Toggle open
Learning Library Blog Kids Teaching Kids — With Chatbots
Expand breadcrumbs

Kids Teaching Kids — With Chatbots

By Nicole Krueger
January 25, 2022
Chatbots blog Version Id81b5ta X Nxvs92l mu Wc3kzsh Md YA4rjb

The client: A second grade classroom learning about nutrition.

The developers: Eighth grade computer science students exploring how artificial intelligence works.

The product: A chatbot capable of dispensing information about local produce.

Today’s students will live and work in a world with artificial intelligence, and computer science teacher Sharon Harrison wants them to be ready. Learning how AI works is a good start, but it doesn’t necessarily prepare kids for the complex decisions they’ll have to make as stewards of this powerful technology.

Instead, Harrison gave her students at University of Chicago Laboratory Schools a taste of what it’s like to be a programmer working with artificial intelligence. Placing students in the role of software developers, she challenged them to create an AI system that meets the needs of a real-world client — in this case, a chatbot that can answer younger children’s questions about nutrition.

“I think it’s important at the moment for students to have an awareness of AI in general and the impact AI is going to have on our future – in the near future,” she says. “For that particular class, I also wanted them to realize it’s quite achievable to make a chatbot yourself.”

While chatbot creation tools make it easy for students to build and deploy their own conversational software, Harrison had her class use raw programming language to code simple chatbots with a command-line interface so they could focus the underlying structure and mechanics instead of the software’s visual components.

“I wanted them looking more at the mechanics of the chatbots, rather than having another level of distraction and not really understanding how it was functioning,” she says.

Many adults interact with chatbots in their daily lives, often without realizing it. But when Harrison’s students encounter this technology in the future, they’ll not only be able to recognize it for what it is, but they’ll understand how and why it’s capable of processing questions and delivering appropriate responses.

Why does it work?


Breaking into small groups, the student developers worked together to determine what information to include and how their chatbots would respond to basic questions. With a flowchart as their road map, some groups divided the work into sections, with individuals coding in parallel and then merging the parts into a cohesive whole.


Having an authentic client added an extra layer of complexity to the project as students learned how to design for a specific audience. “They had to consider their menu choices and how to prompt and get information from second graders who might make spelling mistakes or talking mistakes. We looked up interface design, language and how to use information so that the second graders could understand it,” Harrison says.


Like any software-development project, the assignment had its hiccups. The development team met with their clients only once for feedback, and the second graders, whose field trip to the local farmers market got rained out, didn’t come prepared with as much information as they expected. Just like professional programmers, the eighth graders had to do the best job they could with the resources they had. Some of the chatbots were simple, others more complex, but all were functional.

Despite the setbacks, Harrison says, the positive feedback from students was overwhelming.

“They really enjoyed it,” she says. “I most definitely would do it again.”

AI is already here. Are you ready to teach the future? Learn how when you read Teaching AI.

Nicole Krueger is a freelance writer and journalist with a passion for finding out what makes learners tick.