ChatGPT in your classroom?

Possibility seems endless if implemented properly.

Image generated by DALL-E 2, an artificial intelligence. My prompt was: “oil painting that depicts students using computers in a classroom”

I am sure you have heard so much about ChatGPT by now. According to the University World News, educators’ reactions in Asia seem somewhat cautious so far.

Many teachers are understandably concerned about cheating and/or the unfair advantages some teach-savvy students can gain.

For example, the University of Hong Kong, where I teach, has temporarily prohibited students from using generative AI for coursework and assessment tasks this semester. But this policy has met with some strong pushback.

The university was criticised for becoming the world’s '“first institute of higher education to ban” the technology altogether (instead of embracing it as a pedagogical tool). Meanwhile, in Singapore, the education authority is reportedly encouraging schools to incorporate AI tools in class.

But even though ChatGPT is off-limits to the students at our university, instructors are still free to experiment and use it in our teaching. So, I have been intensively tinkering with it for a few weeks to see how this new tool can help.

Before I share my experience, though, I want to know where you stand.

Please take a few moments to join the super short survey below, which shouldn’t take more than a couple of seconds.

So many ways to use it in our teaching

After playing with ChatGPT and other generative AI tools, I see a huge potential for this technology to be part of our pedagogical methods.

I am convinced that it will become essential in education. In fact, I now have a myriad of ideas about how we can take advantage of it.

Some ideas can be implemented right away without any knowledge or preparation; others require some planning and advanced skills.

But if I try to share my initial thoughts and write everything down, this newsletter will be way too long.

So, as a first of a series of many newsletters to come on generative AI in the future, today I would like to share two possible use cases anyone can adopt immediately.

Warning! TL;DR

Skip to the bottom for key takeaways if you’re in a hurry.

1. Summarise key ideas from a number of small group discussions

Small group discussions are one of the main staples in my inquiry-based courses. I ask a lot of questions for the learners to discuss in small groups.

I use a Google Docs document as a common whiteboard so that each group can freely type what they talked about and also read what the other groups discussed.

The problem with this method is that it’s not always easy to go through what each group writes on the fly and quickly identify emerging themes and new ideas among them, especially when there are many small groups.

One of the common practices in such a setup is to have a person from each group speak up and summarise the discussion. But in my view, that is not an effective way to engage the learners because it is hard to find common threads, agreements, and disagreements across different groups that way.

ChatGPT can solve this problem rather efficiently.

All I need is to copy and paste everything from the Google Docs document and ask the AI to identify what I am looking for — key arguments, main discussion points, opposing views, and so on and so forth.

I can instruct it to give us a bullet-point list, one paragraph summary, or a few hundred words.

Of course, we cannot rely on the summary completely. Once in a while, the AI does miss some important concepts or churns out something a bit off, but that’s where instructors come in.

You can remedy it by using different prompts or narrowing down the focus. As a bonus, since this technology is still a novelty, having an AI summarise the discussions like live broadcasting on a big screen is an easy way to retain learners’ attention and keep them engaged.

2. Create course FAQs and a glossary

I think this is a common experience among us. Even though we try to explain everything clearly, each semester, we get similar questions about foundational concepts, such as “what is critical thinking in the first place?” from our learners again and again.

Sometimes it is disheartening because the answers are what we just go over in class or part of the required reading (and anybody who reads it should have understood it, right?).

Many of us also suffer from the curse of knowledge. We may not quite get why the learners are struggling, or what missing background knowledge is creating difficulties.

It would be nice to have a readily available set of frequently asked questions and glossaries we can distribute, but we all know that it would be really time-consuming to make such material.

ChatGPT can make this process easier and quicker. We can mobilise the learners, ask them to chat with the AI, and send the conversations your way. You can then synthesize them using the same technology — summarising all the questions, answers, and word definitions, which forms the first draft of the material.

For example, here’s an example of a chat.

What is the definition of critical thinking in the context of media literacy?

Critical thinking is a process of actively and systematically analysing information, ideas, and evidence to make informed decisions, solve problems, and form judgements. It involves the ability to objectively evaluate information, consider different perspectives, and apply logical reasoning to reach sound conclusions. Critical thinking also involves questioning assumptions, recognising biases, and being open to alternative viewpoints. In the context of media literacy education, critical thinking skills are essential for evaluating the credibility and accuracy of information and media sources, and for identifying and analysing bias and propaganda.

What about media messages? My teacher also mentioned that.

Media messages can take various forms, including news reports, advertisements, documentaries, entertainment programmes, and online content. These messages can be created by individuals, organisations, or institutions, and may be intended to inform, persuade, entertain, or influence the audience.

In media literacy education, it is important to develop critical thinking skills for evaluating and analysing media messages. This includes being able to identify the purpose and intention behind media messages, recognising the different techniques and strategies used to convey information or influence the audience, and being aware of the potential biases or limitations of the media source. By developing these skills, individuals can become more discerning consumers of media and make informed decisions about the information they consume and the messages they engage with.

Wait. How can we possibly identify “the purpose and intention” without asking the authors directly?

Identifying the purpose and intention of a media message can be challenging, as it may not always be explicitly stated by the author or creator. However, there are several strategies that can be used to infer or interpret the intended purpose or message:

1. Context: Consider the broader context of the media message, including the medium or platform, the timing and location of its release, and the intended audience. This can provide clues as to the purpose and intention of the message.

2. Tone and language: Analyse the tone and language used in the message. For example, is it persuasive, informative, or satirical? Does it use emotional appeals or logical arguments? The tone and language can reveal the intended effect on the audience and provide insights into the purpose of the message.

3. Visuals and symbolism: Examine the visuals and symbolism used in the message. Images, colours, and symbols can convey meaning and help to reinforce the intended message or purpose.

4. Source and authorship: Consider the source of the message and the author or creator. Who is the intended audience? What is their background or perspective? What is the author's reputation or track record?

By considering these factors and asking critical questions, it is possible to develop a more nuanced understanding of the purpose and intention of a media message, even if it is not explicitly stated.

Not too bad, right? Conversations such as the above could form a great starting point for making FAQs and glossaries, which can also be made with the help of AI.

Of course, we need to be vigilant and peruse everything carefully before we finalise them — correcting errors, tweaking the language, and adding references along the way — but hopefully, you can see why I think generative AI can be tremendously helpful for creating FAQs and glossaries.

All right. This letter is way too long. I will stop here. But before I end it, I will ask ChatGPT to give us five key takeaways from the above and copy them below.

Key takeaways (from this newsletter) generated by ChatGPT

1. Educators in Asia have expressed concerns about cheating and unfair advantages that tech-savvy students might gain from AI tools.

2. ChatGPT has the potential for use in pedagogical methods and could become essential in education.

3. ChatGPT can be used to summarise key ideas from small group discussions in classes efficiently.

4. ChatGPT can also be used to create course FAQs and a glossary by synthesizing conversations with learners.

5. The use of ChatGPT in education requires careful consideration and should be used to supplement, not replace, instructors.