How IELTS Teachers Are Actually Using AI in Their Classrooms

If you teach IELTS, you've probably had the same week I used to have. 

The marking pile grows on Friday. The strongest students want more challenge. The weakest want more attention. The mock speaking tests you'd like to do never quite fit in.

AI doesn't solve that on its own. But it can make the job of teaching IELTS easier in some specific ways. 

There are now hundreds of teachers using Keenu to help teach IELTS - here are some of the things that we’ve learned about how they use it. 

What AI is good at (that teachers find awkward)

Giving a band score 

Most teachers I talk to aren't comfortable putting a number on a student's IELTS speaking or writing. It's a hard judgement call, and certified IELTS examiners aren't actually allowed to give band scores outside the real exam. So students often end up in a situation where the people best placed to give them an IELTS score can't do so. AI like Keenu gives a score against the official descriptors every time, which takes the awkwardness out of the conversation.

Telling a student they haven't improved

Teachers care about their students, and there's a natural pull to find the positive. That instinct makes you a good teacher, but it doesn't always help the student. If their score hasn't moved in three weeks, AI will say so. The student can see it, accept it, and act on it.

Catching every error

A teacher listening to a two-minute Part 2 answer can pick up the headline issues, but no one can isolate and log every small mistake in real time. AI can. Every error gets flagged, categorised, and logged, which means students can self-diagnose between lessons and come to class ready to work on the right things.

Scoring without favouritism

Teachers are human, and there's a natural pull to score the students you like, or the ones who try hardest, a bit higher than the work alone justifies. It's not deliberate, but it happens. AI doesn't know which student is which. Two answers at the same level get the same score, whether they came from your star pupil or the one who never does the homework.

The pattern we see in schools using Keenu is that students improve faster with AI and a human teacher than with either alone. The AI does the scoring, error-logging, and repetition. The teacher does the teaching.

How IELTS teachers are using it

As homework between lessons

Students do a full speaking or writing IELTS mock at home. The teacher reviews the scores before the next class and walks in knowing what to focus on. The lesson becomes targeted: linking words in Part 3 for this group, Task 2 introductions for that one. The mock itself isn't taking up class time anymore.

As an in-class activity, especially for writing

Brainstorm and plan together as a group, then set students off on the essay — 20 minutes for Task 1, 40 for Task 2. AI marks them the moment they finish. The lesson can move straight into reviewing what came up, while it's still fresh.

As teaching material in class

Student answers (anonymised) make for some of the best teaching material you'll find. The mistakes are real, the level is right, and the class recognises the patterns because they've made them too. A few common errors pulled from the week's mocks can structure an entire lesson on grammar, vocabulary or structure.

For end-of-week progress tracking

Weekly mocks build a picture over time. Students see the line moving (or not), which is motivating in itself. Teachers spot the ones who are stalling before it becomes a problem.

The common thread is that AI takes over the parts of the job that don't need a human, so the human parts get more time.

Try it with your class

The easiest way to see how this works is to try it yourself. You can take a mock test as a student would, see the scoring in action, and have a look at what the teacher dashboard shows.

There's a 14-day free trial for schools, no card needed.

If you'd rather have a quick call first, drop me a line at francis@keenu.io and we'll set something up.