Grover's Preprint Reflection
The paper seems to brought to light a lot of aspects of teaching to K12 students I hadn't realized, for example how important it is to teach early on the biases that can occur in AI. For me this hadn't really struck me as a topic that would be considered important in educator's eye, but it is commendable that it is. Another interesting excerpt was the concept of whether or not we are the ones to create AI, or are we just correcting and adjusting it. It's also worth pointing out the need to teach data driven approach for teaching rather than algorithimic teaching of the concepts of AI. And while it is imperative to teach students of K12 these concepts and is possible, it should be noted that the issue of teaching AI lies within the approach we choose rather than the subject itself, since all AI really is a sub topic of computer science. And this article emphasizes that point, saying that we should be teaching basic concepts for AI litteracy and not go too deep into it. Teaching AI to K12 students will come to down how we handle the material and not the material itself. An interesting quote I found was when it said, “any subject can be taught effectively in some intellectually honest form to any child at any stage of development”, and I believe that to be fascinating but also true. And I believe that is a valid point to keep in mind while we create our program for the project.
Exploring Children's Attitude's Paper Reflection
It was interesting to read on how the program functioned under the hood, with how the program takes in dynamic words and puts in a prompt to pass into the OpenAI API prompt. I thought there was more going on in terms of the prompt generation, but seeing it simplified made the task of doing something similar less daunting. It was interesting to read that the children asked questions to trust the AI chatbot in a way to trust to see if it was reliable. It showcases some compentacy when utilizing technology, but also hesitation to rely on technology for information at least. But maybe this was only due to the fact that the AI chatbot was being claimed to be knowledgeable on topics these children knew about, and perhaps if it hadn't been on topics that they knew about it, they would be less skeptical on the information being passed. I do find it interesting that after it answered these questions correct that they started to trust it more, and it makes me wonder if that will occur with the program we make during this semester and if we will have to account for that in the five minutes we have. I found it humorous that during the interviewing, that some of the students still responded showing some distrust with it. But it seems overall, the students did trust it to some degree, and it showcases some caution when building the program that we are careful with the information that is being passed and stored.