We all seem to have come to terms with the existence of AI in the world and, especially, in the lives of our students, and we’re finding more and more ways to make the most out of it in our field: language teaching and learning. By ‘coming to terms with it’ I mean that we’ve accepted the fact that students are drawn to using it to alleviate the ‘burden’ of homework, research and studying, and so it seems like the goal now is to find ways to use it in the classroom as a tool, just like the internet or a calculator, and not a cheating device.
Still, there is a lot to keep in mind when planning activities where it is the students who will be using this technology. This is why the aim of this article is to share some practical ideas to continue learning how to maximise its use in the classroom, while also highlighting potential issues that may arise without us even noticing.
According to the British Council’s report on artificial intelligence and English language teaching (2024), teachers are already using AI to create materials, design lesson plans, assess students’ work and handle administrative tasks. Now, let’s explore how we can effectively use AI directly with our students. Here are some ideas.
AI-assisted writing
Students can use AI to generate ideas or suggestions for writing tasks (e.g. essays, reports) or organise their ideas before they begin writing. They will have to do the actual writing and develop those ideas themselves, AI is just a tool that will help them save time and inspire them.
Example task: Write a 300-word essay on climate change. Use AI to generate ideas, then write it in your own words.
AI-facilitated research
Encourage students to ask AI to summarise topics or provide explanations. Then, they must cross-check the information with other sources. Students will learn the content of the lesson and develop critical thinking and fact-checking skills.
Example task: Research the life of Shakespeare. Use AI to get information on his life and then verify it using two other sources.
Personalised vocabulary practice
Students can use AI to generate quizzes or exercises on vocabulary they are learning in class and focus on words they find challenging. This will enhance vocabulary retention and show them they can do extra practice on certain topics when they feel they need it, even if their classmates don’t.
Example task: Create an AI-generated vocabulary quiz on adjectives with –ed and –ing and complete it using your notes from our lessons.
Interactive speaking practice
Students can use AI to simulate (written or oral) conversations on assigned topics. This can help them improve their English by engaging in meaningful, ‘real-world’ practice.
Example task: Have a conversation about school life with ChatGPT. Ask the chatbot to correct you when you make language mistakes.
AI for task breakdown
AI can assist students in breaking down complex assignments into smaller manageable tasks with clear steps, which can help them develop better organisational and time-management skills.
Example task: Use AI to divide a large project (e.g. research paper) into manageable tasks with deadlines.
AI-assisted reading comprehension
Students can use AI to generate questions on assigned readings. This will help them focus on key ideas and improve their reading comprehension skills.
Example task: After reading a short story, use AI to generate comprehension questions. Answer the questions yourself.
Prompting
In order for students to be able to use AI in these ways, it is key for them to learn how to write prompts effectively. I know that because many of our students are younger than us and most of them were born with a device in their hands, we’re drawn to believe that they will be capable of writing good prompts. But let’s not forget that AI is new to everyone, and just like many of us probably remember receiving texts with numbers on top of words, showing that students used Wikipedia so poorly they didn’t even realise they had to delete those numbers for us to actually think they’d written those texts themselves, many students have been submitting writing assignments with the ending phrase ‘regenerate response’. This shows that although many students are tech-savvy when it comes to social media and video games, they don’t necessarily know how to use AI chatbots correctly.
To get the best results, it’s important to be clear and specific with our prompts, just like when students write essays, letters or reviews in class. Having said that, why don’t we structure our prompting lessons similarly to our writing lessons? For example, we always tell them to be straightforward (if they write a letter to a friend, they should tell them why they’re writing to them), the same thing happens with AI: we have to explicitly state what we want the chatbot to do. Just like we ask our students to avoid using vague language when they write proposals, for example, we have to remind them that the more specific their prompts are, the more accurate and tailored the AI’s response will be. And just like when we ask students to provide their readers some context, because the people reading their essays are not inside their heads, they should provide generative AI with all the necessary information for it to be able to provide them with the right information and in the right format.
Tutor prompt: training students to become autonomous
Once students get the hang of prompting, they will be able to work on tasks like the ones above or come up with their own ones whenever they want to learn or practise something, even if no teacher asks them to do so. They can do this by using tutor prompts (Figure 1), a specific kind of instruction that guides an AI to provide explanations, exercises or feedback, tailored to the student’s needs. Essentially, it turns the AI into a personalised, on-demand tutor for the user. By following the prompting steps above, students can turn a chatbot into their own private tutor who’s available to help them around the clock. And although they will still come to class to interact with human beings – the reason they’re learning a language in the first place – they now have a very efficient tool to help them outdo themselves at home.
Figure 1: Example comparing a simple prompt to a tutor prompt for language learning
For a tutor prompt to work, students need to assign the chatbot a framing role, for example, students might ask the AI to act like an English tutor and help with grammar or vocabulary. They then have to specify a task or learning objective, for example, they can request an explanation of a grammatical rule, help with sentence construction or even feedback on a written text. As a result, the AI, following the prompt, will engage the student interactively, explaining concepts, generating examples or giving feedback.
I believe tutor prompts are a big part of the future of language learning. Think about how it can revolutionise instruction in places where teachers work with groups of 30, 40 or more students by providing them with personalised explanations (e.g. if a student doesn’t understand a concept, they can ask the AI to re-explain or break it down further) and immediate feedback on their language tasks. It can also enable a dialogue-based learning experience (talk about interactive learning!) and allow students to learn at their own pace, by continuing to provide additional exercises or explanations, for example, until students are really able to master a skill or structure.
Where do we go from here?
I would like to finish this article with a simple idea: instead of trying to beat students and look for ways to make AI-proof assignments, it is more effective to show them why using AI as more than just a tool can be detrimental for their learning journey. Ethan Mollick (2024) summarised the effects of the careless use of AI with the term ‘illusory knowledge’, the feeling that AI is actually helping students to learn and outdo themselves when doing their homework because of how fast and simple the process is, when it (the AI chatbot solving problems and completing exercises for them and not with them) actually undermines their learning because students are not doing any thinking. I firmly believe that encouraging students to use AI as a learning partner, not a shortcut, and equipping them with effective prompting skills and clear goals, will lead to a more rewarding, autonomous learning experience.
