Will AI Steal My Job? · Role analysis

Secondary Teacher

O*NET 25-2031.00 ESCO: Secondary school teachers

Changing

Secondary teachers teach subject specialisms to pupils aged 11–18, preparing them for GCSE, A-Level, and vocational qualifications. Beyond knowledge delivery, the role involves examination preparation, pastoral tutoring, marking, behaviour management, and being a subject expert who can inspire curiosity in their discipline.

Section 01

Task Map

Task	AI impact	Why
Plan and deliver subject lessons	🟡 Changing	AI can generate lesson resources, explanations, and practice questions rapidly. But teaching — the live human dialogue that builds understanding — is not the same as content delivery.
Mark essays, assignments, and exam papers	🟡 Changing	AI marking tools are improving rapidly (Turnitin, GCSE essay markers) but struggle with nuanced evaluation of argument quality. Summative exam marking oversight remains human.
Prepare pupils for public examinations	🟡 Changing	AI tutoring tools (Khan Academy, Khanmigo) can deliver excellent exam prep content, but the teacher's role in maintaining motivation through a stressful exam cycle is irreplaceable.
Write predicted grades and university references	🟡 Changing	AI can draft references from pupil data, but personalised, credible references that universities trust come from teachers who genuinely know their students.
Manage classroom behaviour	🟢 Safe	Adolescent behaviour management — de-escalating confrontation, building authority with challenging groups, maintaining a culture of respect — is a live, relational skill.
Provide subject-specialist tutoring and intervention	🟡 Changing	AI tutoring can provide excellent one-to-one practice, but diagnosing deep misconceptions and adapting explanation to a specific student's thinking requires expert human judgment.
Contribute to curriculum planning and schemes of work	🟡 Changing	AI can generate schemes of work, but decisions about sequencing, emphasis, and local context require professional teacher judgment and departmental discussion.
Input and analyse pupil data in tracking systems	🔴 High exposure	Data entry and basic analysis in MIS systems (SIMS, Bromcom) is being increasingly automated. Progress tracking dashboards generate much of this view automatically.

Section 02

What Stays Human

Inspiring subject passion The teacher who makes a student fall in love with history, mathematics, or literature is doing something that changes a life trajectory. That spark comes from a human being who cares deeply about their subject sharing that enthusiasm — it cannot be replicated by a chatbot, however knowledgeable.
Pastoral care through adolescence Secondary students navigate identity, mental health, family breakdown, and peer pressure. The teacher who spots a struggling student, creates a moment of connection, or makes a safeguarding referral is doing protective work that no algorithm can substitute.
Academic integrity and examination credibility As AI-generated work proliferates, the teacher's role in assessing genuine understanding — through discussion, oral questioning, and observed performance — becomes more important, not less. Live assessment of what a student actually knows cannot be gamed.

Section 03

What to Do Next

Develop a clear, informed position on AI in your subject. Students are using AI tools whether you address it or not. Teachers who can teach students to use AI critically and ethically — and who design assessments that still test genuine understanding — are doing the most future-relevant work in schools right now.
Use AI to cut your workload, not your quality. Lesson resource generation, differentiation, and first-draft feedback are all places where AI tools genuinely save time. Reclaim those hours for the relationship-based teaching that only you can deliver.
Move into leadership, curriculum design, or specialist roles (SENCO, pastoral lead, exams officer). The more your role combines professional judgment, accountability, and human relationship, the more protected it is from automation pressure — and the better it tends to pay.

Sources: O*NET Online (onetonline.org) · ESCO (esco.ec.europa.eu) · All task data cross-referenced against O*NET occupation profiles. This analysis uses task-level exposure, not occupation-level prediction.