Role Definition
| Field | Value |
|---|---|
| Job Title | Tutor |
| Seniority Level | Mid-Level (2-5 years experience) |
| Primary Function | Provides individualised academic instruction to students outside the classroom — typically 1:1 or small-group sessions in subjects like maths, science, reading, writing, and test prep. Assesses student learning gaps through diagnostic questioning, adapts instruction to individual needs, builds rapport and motivation, creates practice materials, tracks progress, and communicates with parents and teachers. Works independently, for tutoring companies, or on online platforms. |
| What This Role Is NOT | NOT a classroom teacher (licensed, institutional employment, safeguarding duties, curriculum authority — scored 68-75 Green). NOT a self-enrichment teacher (non-academic, recreational subjects — scored 32.4 Yellow). NOT a teaching assistant (works under teacher direction in a school — scored 51.2 Green). NOT a corporate training specialist (occupational focus — scored 27.6 Yellow). |
| Typical Experience | 2-5 years. Bachelor's degree typical but not universally required. Subject expertise is the primary credential. Some specialise with certifications (e.g., Orton-Gillingham for dyslexia, certified test-prep instructor). BLS SOC 25-3041. |
Seniority note: Entry-level tutors (0-2 years) delivering routine homework help online would score deeper Yellow or borderline Red — they compete directly with AI tutoring tools and lack a student base. Senior tutors (7+ years) specialising in complex learning needs, executive function coaching, or elite test prep would score higher Yellow or borderline Green — their diagnostic expertise, reputation, and client relationships create durable protection.
Protective Principles + AI Growth Correlation
| Principle | Score (0-3) | Rationale |
|---|---|---|
| Embodied Physicality | 1 | In-person tutoring involves sitting with a student, pointing at their work, using whiteboards, and physically managing materials. But a large and growing share of tutoring is online — Zoom/platform-based sessions with shared screens. The physical component is real for in-person tutors but absent for online tutors. Average across the role: minor. |
| Deep Interpersonal Connection | 2 | For struggling students, the tutor-student relationship is central. Building trust with an anxious teenager, motivating a child who hates maths, reading non-verbal cues of confusion — these interpersonal skills drive outcomes. Parents often choose a tutor based on personal connection with their child. Less relational for transactional homework help; deeply relational for ongoing student development. |
| Goal-Setting & Moral Judgment | 1 | Tutors make pedagogical judgments — diagnosing root causes of misunderstanding, choosing when to push vs. support, adapting instruction in real time. But these are applied within a subject domain, not ethical or high-stakes decisions. No one faces legal consequences for a bad tutoring session. |
| Protective Total | 4/9 | |
| AI Growth Correlation | -1 | AI tutoring tools directly compete with human tutors. Khanmigo, ChatGPT, Photomath, and Duolingo Max provide personalised academic instruction at near-zero marginal cost. More AI adoption = less demand for routine human tutoring. Not -2 because specialised, in-person, and complex-learner tutoring retains demand independent of AI. |
Quick screen result: Protective 4/9 with negative AI growth correlation — predicts Yellow Zone. Moderate interpersonal protection but weak physical and structural barriers. The negative growth correlation distinguishes tutors from classroom teachers.
Task Decomposition (Agentic AI Scoring)
| Task | Time % | Score (1-5) | Weighted | Aug/Disp | Rationale |
|---|---|---|---|---|---|
| Assess learning gaps & adapt instruction plans — diagnostic questioning, identifying misconceptions, adjusting teaching approach | 20% | 3 | 0.60 | AUGMENTATION | AI adaptive testing (Khan Academy, IXL) can identify knowledge gaps through standardised diagnostics. But a mid-level tutor's real-time diagnostic skill — hearing a student explain their thinking and pinpointing the exact misconception — remains human-led. AI handles structured assessment; the tutor interprets and adapts in context. |
| Deliver 1:1 or small-group tutoring sessions — explain concepts, guide practice, answer questions, scaffold understanding | 30% | 2 | 0.60 | AUGMENTATION | The core service. Khanmigo can explain concepts and guide through problems, but a human tutor reads confusion on a face, rephrases three different ways until understanding clicks, and manages the pacing of a live session. For in-person and complex learners, AI assists (generating examples on the fly) while the human leads. Online routine tutoring scores higher (3-4) but the mid-level average is 2. |
| Build rapport, motivate & provide emotional support — encourage struggling students, manage frustration, build confidence | 15% | 1 | 0.15 | NOT INVOLVED | A child who cries when they see fractions. A teenager paralysed by test anxiety. A parent desperate for someone who "gets" their kid. The motivational and emotional dimension of tutoring is irreducibly human — trust, patience, knowing when to push and when to back off. AI has no relationship to offer. |
| Create/curate practice materials, worksheets & assignments — generate problem sets, find resources, prepare session materials | 15% | 4 | 0.60 | DISPLACEMENT | AI generates practice problems, worksheets, flashcards, and study guides tailored to any topic and difficulty level. ChatGPT, MagicSchool.ai, and Eduaide.AI produce materials 5-10x faster than manual creation. The tutor selects and customises, but the production work is AI-executable. |
| Track progress, write session notes & communicate with parents/teachers — document student performance, report to stakeholders | 10% | 4 | 0.40 | DISPLACEMENT | AI generates session summaries, progress reports, and parent communications from structured data. Tutoring platforms (Wyzant, TutorMe) automate much of this tracking. The tutor reviews and personalises, but the documentation workflow is largely automatable. |
| Administrative tasks — scheduling, billing, invoicing, platform management, marketing | 10% | 5 | 0.50 | DISPLACEMENT | Fully automatable. Scheduling platforms, payment processors, and AI-generated marketing copy handle the business side. Independent tutors still manage logistics, but the tasks themselves are deterministic and rule-based. |
| Total | 100% | 2.85 |
Task Resistance Score: 6.00 - 2.85 = 3.15/5.0
Displacement/Augmentation split: 35% displacement, 50% augmentation, 15% not involved.
Reinstatement check (Acemoglu): Moderate reinstatement. AI creates new tasks for tutors: teaching students how to use AI tools effectively, validating AI-generated explanations for accuracy, coaching "learning to learn" skills that AI can't develop, and serving as the human quality-control layer when students use AI for homework. The tutor's role shifts from "source of knowledge" to "learning coach" — but this transformation favours senior tutors with coaching skills and disadvantages junior tutors whose value was primarily subject-matter delivery.
Evidence Score
| Dimension | Score (-2 to 2) | Evidence |
|---|---|---|
| Job Posting Trends | -1 | BLS projects 6% growth for tutors (SOC 25-3041) 2022-2032 — average across all occupations. But aggregate data masks the split: specialised test-prep and special-needs tutoring postings remain strong while routine homework-help postings face downward pressure from AI alternatives. The 215,500 employed figure is stable but growth is decelerating as AI tutoring platforms absorb the routine segment. |
| Company Actions | -1 | Chegg — the largest online homework-help platform — saw its stock collapse 99% after ChatGPT launched, the single most dramatic AI casualty in the education sector. Varsity Tutors pivoting to AI-integrated models. Course Hero acquired and restructured. Tutoring centres (Kumon, Sylvan, Mathnasium) continue operating but are integrating AI into their delivery models. No mass layoffs of human tutors, but the market structure is shifting. |
| Wage Trends | -1 | BLS median $36,290/yr ($17.45/hr) — well below the national median for occupations requiring post-secondary education. Wages stagnant in real terms. Independent premium tutors (test prep, STEM) command $50-100+/hr, but the median reflects a large population of modest earners. AI-driven downward pressure on routine tutoring rates. |
| AI Tool Maturity | -1 | Production AI tutoring tools directly compete with human tutors. Khanmigo (Khan Academy + GPT-4) provides Socratic tutoring across maths, science, and humanities. ChatGPT Study Mode offers personalised explanations and practice. Photomath solves maths problems with step-by-step explanations. Duolingo Max provides AI language tutoring. The global AI tutors market: $1.63B in 2024, projected $7.99B by 2030 (CAGR 30.5%). These tools handle 50-80% of routine tutoring tasks autonomously. |
| Expert Consensus | 0 | Mixed. The broad education consensus (WEF: 78% augmentation) applies to licensed classroom teachers, not unlicensed tutors. Sal Khan explicitly predicts AI will provide "a virtual personalized tutor for every student" — describing the displacement of the exact service human tutors sell. But Khan also says human tutors become more important as coaches and mentors. Academic research (2025) shows AI-tutored students achieving comparable outcomes in structured subjects. No strong consensus specifically for human tutors — the data points in both directions. |
| Total | -4 |
Barrier Assessment
Reframed question: What prevents AI execution even when programmatically possible?
| Barrier | Score (0-2) | Rationale |
|---|---|---|
| Regulatory/Licensing | 0 | No licensing required to be a tutor. Anyone with subject knowledge can tutor. No regulatory framework mandates human instruction in private tutoring. Background checks may be required for working with minors through companies, but no professional licensing barrier exists. |
| Physical Presence | 1 | In-person tutoring — sitting with a student at a table, pointing at their worksheet, using physical manipulatives — is a meaningful segment. Parents of young children often prefer in-person. But online tutoring is equally established and growing. The occupation spans both modalities, and the online segment faces no physical barrier to AI delivery. |
| Union/Collective Bargaining | 0 | Tutors are overwhelmingly independent contractors, part-time workers, or platform gig workers. No collective bargaining agreements. No union representation. Among the least-protected employment structures in education. |
| Liability/Accountability | 1 | Tutors work with minors. Parents entrust their children to tutors — there's an implicit duty of care and a trust relationship. Companies require background checks. If a tutor's advice damages a student's academic trajectory (e.g., poor test-prep strategy for a college admissions exam), there's reputational but not legal liability. The stakes are moderate — above zero but below licensed professional accountability. |
| Cultural/Trust | 1 | Parents want a trusted human working with their child, especially for struggling or anxious learners. The idea of replacing a child's tutor with an AI chatbot provokes cultural resistance among many families. But this barrier is age-dependent and eroding: university students and self-directed adults already prefer AI tutoring for convenience and cost. Cultural trust is strongest for young children and weakest for adult learners. |
| Total | 3/10 |
AI Growth Correlation Check
Confirmed -1 (Weak Negative). AI tutoring tools — Khanmigo, ChatGPT, Photomath, Duolingo Max — directly substitute for human tutors in routine academic subjects. The global AI tutors market growing at 30% CAGR represents demand shifting from human to AI delivery. However, the correlation is -1 not -2 because: (a) specialised tutoring (complex learning needs, test prep strategy, executive function coaching) retains human demand regardless of AI; (b) AI tools may expand the total tutoring market by making it accessible to families who couldn't afford human tutors, creating a larger ecosystem where human tutors serve the premium segment; (c) the "learning coach" reinstatement effect creates new human tasks around AI-assisted learning.
JobZone Composite Score (AIJRI)
| Input | Value |
|---|---|
| Task Resistance Score | 3.15/5.0 |
| Evidence Modifier | 1.0 + (-4 × 0.04) = 0.84 |
| Barrier Modifier | 1.0 + (3 × 0.02) = 1.06 |
| Growth Modifier | 1.0 + (-1 × 0.05) = 0.95 |
Raw: 3.15 × 0.84 × 1.06 × 0.95 = 2.6645
JobZone Score: (2.6645 - 0.54) / 7.93 × 100 = 26.8/100
Zone: YELLOW (Green ≥48, Yellow 25-47, Red <25)
Sub-Label Determination
| Metric | Value |
|---|---|
| % of task time scoring 3+ | 55% |
| AI Growth Correlation | -1 |
| Sub-label | Yellow (Urgent) — AIJRI 25-47 AND ≥40% task time scores 3+ |
Assessor override: None — formula score accepted. The 26.8 sits 1.8 points above the Red boundary. This proximity is honest, not an error — tutoring is one of the most AI-exposed roles in education precisely because it lacks the licensing, institutional employment, and safeguarding mandates that protect classroom teachers. The score calibrates correctly against Self-Enrichment Teacher (32.4, similar task resistance but weaker AI competition) and Substitute Teacher (50.2, similar interpersonal work but institutional employment and licensing).
Assessor Commentary
Score vs Reality Check
The 26.8 AIJRI score places tutors near the bottom of Yellow Urgent, 1.8 points above Red. The label is honest but requires context: this is a bimodal role. The aggregate score hides a sharp split between online tutors delivering routine homework help (effectively Red Zone — directly displaced by AI tutoring tools) and in-person tutors working with struggling students, complex learning needs, or high-stakes test prep (solidly Yellow, approaching Green for the most specialised). The barriers (3/10) are doing meaningful work — the cultural trust of parents placing their child with a human tutor is real but eroding as AI tools improve and younger parents become more AI-comfortable.
What the Numbers Don't Capture
- Bimodal distribution is extreme. A Chegg tutor answering algebra questions online faces near-total displacement — Khanmigo does this better, for free. A tutor sitting with a dyslexic 10-year-old, using multi-sensory techniques and building a trust relationship over months, is in a fundamentally different occupation. The same SOC code covers both, and the 26.8 average is truthful for neither.
- The Chegg collapse is the canary. Chegg's 99% stock decline is the education sector's most visible AI casualty and directly represents the online homework-help segment of tutoring. The human tutoring market hasn't collapsed yet, but the business model that connected students with remote tutors for routine academic questions is being hollowed out.
- Platform economics amplify displacement. AI tutoring costs $0-20/month vs. $30-100+/hour for a human tutor. For families making cost-benefit decisions about routine academic support, AI wins on pure economics. The human tutor's protection comes from situations where the relationship, the in-person presence, or the diagnostic complexity justifies the premium.
- Market growth vs headcount growth. The total tutoring market (human + AI) is growing — more students than ever receive some form of tutoring support. But the human share of that market is shrinking as AI captures the routine segment. Revenue may grow while headcount stagnates or declines.
Who Should Worry (and Who Shouldn't)
In-person tutors working with complex learners — students with learning disabilities, test anxiety, ADHD, or multi-year academic gaps — are safer than this score suggests. Their work requires diagnostic judgment, adaptive patience, and a trust relationship that AI cannot replicate. Parents paying $80/hour for a tutor who "gets" their child are buying a relationship, not an information service. Online tutors delivering routine homework help in standard academic subjects should be most concerned. This is exactly what Khanmigo, ChatGPT, and Photomath do — explain concepts, generate practice problems, provide step-by-step solutions — at a fraction of the cost with 24/7 availability. The single biggest factor separating the safe version from the at-risk version: whether the student needs a human relationship or just an explanation. If they need someone to believe in them, read their frustration, and adapt in real time — you're protected. If they need someone to explain quadratic equations — AI already does that.
What This Means
The role in 2028: The surviving mid-level tutor is a learning coach, not a knowledge dispenser. They use AI tools to generate practice materials, track progress, and handle administrative tasks, then invest their time in what AI cannot do: building trust with students, diagnosing complex learning barriers, providing emotional support, and coaching metacognitive skills. The "explain this concept" request goes to AI; the "help my child believe they can succeed" request goes to the human tutor. Hourly rates polarise — routine tutoring collapses toward AI pricing while specialised, in-person, relationship-based tutoring commands a premium.
Survival strategy:
- Specialise in what AI can't do. Complex learning needs (dyslexia, ADHD, executive function), high-stakes test strategy (not drill — strategy), and in-person coaching for anxious or struggling students. The generalist homework-help tutor is the most exposed version of this role.
- Master AI tutoring tools and integrate them. Use Khanmigo, ChatGPT, and adaptive platforms as force multipliers — let AI handle concept explanation and drill while you focus on diagnosis, motivation, and personalised strategy. Tutors who fight AI lose to it; tutors who wield AI become more valuable.
- Build a client base through relationships, not platforms. Platform-based tutoring (Wyzant, TutorMe, Chegg) is most vulnerable to AI substitution because the platform mediates the relationship. Direct client relationships, parent referrals, and school partnerships create a moat that AI platforms can't easily replicate.
Where to look next. If you're considering a career shift, these Green Zone roles share transferable skills with tutoring:
- Teaching Assistant / Paraprofessional (Mid) (AIJRI 51.2) — your diagnostic skills, student engagement, and one-on-one instruction transfer directly; institutional employment provides stability that independent tutoring lacks
- Elementary School Teacher (Mid-Career) (AIJRI 70.0) — your pedagogical instincts and student rapport are the foundation; requires licensure but tutoring experience accelerates teacher preparation programmes
- Special Education Teacher, K-Elementary (Mid-Level) (AIJRI 75.1) — for tutors already specialising in learning differences; requires certification but the diagnostic and adaptive instruction skills are directly transferable
Browse all scored roles at jobzonerisk.com to find the right fit for your skills and interests.
Timeline: 3-5 years for significant structural change. The AI tutoring market's 30% CAGR means the competitive landscape doubles roughly every 2.5 years. Online and routine tutoring segments face faster displacement (2-3 years); in-person and specialised segments face slower transformation (5-7 years). The Chegg collapse (2023) marks the beginning, not the end, of the transition.
