Teens and Generative AI

Generative AI, which can generate new and original content such as written texts, images, codes and music, has attracted a lot of people’s attention including English language teachers and learners, especially since the release of Chat GPT in late 2022. Nearly 80 percent of 13–17-year-olds have been using generative AI tools in the UK (Ofcom, 2023). You might notice that some of your teenage learners have started using different tools including Google Gemini (used to be Bard) and Microsoft Co-pilot in their writing. Perhaps it is time for you to step in and use AI tools along with your learners and help them use the tools effectively in a way to enhance their learning. In this blog post, we will look at some practical ways to incorporate generative AI tools in our language lessons. 

Age limit

Before we integrate the use of AI tools in our classroom, we should be aware that it is NOT recommended to use AI with learners under age 13 and if the learners are under 18, they must have the permission from the parent or the guardian to use the services. Also check with your school and country’s policy on further restrictions depending on where you are.

Getting started

The first thing we could do is to raise the learners’ awareness of capabilities and limitations of AI and using it in a responsible way. One resource I find useful is Lesson Plan: Teaching and Learning in the Era of ChatGPT by the New York Times. Your lessons could include the following stages: 

• Activate their schemata or tap into their existing knowledge through a warm-up discussion about Generative AI 

• Live demo of the tool(s)

• Show examples of the output and critique them

• Discuss the opportunities and challenges of using AI 

• Co-create a classroom contract on the acceptable and unacceptable use of AI (If your school has certain guidance and policy, you can make sure that learners are aware of them too.)

Prompt writing

One important aspect of creating suitable output using Generative AI is prompt writing. Prompts are the instructions or guidelines that we give to the AI to generate content. They can be as simple as ‘Write a story about two friends on an adventure’. The quality of the output depends a lot on the quality of the prompt which is why teaching learners how to write effective prompts can be a valuable skill. 

One model you can teach your students is the RTRI model which stands for Role, Task, Requirements and Instructions. 

Role – Act as a helpful grammar police

Task – Check my use of narrative tenses in writing a story

Requirements – Please tell me if they are correct or not. 

Instructions – First, read my story. Next, highlight the mistakes I have made. Then, correct the mistakes. Finally explain why I have made a mistake. 

Please note that there is no single formula or model of writing a prompt for AI but the RTRI model can serve as a useful guide. It provides a structured approach to prompt writing and encourages the user to clearly define the role of the AI, the task it needs to perform, the requirements of the task, and the instructions on how to carry it out. This can lead to more effective and targeted prompts. 

However, it’s important to remember that AI’s responses can be influenced by subtle changes in the prompt. Experimentation, refinements of the prompt and practice are key to mastering prompt writing which is why it’s beneficial to teach students about prompt writing and give them plenty of opportunities to practice and refine their skills.

To teach prompting, you can do a prompt writing practice activity in pairs or in groups. Before this activity, students should already be familiar with how Generative AI works. Here are some lesson stages you can do. 

• Introduce the context or the topic you want to focus on 

• Provide learners with a set of prompts and the corresponding outputs from the AI

• Ask them what they notice about the prompts and the outputs 

• Elicit what they think makes good prompts

• Let them notice a suggested model ‘RTRI’ or other models you have used

• Discuss the importance of providing clear instructions to AI tools and experimenting with the tool with several refinements

• Practice writing prompts in pairs and share them with different groups 

• Use the prompts with AI tools, critique the output and vote for the most effective ones

Ethical considerations

While Generative AI can be a powerful tool for language learning, it’s important to discuss the ethical considerations as well. For instance, learners should understand that while AI can generate creative content, it doesn’t have feelings or consciousness. It’s also important to discuss issues such as data privacy and the potential misuse of AI. 

Here is one activity which has been generated by Microsoft Copilot. 

“Activity: The AI Ethics Court

Objective: To understand the potential misuse of AI, develop argumentation skills, and practice using persuasive language.

Roles:

1. AI Developer: You have developed an AI system that can generate realistic images of people. You believe in the potential of your creation but also understand the risks.
2. Privacy Advocate: You are concerned about the privacy implications of the AI system. You believe that it could be misused to invade people’s privacy.
3. Government Regulator: You are responsible for ensuring that AI systems are used responsibly. You are considering whether to allow the use of this AI system.
4. Judge: You listen to all sides and make the final decision. You need to ensure that the decision is fair and considers all perspectives.

Language Learning Objectives:

• Use persuasive language to argue their point of view.
• Understand and use vocabulary related to AI and ethics.
• Practice listening skills by understanding others’ viewpoints.
• Develop speaking skills through debate and discussion.

Steps:

1. Preparation: Each role should research their position and prepare arguments for the discussion. They should also prepare a list of vocabulary words related to their role.
2. Court Session: The AI developer presents their AI system. The privacy advocate and government regulator present their concerns. Each role should use persuasive language to argue their point of view.
3. Debate: After each role has been presented, the judge facilitates a debate. Each role can respond to the others’ arguments and defend their own position.
4. Decision: The judge makes the final decision based on the arguments presented.
5. Reflection: After the role-play, students should reflect on the activity. They should discuss the vocabulary they learned, the arguments they found most persuasive, and how they felt about the final decision.

Remember, the goal of this activity is not just to teach students about the potential misuse of AI, but also to develop their language skills and critical thinking. It’s important to create a respectful and open-minded environment where all opinions are valued and considered.

I hope this helps! Let me know if you need more information. 😊”

Conclusion

Generative AI can be a very useful  tool and like any other tool, its effectiveness depends on how we use it and how our learners use it. In this blog post, we have looked at three practical activities we can do to incorporate Generative AI into our lessons. These are suggested activities which you might already have been doing for other topics. You might want to adapt these activities based on your learners, teaching context and your learning aims. Try them out in your class and let us know how they work for you. Share your experiences, adaptations, and results. Your feedback can help us all learn and grow together in this exciting field.

If you’d like to read more about the use of AI tools to enhance teaching and learning, click here.