Skip to content
AI, TECHNOLOGY & INNOVATIONTwo editions. One publication. / About us
News

The Teacher's New Job: Why AI Makes Great Teachers More Important, Not Less

AI can draft a lesson plan in seconds. It cannot notice the quiet child in the third row. The teachers who thrive will be the ones who let machines do the paperwork and keep the teaching for themselves.

The Teacher's New Job: Why AI Makes Great Teachers More Important, Not Less

The 2 a.m. lesson plan

It is late, and a teacher is still at her desk, preparing tomorrow's lesson. She gives an AI tool the topic, and within seconds she has a lesson plan, a worksheet, and quiz questions. She does not hand them over as they are. She reads them, cuts them, changes them, adapts them to her students, and adds her own examples. This is the new workflow, and it is quietly spreading through classrooms everywhere.

The numbers confirm it. In a Gallup and Walton Family Foundation survey of 2,232 US school teachers, six in ten had used AI that year, and about 32 percent used it weekly. The most common uses were preparation: 37 percent used AI at least monthly to prepare to teach, 33 percent to make worksheets, 28 percent to modify materials for student needs, 28 percent for admin work, and only 16 percent for grading. In England, just 7 percent used generative AI to deliver a live lesson and 5 percent for marking. The pattern is clear: behind-the-scenes work first, live teaching last. Teachers are keeping the classroom human.

The hours question, answered honestly

Weekly users self-reported saving an average of 5.9 hours a week, roughly six weeks over a school year. But self-reported is not measured, and the distinction matters. In an Education Endowment Foundation randomised trial with 259 science teachers in 68 schools in England, teachers using ChatGPT spent 56.2 minutes a week on lesson and resource preparation versus 81.5 minutes without it, a 31 percent saving, and a blinded expert panel found no drop in quality.

A second trial, of Oak National Academy's Aila tool with 464 teachers across 108 schools in autumn 2025, saved about 49 minutes a week, 2.51 hours versus 3.32 hours, again with no worse quality. But the tool was used for only about 7 percent of lessons in an average week, usage declined over the term, and fewer than one in five teachers said it meaningfully changed how they taught. The honest read: real time saved, not a revolution. And those hours matter because they buy the one thing teachers lack most: time with students.

What AI cannot do in a classroom

Notice the quiet child. Lift the discouraged student back up. Teach values by example. Build trust. Handle a sensitive disclosure. Read a room. These are the teacher's work, and they will stay that way. A machine can deliver content. A teacher forms people. That was always the real job; delivering content was just its visible part.

This is why live lesson delivery has the lowest AI use of all. Teachers know that a lesson is not only information. A lesson is a relationship: between teacher and student, and among students themselves. A machine cannot sit inside that relationship.

The guidance vacuum

The problem is that teachers have mostly been left alone with this new role. UNESCO's 2023 global survey of more than 450 schools and universities found that fewer than 10 percent had institutional policies or formal guidance on generative AI. UNESCO stressed the teacher's role as facilitator and said teachers need guidance and training. Three years later, many institutions still leave the rules to individual teachers. The result is that one teacher integrates AI into lessons while the next one bans it, and students sit confused between the two.

The Pakistan angle

In Pakistan, teachers sit at the centre of this story for a reason that is often forgotten: someone has to deliver the national targets. The World Bank's October 2026 report records a target of training 200,000 people in AI every year, alongside 1,000 PhD scholarships and training for one million non-IT professionals. That training will happen in classrooms, and it will be delivered by Pakistan's teachers.

The second issue is language. Urdu accounts for just 0.03 percent of the web addresses in Common Crawl, one of the largest datasets used to train AI models, against 40.6 percent for English. So a teacher working in Urdu finds less ready-made Urdu material. But here AI can also help her: a first draft of a worksheet, a quiz, or a letter to parents in Urdu, which she checks and adapts to her students. In a low-resource classroom, where books are few and time is short, a teacher with a phone and a printer can now produce differentiated materials she never had time to make before.

How a teacher uses it well

The workflow has two stages: curriculum grounding and the teacher edit. Take a draft from AI, then work on it like an editor: check the facts, align it to your syllabus, adapt it to your students' level, add your examples. Never hand an AI draft straight to students. Keep grading and feedback for yourself; only about 16 percent of teachers use AI for grading, and that restraint is wise, because giving feedback is itself teaching. Give the machine: first drafts, variations for different levels, quizzes, parent letters, admin work. Keep for yourself: live teaching, feedback, mentoring, judgment.

The future of schooling

Pedagogy is shifting from delivering content to coaching thinking. Less "teacher tells, students copy," more "students attempt, teacher guides, AI assists." Schools that teach AI literacy — how to use AI, how to check it, how to question it — will produce students who can use it well. Schools that only ban it will produce students who use it secretly and badly. That was the lesson of the previous article in this series, and it is this one's too.

The last word

The teacher's job was never really information delivery. That was only the part machines could copy. The real job was always formation: shaping minds, character, and confidence. AI, used well, gives teachers their time back for that job. The teachers who thrive will not be the ones who use AI most. They will be the ones who know exactly what to hand to the machine and what to keep for themselves.

Sources & attribution

Gallup and Walton Family Foundation, "Teaching for Tomorrow" survey (RAND American Teacher Panel, 2,232 teachers, March–April 2025); Education Endowment Foundation and NFER, Aila trial (autumn 2025); UNESCO global survey (2023); World Bank, MENAAP Economic Update, October 2026

View original source
About the author

Watan Zaib Mehsood

Watan Zaib Mehsood is a journalist from South Waziristan who reports on the issues of the former FATA region and Khyber Pakhtunkhwa. He is the founder of two digital publications: The Journal Dev (thejournaldev.com), a bilingual site covering artificial intelligence, technology and innovation in English and Urdu, and PrimeUrdu (primeurdu.com), an Urdu news website. He also writes about how emerging technology is reshaping work, education and everyday life in Pakistan, and builds digital learning tools, including Medricks, an app for medical students.