Learning Loft Editorial Team · Reviewed by qualified UK teachers

Responsible use of AI at Learning Loft

Learning Loft uses AI to remove mechanical work — never to replace teacher judgement. Every model output is reviewable, every design decision keeps the teacher, school or parent in control, and every byte of pupil data is handled under UK GDPR.

Our principles

  1. Teacher-in-the-loop. AI drafts; teachers decide. No pupil-facing output is auto-published.
  2. Purpose-built, not general chat. Each tool is scoped to one educational task with clear inputs and outputs.
  3. Specification-aware. Marking and feedback are grounded in the mark scheme or rubric the teacher supplies.
  4. UK data, UK schools. Data is processed in UK-hosted infrastructure under UK GDPR.
  5. No training on pupil work. Pupil submissions are not used to train third-party models.
  6. Transparent by default. Teachers can see what the AI produced, edit it, and export the audit trail.

Where AI is used across the platform

  • LoftTick — drafts feedback comments against the mark scheme; the teacher confirms every judgement.
  • LoftTalk — transcribes spoken responses and applies a rubric; the teacher awards the final grade.
  • LoftTrack — summarises question-level analysis into readable insights for teachers and leaders.
  • LoftTest — generates quiz variants and distractors from curriculum topics teachers approve.
  • LoftTeach — produces draft observation notes, coaching scripts and report paragraphs.
  • LoftLibrary — supports reading fluency measurement and comprehension question generation.

What we do not do

  • We do not send pupil work to unvetted third parties.
  • We do not auto-publish AI-written comments to pupils or parents.
  • We do not use AI to make safeguarding decisions.
  • We do not claim AI replaces qualified teachers.

Safeguarding considerations

Safeguarding disclosures, welfare concerns and pastoral judgements are never delegated to AI. Any content flagged for concern is routed to a human reviewer through your school's existing safeguarding process, in line with UK statutory guidance.

See it in the tools

Each product page explains where automation sits in the workflow and where the teacher confirms: LoftTick for marking and feedback, LoftTrack for question level analysis, and LoftTalk for oracy assessment. Whole-school rollout, DPA and CPD are covered on For Schools, and our position on what should and should not be automated is set out in the teacher workload guide.

Contact

For due-diligence packs, DPA copies or specific questions about model use, email info@learningloft.co.uk.

Explore further

Security and data protection

UK hosting, GDPR, RBAC and encryption.

Platform overview

How the Learning Loft apps fit together.

For schools

Whole-school onboarding, DPA and CPD.

For teachers

Guides and tools for the classroom.

Marking and feedback

How teachers use LoftTick responsibly.

LoftTick

Smart marking with the teacher in the loop.

LoftTrack

Question level analysis, reported not decided.

LoftTalk

Oracy assessment with teacher-confirmed rubrics.

Teacher workload

What we automate, and what we never will.

Frequently asked questions

Where does Learning Loft use AI?

AI supports specific tasks inside our tools: drafting marking feedback in LoftTick, transcribing spoken responses in LoftTalk, summarising question-level analysis in LoftTrack, generating quiz variants in LoftTest, and producing draft observation notes and report paragraphs in LoftTeach. In every case a teacher reviews and confirms the output.

Does AI replace the teacher?

No. Learning Loft is designed on a teacher-in-the-loop principle. AI produces a first draft or a data summary; the teacher, tutor or school leader always makes the final professional judgement. Nothing is auto-published to pupils or parents without human confirmation.

How is pupil data protected?

Pupil work and recordings are processed under a Data Processing Agreement with each school, stored in UK-hosted infrastructure with role-based access, and are not used to train third-party models. Access is time-limited, signed and auditable.

Which models does Learning Loft use?

We use a small set of vetted, commercially licensed models chosen for accuracy, safety and cost. Deterministic work (parsing, formatting) is cached to avoid unnecessary calls. We do not use pupil work to train external models.

How should teachers review AI output?

Treat every AI suggestion as a draft. Check that feedback matches the specification and the pupil's actual response, that data summaries reflect the underlying evidence, and that anything shared with pupils or parents reads in your professional voice.

Who do I contact with questions?

Email info@learningloft.co.uk for questions about how we use AI, our Data Processing Agreement, or to request additional detail for your school's due-diligence process.

One connected platform

The Learning Loft ecosystem

Each app shares one login, one progress record and one pedagogy — so assessment, teaching and tutoring stay joined up.

LoftTrack

QLA tracking & gap analysis

LoftTick

Spec-aware exam marking

LoftTeach

Observations & coaching

LoftTalk

Oracy & speaking practice

LoftMath

Fluency & reasoning

LoftLibrary

Reading & comprehension

LoftClass

Live tutoring

LoftTopia

Gamified learning

  • Made by qualified UK teachers
  • National Curriculum aligned
  • Safeguarding-ready
  • GDPR compliant · UK-hosted
  • Built with UK schools and educators
  • Evidence-informed pedagogy

Bring Learning Loft to your school

Trusted by UK schools. Made by teachers. Built around safeguarding, evidence-based practice and reducing teacher workload.