1 hour. No accounts, no passwords, no infrastructure changes.
