How did we come up with the idea of “scaffolding chatbots”?
During our first brainstorming session, we quickly agreed on one thing: basic chatbots have become widespread, but most of them leave much to be desired. These chatbot implementations are uninspired, lack utility, or provide answers and explanations so efficiently that it actively undermines the learning process. Teachers hate them. Students are tired of them. These tools often ignore decades of learning science research about how learning actually works. They either lack real educational value or make learning so “efficient” that they bypass the deep thinking essential to real understanding.
That got us thinking—what if chatbots could support learning rather than replace it? What if they could guide learners through critical thinking using strategies like Socratic questioning? This idea led us to Vygotsky’s concept of the “more knowledgeable other”. We believe there is significant opportunity in the gap between AI chatbots and pedagogical theory, one which keen investors should review.
What can you expect on our website?
You’ll start by learning (or reviewing) what chatbots are and the different types that exist. Then, we introduce our concept of “scaffolding chatbots”—and yes, we’ve built one for you to try!
Next, you’ll explore our market analysis of current chatbot products, our thoughts on the technological trajectory of AI in education, and the potential risks these tools may pose. Along the way, you’ll find interactive Padlet discussions where you can reflect on your own experiences with chatbots and read what your peers have shared.
We’d love to hear from you!
After exploring our project, please let us know:
- What’s your biggest takeaway from our website?
- Do you have any suggestions for how we could improve it?
Feel free to leave your comments below—we’re excited to hear your thoughts!
Hi everyone, just a quick reminder—please don’t forget to add your name to your post on the Padlet discussion board. We need everyone’s names for recording participation. Thank you!
Hello, team! I enjoyed browsing your OER. It is insightful and shows that the topic is well-researched. The interactive elements make it more engaging, and I love how you clearly explained chatbot. My biggest takeaway from the chatbot is how it emphasizes the importance of aligning chatbot design with educational theory on MKO, or the More Knowledgeable Other. What I like about your idea is it’s focused on supporting learners through guided questioning and structured reasoning. I learned that chatbots can move beyond passive tools to become active learning partners that scaffold thinking and promote deeper understanding. I agree that chatbots should be used as a tool for learning rather than replacing it. With their potential, chatbots can be an effective learning tool.
Hi Yan and Duncan,
Great work this week on your OER! I really enjoyed learning about chatbots and never realized that there were so many different types of chatbots (Rule-based, Chat-based, AI-powered, etc). I appreciated how you included a little activity on each page to engage your learners and help them revisit key concepts. My biggest takeaway from the OER is the potential of AI-powered chatbots to be used in education. The emerging tools you mentioned, MATHia and Squirrel AI, provides personalized education and even helps identify any learning gaps. My one suggestion for improvement would be to make the quizzes at the end of the page longer rather than just one questions. Other than that, great work this week and thanks again for teaching me all about chatbots!
Hi Yan and Duncan,
I really enjoyed your OER! I appreciated how you two wove in Vygotskyian concepts relating to scaffolding with the topic of AI chatbots. I thought the way you two presented it was very clear-minded and persuasive in terms of the benefits that MKO-style chatbots could confer! Interacting with the AI chatbot that you created was a helpful experience since its socratic-style questioning offered a different experience than the popular models we are typically used to. Overall, great job!
Hi Yan and Duncan, thank you for this OER. It was very educational for me – its amazing to see how:
1. AI develops every day
2. Makes it accessible
3. Not all AI is so fantastic!
I takeaway what you mentioned – you need to add value to a ChatBot, by trainning it to do what you want it to, or else its just not adding anything to the user.
I think that well trainned chatbots can be a great educational tool (socratic trainned, for instance) and they are a way to avoid the laziness that can spike if we just keep using GPT for everything we do!
Seeing the prompts and info one has to imput in order to train was also very interesting!
Thank you for this detailed exploration. I thought this was a really well polished set of activities and I really enjoyed the custom chat bot experience you built. I also hadn’t looked very closely at squirrel AI before this week but I now see that it’s one of many exciting new ventures in this space.
I think that real-time AI powered support/interactions will become ubiquitous very soon and the applications for the classroom are endless so this is a very timely topic. My biggest take away from the content this week is that we are on the cusp of some incredibly exciting developments in the area of chat bot style supports for classrooms. I hope somebody creates an effective solution for k-12 teachers and students soon because it would have an incredibly powerful and positive impact.
Hi Duncan and Yan,
Amazing work on this week’s OER. I really appreciate the effort you put into coding your own chatbot and sharing the source code. I am inspired to use it as a base to create a little chatbot of my own. And I think that was the thought process when I explored the OER – Chatbot gives avenue to output, but input needs to be co-created between human and machine. That being said, the creation process by human has more or less been influenced by AI. As a result, we need to be aware of the risks and flaws when integrating chatbots into school or work settings.
Hey Yan and Duncan,
I learned a lot from this module!
What’s your biggest takeaway from our website?
– I loved the chatbot! It was the activity I spent a lot of time on just asking it questions and seeing how far I could take it with the content. Thank you for sharing the source code.
– I appreciated know about the accessibility and option I can have with chatbots. I have never considered it as a teaching tool and before the OER, I’ve had a story that it was all marketing.
– The input you put matters. One of the other things I learned in the module is the amount of human input needed to have the chatbot behave in a certain way. It sounded like I was speaking to a person which I thought was a nice touch.
Thank you for this.
Hi Duncan and Yan,
Sorry for the late response. I finished this OER and forgot to add to the discussion board!
First, amazing job on this resource. It was incredibly easy to follow along, and the information was presented in a way that was easy to understand and digest. I love your idea of a scaffolding chatbot. Too often are chatbots just giving answers instead of trying to elicit them for the user. Your take on what chatbots could do to support learning is incredibly innovative, and I could see it helping ESL learners gain confidence in their language abilities. I would say my biggest takeaway is that we should consider how to effectively implement tools in educational settings instead of just outright banning them. It is definitely easy to prohibit or block chatbot websites on school networks, but if we use them in a way that can promote and encourage students’ motivation and learning, then perhaps there is no need to ban such tools.