Week 4

Teaching AI to K-12 Learners: Lessons, Issues, and Guidance

Grover’s paper does an excellent job of presenting the current state of AI/ML education field. The paper covers most of what we’ve read in the literature, touching on the Five big Ideas Framework and discussing the need for AI Education for K-12 students. Where we really start to see the main idea of the paper is Grover’s review of the different tools that have become available in the field in the last few years.

Grover’s emphasis on including teachers in the design process is important when approaching AI education for K-12 students. A challenge that educators and researchers are going to have to face as AI becomes more sophisticated, is STEM teachers understanding of the field as it currently is. Although there is a myriad of tools at teacher’s disposal to educate their students about AI, it is ultimately meaningless if they don’t understand the AI concepts themselves. Bringing teachers into the design process is an excellent suggestion to keep them engaged with the material before it hits their students and serves as a way for researchers to understand how they should design their tools for classroom settings.

I also liked how Grover made a point to broach the question of if AI education should be included in regular K12 CS education, or if it should be spun off into its own curriculum sect. As AI becomes more separated away from actual programming and heads more into an emphasis on data science, it will become more pressing to figure out how to orient AI education in a school curriculum.

Exploring children’s attitudes toward an age-tailored AI-powered chatbot

The researchers in this study had students, aged 7-14, interact with an AI chatbot that was prompt engineered to only answer questions about a few select topics. One of the primary goals of the research was to understand the children’s attitudes towards the chatbot and gain insight into critical design elements that could be used for future projects.

The researchers noted the students’ interactions with the chatbot to assess the student’s learning engagement. They measured student engagement level based on the number of questions each student posed and the number of topics that each student chose to engage with. The researchers also identified three distinct themes that characterized the interactions that the students had with the chatbot: Trust and Developing Confidence; Expressing Curiosity, Wonder, and Surprise; and Building Relationships and Anthropomorphizing.

Finally, the researchers developed pre- and post- survey questionnaires to see how the student’s opinion of the chatbot had changed after they interacted with the project. To ensure that they captured good data from all their participants, they altered their survey methodology for their younger participants by simplifying the language and reading the questions out loud.