AI for schools and trusts.
Marking support built with the DfE, with admin work alongside it and the teacher confirming every grade.
What AI can do in a school.
AI earns its place on the jobs that come round every week and run against a scheme the school already has: marking a class set, drafting reports from the term’s marks, adapting resources across reading levels, and turning attendance data into letters home.
Where schools like yours start.
It marks the class set against your own scheme and shows the reasoning behind each grade.
It drafts each report from the term’s marks and notes, in the school’s own format.
It adapts the materials you already have across reading levels and drafts the variants a class needs.
It turns attendance data into the letters that go home and files what it sent.
It works inside Google Classroom and Microsoft Teams, so nobody signs into a separate system.
Evenings back from the marking, and the same standard applied to every class set. It also cuts the licences the school no longer needs, and makes the material you already have easier to find.
The audit works out yours.
How we find where AI fits in a school.
An AI Audit and Transformation comes first, fixed in scope and fee: we time the work that repeats every week and hand back a costed, prioritised 12 month roadmap naming who owns each workflow. OpenKit becomes your business’s embedded AI team: the people who ran your audit stay on to deliver it. If the answer is a change to the marking policy, we will say so.
The build is part of the same engagement, and what goes live in Classroom or Teams is measured against the evenings the audit counted.
The AI readiness check is free, and it tells you whether an audit is worth your term.
How we engage.
Most engagements start with the AI Audit and Transformation. What follows depends on what it surfaces: many teams keep us on as their Embedded AI Lead, and some commission a bespoke build where nothing off the shelf will fit. The AI Charter runs underneath whichever way it goes.
- AI Audit and Transformation Fixed in fee and scope: we measure where the evenings go and hand back a costed, prioritised 12 month roadmap, yours to keep.
- Embedded AI Lead We stay inside your team, build the plan into the school’s week and train the staff who run it.
- AI Charter It writes down what AI is used for in the school and who owns each workflow it touches.
Related work.
Rubrical is our own platform, built with the Department for Education, and CodeKit is the AI programming tutor we built for Wolsingham School in County Durham.
Rubrical Marking a class set against the school’s own mark scheme Built with the Department for Education and live in fourteen schools. Every draft shows the criteria it marked against, and the teacher confirms or overrides each grade before it reaches a student. 52% Less marking time 
CodeKit Coaching a student through Python in the browser, on the school’s own lessons It reviews what a student has written and comes back with feedback and a next step to try, rather than handing back the answer. CodeKit became the basis of our bid when the Department for Education opened a competition for AI tools in education, and we won the first phase. Pro bono For Wolsingham School The Department for Education records that figure on page 98 of its schools white paper, Every child achieving and thriving (February 2026), alongside the Department’s initial £1m award to Rubrical in 2024 and a further £1m from Innovate UK.
Secure and private, as standard.
Every automated step leaves a record the school can stand behind, and pupil data never leaves to train another model. Private AI
ISO 27001, ISO 9001 and Cyber Essentials are certifications we hold, and we operate to UK GDPR, pupil data included.
The questions that decide it.
School leaders ask us these before they let anything near a mark.
What does an AI consultancy for UK schools actually do?
OpenKit runs a fixed-scope, fixed-fee audit of where AI fits in a school or trust as the opening stage of an AI Audit and Transformation, builds the workflows that justify the work, trains staff to use them safely inside Google Classroom or Microsoft Teams, and either hands over or stays on as an Embedded AI Lead on retainer. What we build belongs to the school and runs on its own mark schemes and materials. Most engagements open with the audit because it settles what a school or trust already has and what is genuinely worth building, before any budget is committed.
Will AI replace teachers?
No, and the DfE product safety standards explicitly require AI tools to mitigate cognitive deskilling. The workflows that earn their place are marking first-passes, lesson planning support, parent comms drafting, and admin triage: places where the teacher’s professional judgement is being eroded by volume, not where it is being applied. The EEF lesson-planning study found teachers saved around 31% of planning time but still owned every plan.
How does this work with the DfE Generative AI product safety expectations?
The DfE updated its expectations into formal product safety standards on 19 January 2026, covering content filtering, distress detection, cognitive deskilling, social and emotional development, manipulation defences, and an absolute ban on using learner or teacher IP for training. We design against the standards before the build starts, document the alignment in writing, and hand the documentation to the DSL and the DPO so it can be presented at a governor meeting or an ICO audit without rework.
What about JCQ and exam integrity?
JCQ’s "AI Use in Assessments" guidance treats undeclared AI use in submitted work as malpractice, with sanctions including disqualification and multi-year debarment, and recorded 1,125 GCSE/A-level disqualifications plus around 2,000 mark deductions for AI malpractice in 2025 alone. Any system we build that touches assessed work keeps the student declaration trail intact, the teacher in the moderation loop, and the moderation evidence in a form the centre can defend at a JCQ inquiry. Rubrical was built against this exact framework.
Can we keep pupil data on-prem or in the UK?
Yes. Default deployment is Azure UK tenant or equivalent UK/EU-hosted region; on-prem is supported for trusts and universities with full sovereignty requirements. Models run behind the trust’s own identity provider, logs are retained for moderation rather than training, and the contract grants the trust full data export plus permanent deletion at the end of the engagement.
Does this integrate with Google Classroom, Microsoft Teams, and our MIS?
Google Classroom and Microsoft Teams for Education are evidenced production integrations: Rubrical syncs assignments and returns grades through both. Microsoft 365 / SharePoint and Entra ID identity are standard. MIS integration with SIMS, Arbor, or Bromcom is scoped per engagement; where a standard connector does not exist we build a custom API, but ask us about your specific MIS at the audit stage.
What happens with the EU AI Act if we deploy an assessment tool?
The EU AI Act Annex III classifies systems that evaluate learning outcomes, determine access to education, or monitor candidate behaviour during tests as high-risk. The June 2026 Digital Omnibus agreement deferred the stand-alone high-risk compliance deadline to 2 December 2027, but the classification is settled and the deploying institution still carries its own duties, including a Fundamental Rights Impact Assessment under Article 27. We write the FRIA alongside the build, document the risk controls, and stay available through enforcement.
How is an education AI engagement structured?
Every engagement opens with the audit stage of an AI Audit and Transformation: fixed scope, fixed fee, scoped up front, mapping where AI fits and where it should not go. The build that follows puts that work live inside your stack. Rubrical, our DfE partnership, ran from discovery through to a fourteen-school production rollout; what a Bespoke Build involves is scoped against what the audit surfaces, not fixed in advance.
What does it cost for a school or a MAT?
The audit is a fixed fee and fixed scope, agreed before kickoff, so there are no surprises on the invoice and no retainer is required to reach it. What follows, whether that is the transformation build, an Embedded AI Lead on retainer or a Bespoke Build, is scoped and priced after the audit against what it surfaces. We give you indicative figures on a discovery call.
What makes OpenKit different from Century, Sparx, Eedi, or Oak’s Aila?
Those are products you subscribe to. We are a build engagement: the audit chooses the workflow, the transformation build makes it work inside your stack, and we either hand over or stay on as an Embedded AI Lead. We are the only UK consultancy on this list with a DfE programme-partner build in production (Rubrical) plus a second school-built tutor (CodeKit), and ISO 27001, ISO 9001, Cyber Essentials and UK GDPR all held at once.
How does OpenKit compare to Faculty for education AI work?
Faculty (now part of Accenture) operates at enterprise scale: DfE central programmes, FTSE 100 buyers, Big Four day rates. OpenKit operates at trust scale: independent schools, MATs of three to thirty schools, FE colleges, university departments. The audit-first engagement keeps the entry point at a scale most trusts can sign off without a full procurement exercise. If a Faculty engagement is the right fit we will say so during the audit.
Find your first workflow.
We start with a conversation, audit where AI actually pays back, and build the first automation into how your team already works. We reply within one working day.