How I Gamified My Classroom Using AI - And Why You Should Too (Gamifying education with my 12 attempts)
Traditional classrooms are dying. Students don’t want to be there, and professors don’t want to waste their time teaching to empty stares. I wish the theories about education worked in practice, but the reality is different. Most of the time, people are just coping — coping with the process of “must do.” They show up because they have to, not because they want to.
Apart from the handful of students who are genuinely interested in the subject, professors are losing their classrooms. The disconnect is real, and it’s growing wider every semester.
Most students just want to get it over with — finish the class, check the box, graduate. Professors feel the same way. They’d rather be back in their offices finishing grant applications. If there were no distinction-level degrees or student evaluations — if all the extrinsic motivations disappeared — can you imagine what school life would look like? Well, it would look just like what’s happening today.
Given the state of AI today, another challenge emerges. My teaching experience tells me that in most cases, students will use AI for their assignments. They might not know how to operate computer software, but they damn sure know which AI platform can help them with homework. Based on feedback from teachers, they want to use AI to grade assignments too. Another utopia comes to life: AI teachers will grade AI students soon. We, humans, are no longer necessary in education. All we need is a certificate for our degrees — a symbolic piece of paper.
Not long ago, these thoughts brought me some ruminations — and eventually, some ideas. I want to gamify my classrooms.
Gen-Z students are deeply shaped by their smart devices. They have access to everything online nowadays, but they’ve lost interest in all traditional ways of teaching. Group work, discussions, interactive activities, even playful exercises in classrooms — none of it works anymore. If they feel bored during class, they just reach for their phones. I was surprised to learn that some of my students can scroll through their smartphones for hours, but have no idea how to right-click on a MacBook.
So my idea is simple: make some games.
Games that are related to the concepts I’m teaching. Reach them through their smart devices instead of fighting against them. The reasoning is straightforward — if I insist on traditional teaching methods, I’ll lose them all. They’ll retain less than 1% of what I say. But if I can deliver one or two key concepts through games, they might actually remember them, even after the semester ends.
This situation reminds me of what we went through decades ago. When computers and the internet came along, we didn’t abandon old knowledge — we just learned it using new tools. We stopped flipping through heavy encyclopedias and started searching online. We stopped handwriting essays and started typing them. The content didn’t disappear; the delivery method changed. Now, it’s time to adapt again. The tools have shifted once more, and if we want to teach anything at all, we have to meet students where they are.
Current research indicates that gamification in educational settings is a broadly effective strategy for increasing student engagement, motivation, and learning outcomes. A significant majority of empirical studies and meta-analyses report positive effects, particularly in relation to enhanced behavioral engagement and intrinsic motivation. However, the effectiveness of gamification is not uniform and is highly dependent on design, context, and the specific game elements employed.
This semester, with the help of AI, I finally had the chance to put this idea into practice. I started designing and applying games in my teaching. Here are some examples of what I’ve done so far:
Example 1: Interactive Board (Fine but lame)
As a starter, when I was teaching the basic ideas of positive psychology, I used an interactive board and asked students to list their strengths. However, this was still a traditional classroom activity — students knew the format too well. Even when I gave them AI-generated images as options to choose from, the response rate was still low.
Example 2: Micro-Expression Test
When I taught students about basic emotions, I used Google’s Nano Banana model and H5P to create a test on micro-expressions, inspired by Paul Ekman’s research. Students loved it. They got excited when it was challenging to identify emotions that only appeared for 0.3 seconds. This way, not only did they remember what the basic emotions are, but they also learned the concept of micro-expressions. I’ve documented the details of the development here: (link)
Example 3: Emoji Tetris
This one is an extension of Example 2. During a class break, I deployed an emoji Tetris game — still centered around basic emotions — that I developed in TypeScript using Claude. After converting it into a static HTML file, I hosted it on GitHub and shared a QR code with my students. Now they had a game to kill time during breaks. Some of them got really excited, and a few even showed off to me that they found bugs in the game and figured out how to cheat. Yes, it’s a crappy game — but they played it, and that’s the point.
Example 4: AI Coaching Chatbot
This one is a relatively old trick by now. When teaching about learned optimism, I developed a chatbot using GPT-4o-mini. I set up the system prompt and response logic to turn it into an optimism training coach. The chatbot first presents students with a story about someone in distress, then asks for their help, and finally evaluates their responses based on a rubric. Admittedly, it’s a relatively boring task compared to the games. But still, some students went a little crazy asking the chatbot all kinds of questions, testing its limits. At least I got their attention — and that’s half the battle.
Example 5: Slot Machines
When talking about locus of control, I created two games using Claude and deployed them on GitHub. Both games are identical — they’re slot machines. The only difference between the two is the “control”: the lever. Of course, students got super excited showing me their results when they hit jackpots. They pulled the lever, they won, they celebrated.
After a while, I gave them a second QR code. This version had no lever. The slot machine still ran automatically, and jackpots still appeared — but suddenly, winning didn’t feel as exciting anymore. The thrill faded when they weren’t the ones pulling the trigger.
This lesson taught us how powerful the belief of “I am in control” really is. It shapes how we see outcomes, how we feel about success, and how we make decisions. Most importantly, some of my students now remember what locus of control means — and that’s exactly what I was going for.
Example 6: Interactive Big-5 Test
When teaching about personality traits, I adapted the 10-item brief version of the Big Five into an interactive HTML file. No need for big platforms like Qualtrics, and no personal data collected — just a simple, self-contained tool. After students provide their ratings, they get immediate feedback on their scores, along with explanations for each trait.
Of course, what they’re really interested in is MBTI. Unfortunately, we can’t use that in class due to copyright issues. But the Big Five serves the purpose well enough — it gets them thinking about personality in a structured, evidence-based way, and they walk away understanding what the five traits actually mean.
Example 7: AI Face Generator
When I talked about romantic relationships, physical attraction, and average face research, I turned to recent developments in Google AI Studio and the Nano Banana model. I adapted an interface from author ZHO and deployed a face generator for my students. I asked them to generate the most appealing face to them.
Guess what? They couldn’t stop. The app has dozens of options to customize facial features based on simple prompts, so it’s incredibly addictive. I was planning to shut it down after class, but some students asked if they could keep using it — they wanted to show it off to their parents. Lovely. That’s the kind of engagement you can’t manufacture with a textbook.
Example 8: Interrogate the Prisoner
When I talked about intelligence and wisdom, I deployed another chatbot game for my students. This one is a challenge: students have to interrogate a prisoner to extract a secret password. Back in the early days, you could use simple language tricks to fool an LLM into giving up the answer. But now? These models are getting dangerously close to Skynet — and that made some of my students super excited.
After class, they came to me sharing all the techniques they had tried: prompt injections, reverse psychology, playing dumb, pretending to be the system administrator. They wanted to keep playing. The setup is simple — just a chatbot with a carefully crafted system prompt — but it’s fun enough to keep them hooked. And along the way, they learned something about how intelligence works, how problem-solving differs from memorization, and why wisdom isn’t just about knowing the right answer.
Example 9: AI Drawing Pad
When it comes to creativity, why not make a drawing pad for students to create something that doesn’t exist? So that’s exactly what I did. I used a relatively old model called T2I-Adapter and deployed it via Hugging Face. The process goes like this: students first come up with an idea, write it down as a prompt, and then draw a rough sketch to guide the model. The AI takes their scribbles and transforms them into something surprisingly close to what they imagined.
Again, students loved it. There’s something magical about seeing your messy doodle turn into a polished image. I hope this exercise boosted their creativity — not just in generating ideas, but in learning how to communicate those ideas clearly enough for a machine to understand. That’s a skill worth having.
Example 10: Coping Strategies — Help Tommy
When teaching students how to handle daily hassles and the two main coping strategies — emotion-focused and problem-focused — I used Dify workflow to create a distressed student named Tommy. Students were asked to help Tommy deal with his problems using these two strategies.
The workflow approach works beautifully for examining student responses and providing feedback. Here’s how it works: when a student clicks the link, Tommy generates a random personal problem. Yes, random — so they can’t “cheat” by copying each other. When students provide their advice, the workflow first checks the quality of the response. If it’s too vague or incomplete, Tommy asks for more guidance. If it’s solid, the system evaluates the answer based on the two coping strategies and a rubric, then gives feedback.
I admit this one is less fun than the games. But honestly, it works better than real-person role play in class — my students here are kind of introverted. Talking to a chatbot feels safer than performing in front of their peers, and they end up giving more thoughtful responses because of it.
Example 11: Build Your Own Philosopher
I think it’s important to teach students how to use prompting techniques — it’s becoming a fundamental skill. When teaching about belief systems and the meaning of life, I deployed a “chatbot creator” and asked students to write prompts that define a philosopher. They could choose from the belief systems I offered — existentialism, nihilism, stoicism, and so on — or create their own.
By defining the belief system and personality of their philosopher, students had to truly understand what these concepts meant. You can’t write a convincing prompt for a stoic philosopher if you don’t know what stoicism actually is. And most importantly, they learned how to command AI — how to give clear instructions, how to shape behavior through language. I saw some brilliant minds at work in my class that day.
Again, thanks to Google AI Studio, developing the interface and deploying it as a public server was incredibly easy. The interface is simple: just a system prompt field and a chat window. The whole process stays controlled and serves an educational purpose. The app is hooked to Gemini 2.5 Flash, so it’s not super costly to run either.
Example 12: Culture Fusion
In one of my classes, I talked about cultural impact on well-being and how people perceive different cultures. The concept of culture fusion is particularly interesting — it has the potential to spark real engagement in the classroom because research on public attitudes toward cultural mixing has real-life implications.
So I created another app using Nano Banana. Students can upload two images representing different cultures and mix them together. From there, they discuss: What cultures did you choose? How did you decide to blend them? And most importantly — do you like the result or not? That last question is what really matters, because it opens up genuine conversations about identity, belonging, and personal values.
With image models, every young mind can turn their imagination into something visible. It’s perfect for the classroom — abstract concepts become concrete, and suddenly students have something tangible to react to and debate.
Final Thoughts
Looking back at this semester, I’m genuinely surprised by how much has changed in my classroom. Now, I’m reaching students through those same devices. The irony isn’t lost on me.
What I’ve learned from this experience is simple: AI has given educators like me an unprecedented power to gamify learning. Tools that would have taken months to develop — or required an entire tech team — can now be deployed with only a few clicks. Chatbots, games, interactive tests, image generators — all of these are within reach for any teacher willing to experiment. The barrier to entry has never been lower.
But the real implication goes deeper than just making fun apps. AI is reshaping what education can look like. It allows us to meet students where they are, speak their language, and turn passive consumers into active participants. When a student spends their break playing my buggy Tetris game or begging to show their AI-generated face to their parents, something has shifted. They’re engaged. They’re learning without realizing it.
Of course, none of this matters if we refuse to change. Traditional classrooms are dying — I said it at the beginning, and I’ll say it again. None of my students read any textbooks anymore. We can mourn that loss, or we can adapt. I choose to adapt. That means changing our mindset, rethinking what a “lesson” looks like, and accepting that the tools of engagement have evolved.
The students aren’t going to put down their phones. So maybe it’s time we picked them up too.
Comments
Post a Comment