Centaur ELT: Syllabus Design

In working with AI, Ethan Mollick defined two ways we can work with AI: Centaur and Cyborg. Both relationships keep the 'human in the loop' and mitigate the risks of hallucinations. 

Based on the title of this series, you can probably guess what type of relationship I have with AI. So, let's focus on the definition of a centaur like relationship with AI using the example of planning lessons in ESL.

If we are using a centaur-like model, the curriculum and lesson plan structure would be developed by the teacher. AI then supplements the learning experience with matching resources, tasks, or games. 

Discover how I incorporated AI into my syllabus update project. Keep reading to see who was in control - the human or the AI.

Centaur ELT: Syllabus Design

This series is an extension of my deep dive article, AI: The State of Play in Language Teaching in 2024.

If you want to find out more about the AI terms I use below including Large Language Models — and no I don't mean plus-sized models who like to study languages —then the ‘State of Play’ article is definitely for you.

In my research for the article above, Ethan Mollick advised you to spend at least ten hours using AI in your day-to-day tasks. He says:

Just use it and see where it takes you.

So, that's what I'm doing.

For the rest of the year, I'm immersing myself in the 'jagged frontier' of ELT.  I hope my observations help you figure out when AI can enhance your teaching and when you might want to steer clear of it.

Each article in the series will include:

1. Prompt crafting.
2. A live demo or screenshots of the prompt in action and,
3. My reflections on the experience.

My first two prompt experiments were about Wellbeing in ELT and Vocabulary. While in this post, I'm going to share with you how I used ChatGPT to help in designing lessons.

Syllabus Design Prompt

For this experiment, I moved away from a highly detailed prompt and instead engaged in a more iterative process, involving some back and forth in the conversation box.

Before we get into the prompt, I thought it would be helpful to give you some background on why the syllabus was updated at my school. The problem was that teachers struggled with incorporating storytelling into their lessons, so I decided to provide a more structured approach by offering a consistent format along with suggestions for storytelling, as well as follow-up game and craft activities.

You can see in the initial prompt with ChatGPT below, where I introduced the topic, provided a simple lesson outline, and had the AI read a story or fable that I uploaded.

From that point it would produce a rather generic type lesson plan. It gave me a starting point rather than a finishing point. 

Read on for real examples of the prompts I used and responses I received from ChatGPT. 

Prompt in Action

Reflections

1. The iterative discussion was my preference over the lengthy task of creating prompts with Centaur ELT Vocabulary, as it was more organic and less time-consuming.
2. This has saved me a lot of time, either by sparing me the effort of brainstorming related craft activities or searching the Internet for them, especially when I had to create 20 lessons following this structure. 
3. If I didn't like any of the suggest games or activities I'd just keep asking for five more until I got the an idea I could work with. 
4. On one occasion, it unexpectedly generated a Word document featuring a craft activity template during an interaction. Similarly, in another project, I uploaded an Excel spreadsheet, and it successfully detected duplicate entries, delivering an updated sheet afterward!!! 
5. ChatGPT is skilled at grading content and producing simplified versions of any input. 

‍