Role Definition
| Field | Value |
|---|---|
| Job Title | Home Tutor — PRU / Medical Needs |
| Seniority Level | Mid-level (3-10 years teaching experience) |
| Primary Function | LA-employed qualified teacher who delivers 1:1 education to children unable to attend school. Reasons include cancer treatment, chronic illness, school refusal/anxiety, permanent exclusion, and awaiting placement. Teaches full curriculum (maths, English, science) in pupils' homes, hospitals, or community settings. Manages complex SEN/SEMH needs, safeguards vulnerable children as sole professional in private settings, and liaises with schools, social services, CAMHS, and medical teams. Operates under Section 19 of the Education Act 1996 — a statutory duty on local authorities. |
| What This Role Is NOT | NOT a private tutor (no statutory role, no safeguarding duty, no LA employment, no vulnerable cohort). NOT a Tutor (25-3041 — scored Yellow Urgent 26.8 — commercial, no licensing, no safeguarding mandate). NOT a Special Education Teacher (25-2050 — school-based, classroom setting, higher autonomy over IEP/EHCP design). NOT a Substitute Teacher (25-3031 — temporary classroom cover, no 1:1, no home visits). NOT a Teaching Assistant (25-9045 — support role, works under teacher direction). |
| Typical Experience | 3-10 years. QTS (UK) or state teaching licence (US equivalent). Enhanced DBS/background check mandatory. Specialist training in SEMH, trauma-informed practice, medical needs education, safeguarding Level 3. Often former mainstream teachers who moved into alternative provision. |
Seniority note: Newly qualified teachers rarely enter this role — it requires significant classroom experience before working alone in private homes with vulnerable children. More experienced home tutors managing the most complex cases (palliative care, severe psychiatric conditions) carry the strongest protection.
Protective Principles + AI Growth Correlation
| Principle | Score (0-3) | Rationale |
|---|---|---|
| Embodied Physicality | 2 | Must be physically present in the child's home, hospital ward, or community setting — unstructured, unpredictable environments that change daily. Not a structured classroom. However, the work is primarily instructional rather than involving physical care (unlike SEN TAs), so scores 2 rather than 3. |
| Deep Interpersonal Connection | 3 | Trust IS the value. These children are often anxious, traumatised, medically fragile, or behaviourally challenging. The tutor is frequently the only consistent professional adult in the child's life. Building trust with a school-refusing teenager or a child undergoing chemotherapy is irreducibly human. Parents allow this person into their home because they trust them — not a credential, but a relationship. |
| Goal-Setting & Moral Judgment | 2 | Makes significant independent judgment calls — when to push academically, when to stop because the child is in pain, when to report a safeguarding concern, how to adapt the curriculum in real time. Works alone without supervision in private homes. Higher autonomy than a TA but reports to a service coordinator. Mandatory reporter in vulnerable settings. |
| Protective Total | 7/9 | |
| AI Growth Correlation | 0 | AI adoption neither creates nor destroys demand. Demand driven by exclusion rates, medical need, EHCP numbers, and LA budgets — not by technology adoption. |
Quick screen result: Protective 7/9 with Neutral Correlation — strong Green Zone. Deep interpersonal connection at maximum; physical presence and judgment both significant. This role is structurally protected by the nature of the children it serves.
Task Decomposition (Agentic AI Scoring)
| Task | Time % | Score (1-5) | Weighted | Aug/Disp | Rationale |
|---|---|---|---|---|---|
| 1:1 curriculum delivery — maths, English, science adapted to individual needs, differentiated for SEN/SEMH, delivered in pupil's home or hospital | 30% | 2 | 0.60 | AUGMENTATION | AI tools (MagicSchool.ai, Khanmigo, ChatGPT Study Mode) generate resources and practice materials. But the tutor delivers them face-to-face, adapting in real time to whether the child is in pain, anxious, or having a good day. Human-led, AI-assisted. |
| Relationship building & pastoral care — establishing trust with anxious/ill/traumatised children, managing school refusal, providing emotional stability and routine | 20% | 1 | 0.20 | NOT INVOLVED | Irreducibly human. A child who refuses to leave their bedroom for a stranger will not engage with an AI. Building rapport with a teenager on chemotherapy requires human empathy, patience, and physical presence. This is why the role exists. |
| Safeguarding & welfare — mandatory reporting, monitoring for abuse/neglect, working with children in crisis, lone working in private homes | 15% | 1 | 0.15 | NOT INVOLVED | Sole professional adult in a private home with a vulnerable child. Legal duty to observe, assess, and report. Cannot be delegated to AI — requires physical presence, professional judgment, and personal legal accountability. |
| SEN/SEMH needs management — de-escalation, managing complex behaviour, supporting children through medical treatment, sensory/emotional regulation | 10% | 1 | 0.10 | NOT INVOLVED | Reading a child's emotional state, calming a panic attack, knowing when to stop teaching because the child is in pain — embodied, relational, judgment-intensive. No AI involvement. |
| Multi-agency liaison — communicating with schools, social services, CAMHS, medical teams, parents/carers, contributing to TAC/CIN meetings | 10% | 2 | 0.20 | AUGMENTATION | AI can draft meeting notes and generate progress summaries. But advocating for a child in a multi-agency meeting, negotiating reintegration with a headteacher, and managing parent relationships require human communication and professional credibility. |
| Lesson planning & curriculum adaptation — creating individualised schemes of work, adapting full curriculum to 1:1 home setting, selecting resources | 10% | 3 | 0.30 | AUGMENTATION | AI significantly accelerates lesson planning — generating differentiated worksheets, adapting reading levels, creating visual resources. Human still leads because curriculum must be adapted to the child's emotional state, medical schedule, and home environment in ways AI cannot anticipate. |
| Administration & reporting — progress reports, attendance records, reintegration plans, EHCP contributions, LA paperwork | 5% | 4 | 0.20 | DISPLACEMENT | Structured, rule-based documentation. AI auto-generates progress reports from observation notes, tracks attendance, and drafts EHCP contributions. School MIS systems handle data entry. |
| Total | 100% | 1.75 |
Task Resistance Score: 6.00 - 1.75 = 4.25/5.0
Displacement/Augmentation split: 5% displacement, 50% augmentation, 45% not involved.
Reinstatement check (Acemoglu): Limited new task creation. Some emerging work: configuring AI-powered adaptive learning tools for individual students, interpreting AI-generated progress analytics, managing children's use of AI study tools. These supplement rather than transform the role.
Evidence Score
| Dimension | Score (-2 to 2) | Evidence |
|---|---|---|
| Job Posting Trends | 0 | UK-specific niche role — no aggregated posting data. LA home tutor positions are advertised through council job boards, not major platforms. Chronic difficulty recruiting into alternative provision broadly (AP teacher vacancy rates exceed mainstream). Stable but small market — not growing explosively. |
| Company Actions | 0 | No LAs cutting home tutor positions citing AI. UK government announced AI tutoring tools for 450,000 disadvantaged mainstream pupils (DfE 2026), but this targets classroom supplementation, not replacement of statutory 1:1 provision for medically unfit or excluded children. No evidence of AI-driven workforce reduction in AP. |
| Wage Trends | -1 | LA-employed teachers on Teachers' Pay and Conditions. UK experienced teacher pay 9% lower in real terms than 2010/11 (NEA/IFS data). Home tutors typically on Main Pay Range or Upper Pay Scale — GBP 30,000-45,000. Not declining specifically, but stagnant against inflation. |
| AI Tool Maturity | 1 | AI tutoring tools (Khanmigo, ChatGPT Study Mode, MagicSchool.ai) augment lesson planning and resource creation. No tool can enter a child's home, assess their welfare, build trust, or deliver curriculum to a child too anxious to engage. AI augments admin; core tasks have no viable AI alternative. |
| Expert Consensus | 1 | Strong consensus that vulnerable children require human educators. DfE guidance explicitly mandates suitable education provision — human delivery. EdTech Magazine, CEC, and WEF all frame AI in education as augmentation. Brookings places education among lowest-automation-potential sectors (<20% of tasks). |
| Total | 1 |
Barrier Assessment
Reframed question: What prevents AI execution even when programmatically possible?
| Barrier | Score (0-2) | Rationale |
|---|---|---|
| Regulatory/Licensing | 2 | QTS (Qualified Teacher Status) required in UK. Section 19 Education Act 1996 places statutory duty on LAs to provide education — implies human provision. Enhanced DBS check mandatory. Ofsted inspects AP quality. No regulatory pathway exists for AI to fulfill this duty. |
| Physical Presence | 2 | Must enter the child's home, sit at their kitchen table, accompany them to hospital. Unstructured, unpredictable environments — one day a bedroom, next day a hospital ward, next day a park. Physical presence is the delivery mechanism. |
| Union/Collective Bargaining | 1 | NEU and NASUWT represent LA-employed teachers. Teachers' Pay and Conditions apply. Unions have adopted strong positions that AI enhances teaching, not replaces teachers. Moderate protection — weaker than NHS unions but meaningful. |
| Liability/Accountability | 2 | Sole professional adult with a vulnerable child in a private home. In loco parentis. If a child is harmed, neglected, or a safeguarding concern is missed, the tutor bears personal and professional liability — potential criminal prosecution, fitness-to-practise proceedings, loss of QTS. AI has no legal personhood to bear this accountability. |
| Cultural/Ethical | 2 | Parents of sick, excluded, or anxious children will not accept an AI teaching their child in their living room. The cultural barrier is absolute for this cohort — these are the families who already distrust the system, who need a human face, who require someone to notice when their child is struggling. Society categorically rejects AI as sole educator for the most vulnerable. |
| Total | 9/10 |
AI Growth Correlation Check
Confirmed 0 (Neutral). AI adoption does not create or destroy demand for home tutors. Demand is driven by permanent exclusion rates (rising — up 44% in England 2021-2024), EHCP numbers (up 60% since 2018), school refusal/anxiety prevalence (surging post-pandemic), and LA budgets. The UK government's AI tutoring initiative (450,000 pupils by 2027) targets mainstream disadvantaged students, not the statutory 1:1 provision for children who cannot attend school. This role is Green (Stable) — demand is independent of AI adoption.
JobZone Composite Score (AIJRI)
| Input | Value |
|---|---|
| Task Resistance Score | 4.25/5.0 |
| Evidence Modifier | 1.0 + (1 x 0.04) = 1.04 |
| Barrier Modifier | 1.0 + (9 x 0.02) = 1.18 |
| Growth Modifier | 1.0 + (0 x 0.05) = 1.00 |
Raw: 4.25 x 1.04 x 1.18 x 1.00 = 5.2156
JobZone Score: (5.2156 - 0.54) / 7.93 x 100 = 59.0/100
Zone: GREEN (Green >=48, Yellow 25-47, Red <25)
Sub-Label Determination
| Metric | Value |
|---|---|
| % of task time scoring 3+ | 15% |
| AI Growth Correlation | 0 |
| Sub-label | Green (Stable) — <20% task time scores 3+, Growth != 2 |
Assessor override: None — formula score accepted. The 59.0 correctly positions this role between the SEN Teaching Assistant (61.9 — more physical care, more barrier-protected) and the Substitute Teacher (50.2 — no 1:1 relationship, weaker barriers). The home tutor's higher autonomy and safeguarding responsibility compared to a TA, combined with slightly less physical care work, produces an appropriate score.
Assessor Commentary
Score vs Reality Check
The 59.0 honestly reflects reality. This role is protected by an unusual combination: statutory legal mandate (Section 19), physical presence in private homes, sole-adult safeguarding accountability, and deep trust relationships with the most vulnerable children in the education system. The Green (Stable) sub-label is correct — only 15% of task time involves work scoring 3+, and the daily experience of a home tutor changes minimally with AI adoption. The score sits appropriately below the SEN TA (61.9) because the SEN TA has more irreducible physical care (toileting, restraint) and below Special Education Teacher All Other (62.5) because that SOC includes hospital/homebound teachers with higher curricular autonomy.
What the Numbers Don't Capture
- Budget vulnerability is the existential threat, not AI. LA alternative provision budgets are under severe pressure. When councils need to cut, home tutor hours are reduced — children receive 5 hours/week instead of 15, or are placed on waiting lists. AI is irrelevant to this funding squeeze.
- The role is emotionally isolating in ways no score captures. Home tutors work alone in private homes with children in crisis. No staffroom, no colleagues, no shared break. Professional isolation combined with exposure to domestic dysfunction, severe illness, and trauma creates high burnout and turnover.
- The UK government's AI tutoring initiative creates confusion, not displacement. The DfE's plan to deploy AI tutoring tools for 450,000 disadvantaged mainstream pupils may lead commentators to claim "AI is replacing tutors." This is a different cohort, different setting, different purpose. Home tutors for excluded/medical children are not the target.
Who Should Worry (and Who Shouldn't)
Home tutors working 1:1 with children with severe medical needs — cancer treatment, complex chronic illness, hospital-based education — are among the most protected educators in the country. No technology enters a child's hospital room to teach them GCSE maths while they recover from chemotherapy. Home tutors managing permanently excluded pupils with severe SEMH are equally protected — the behavioural complexity and safeguarding intensity are irreducibly human. The version most exposed: home tutors whose caseload consists primarily of higher-functioning students with mild school refusal, where the work is closer to academic tutoring than pastoral/safeguarding work. If AI tutoring platforms improve and the child's primary need is curriculum catch-up rather than emotional support, budget-constrained LAs might consider reducing human hours. The single biggest separator: whether your value comes from the relationship and safeguarding (protected) or from curriculum delivery alone (more exposed).
What This Means
The role in 2028: Home tutors spend less time on lesson planning (AI generates differentiated resources instantly) and less time on paperwork (AI drafts progress reports and EHCP contributions). More time is freed for the relational and pastoral work that is the actual reason this role exists. Adaptive AI learning platforms become supplementary tools — the tutor assigns AI-generated practice between visits, reviews AI analytics on student progress, and focuses face-to-face time on the human connection these children desperately need.
Survival strategy:
- Specialise in complex needs. Children with severe medical conditions, psychiatric needs, or profound SEMH require the deepest human expertise. Training in trauma-informed practice, medical education, and complex safeguarding makes you the tutor LAs cannot replace.
- Master AI teaching tools. Become fluent with MagicSchool.ai, adaptive platforms, and AI-powered differentiation. Use them to demonstrate impact — better-prepared lessons, faster resource creation, measurable student progress. The tutor who shows data-driven outcomes protects their post.
- Build your multi-agency network. The strongest job security comes from being embedded in the professional ecosystem — known by name to social workers, CAMHS clinicians, educational psychologists, and headteachers. You are irreplaceable when you are the connective tissue between services.
Timeline: 5+ years for core role stability. Planning and admin tasks transform within 2-3 years. The relational, safeguarding, and pastoral functions remain indefinitely — these children will always need a trusted human adult who comes to their door.
