AI Intelligence

Ask questions like "Which Year 9 students missed more than 5 days this term?" or combine attendance with homework completion patterns — instant answers from your school's aggregates without exporting spreadsheets.

See how it works

Retrieval-augmented generation lets staff query school policies and procedures in natural language, always grounded in your actual documents.

See how it works

Generate personalised report card comments from assessment data, learning notes, and student context — reviewed and refined by teachers before publishing.

See how it works

Draft first-pass comments across multiple subjects in one run where your school enables it; principals cannot publish until comments leave AI-draft state and are human-reviewed.

See how it works

In Assessment & Reporting, AI drafts feedback and marking guidance from the task brief and, when you attach a rubric, from criteria, band descriptors, and weightings — surfaced in a marking queue so teachers accept, edit, or discard every suggestion. Nothing is published to students or parents until a human approves it.

See how it works

For configured online exams, staff can request AI-assisted suggestions on short-style responses alongside attempt review — same human-in-the-loop rules as classroom assessment marking.

See how it works

Class-level performance narrative from assessment aggregates — surfaced on each assessment detail page for teachers and HODs.

See how it works

Teachers start from a structured brief (topic, year level, duration) and receive a draft task description to edit before publishing — optional and never sent to families without your save.

See how it works

Class-level summary of who is on track, overdue, or not done — from completion aggregates for a task so teachers can follow up without reading every row first.

See how it works

From learning-analytics aggregates across classes and time, generate an advisory cohort narrative for heads of year and learning support — class means and trends only, no individual student narratives in the prompt.

See how it works

Turn imported NAPLAN cohort statistics into a plain-language gap summary for leadership — grounded in school-held results and labelled AI-generated for review.

See how it works

Automatically prepare comprehensive handover briefs for casual relief teachers, including class profiles, medical alerts, and lesson context.

See how it works

AI-powered analysis of incident patterns and language cues to support wellbeing teams with early safeguarding identification.

See how it works

Generate a concise pastoral briefing from aggregate attendance shape, wellbeing record counts, incident severity tallies, fee pressure, and recent grade averages — counselling notes, medical details, and incident narratives never leave your school boundary.

See how it works

Multi-signal heuristic risk tiers from attendance, behaviour, mood, and incidents — with optional LLM explainability for tier drivers and absenteeism alert narratives on the wellbeing dashboard.

See how it works

Automatically surface unusual patterns in attendance, behaviour, and academic performance so nothing slips through the cracks.

See how it works

Machine learning models identify students at risk of chronic absenteeism weeks before it becomes critical, enabling early intervention.

See how it works

Auto-generate student summaries for parent-teacher meetings, pulling together attendance, grades, wellbeing, and behavioural data.

See how it works

Turn live operational aggregates — attendance today, enrolment pipeline, wellbeing tier counts, and open work orders — into a plain-language executive briefing for leadership.

See how it works

Monday-style executive readout generated from live operational aggregates, 7-day incident signals, and enrolment pipeline shape — for principals and leadership teams.

See how it works

AI narrative on upcoming calendar load — entry-type mix, overlap signals, and busy days from aggregate metrics only, on web and CampusBridge School.

See how it works

Monthly cashflow commentary, payment-plan outreach guidance, and fee-assistance eligibility notes — whole-school aggregates for business managers and registrars.

See how it works

Plain-language summary of application stage mix and conversion pressure from enrolment pipeline aggregates.

See how it works

AI summary for board members from portal aggregates and attendance trends — no individual student data in the model prompt.

See how it works

Summarise open work-order counts, priorities, and age buckets for facilities planning — order titles never sent to the model.

See how it works

AI narrative highlighting staff policy acknowledgement gaps from aggregate compliance metrics on the policy library.

See how it works

Teaching load heatmaps from homeroom assignments and roster totals — department equity watchlists and rules-based overload flags for leadership planning.

See how it works

Break language barriers with real-time translation of notices, messages, and communications into 30+ community languages.

See how it works

On publish, notices can be batch-translated into each family's preferred language — with staleness detection when source text changes. Families also get on-demand translation in the portal and CampusBridge Family app.

See how it works

Draft a full multi-section school newsletter from a section outline in the Communication Assistant — on web and CampusBridge School.

See how it works

When logging a behaviour incident, AI analyses the description and suggests the incident type and sub-type (e.g. bullying → cyberbullying) with a confidence score — staff accept or override before saving. PII-stripped before the model call.

See how it works

For HIGH or CRITICAL incidents, AI suggests Positive Behaviour Support and trauma-informed de-escalation strategies tailored to the incident context — labelled AI-generated for staff consideration, never surfaced to students or parents.

See how it works

AI drafts a suspension letter from incident details, sanction duration, and return conditions — presented for Principal or Deputy review with a mandatory human-approval checkbox before the letter is generated or sent.

See how it works

Staff pick a merit category (and optional seed words); AI suggests polished wording for the reason field — always edited and approved by a human before the merit is saved.

See how it works

Structured prompts help staff turn short answers into a clear award nomination narrative — draft stays AI-labelled until a human edits and submits.

See how it works

From aggregate merit statistics (no student names in the model prompt), AI produces a short equity-style insight for leaders — useful for term reviews and assembly planning.

See how it works

When configured, a single-sentence AI nudge reminds staff to recognise students who have not yet received merits in the current term window — supportive tone, no individual profiling.

See how it works

For schools using School Houses, AI reads term-to-date house point aggregates and explains whether the spread looks healthy or unusually skewed — output is labelled for review before sharing.

See how it works

At season close, AI drafts an end-of-season story from standings and category totals (no student PII) — publish to the community only after staff review.

See how it works

Optional AI assist when staff record inter-house carnival or event results — speeds data entry while organisers verify every score.

See how it works

For boarding students, AI can draft a short end-of-term pastoral summary from category counts and follow-up tallies only — confidential note bodies are never sent to external models.

See how it works

AI-powered moderation screens parent-teacher messages for inappropriate content before delivery, protecting staff and maintaining professional communication.

See how it works

Student data is automatically pseudonymised before any external AI call — names, DOBs, and identifiers never leave your school's boundary.

See how it works

Planned for a future release. Parent message moderation already screens inappropriate content before delivery.

See how it works

Per-school usage logging (model, tokens, feature), estimated cost visibility, optional UTC-monthly caps, and clear UI when AI is disabled or not configured — pilot-friendly controls without surprise spend.

See how it works