How AI Makes Decisions

This lesson will introduce students to the fundamentals of Artificial Intelligence (AI) and machine learning by focusing on predictive data models. Students will get hands-on practice with supervised and reinforcement learning. They’ll start with data that has already been sorted and progress to making predictions on their own of how to sort items on screen. In the final level, students will try their hand at sorting data on their own and will enlist a partner to predict how the items have been categorized.

This hands-on experience will be complemented by discussions on the significance of data categorization and the challenges posed by limited datasets. Students will explore a real-world example where AI predictions have been used to help humans. The lesson aims to demystify AI technology, emphasizing its pattern-based logic rather than perceiving it as magical or random.

In this lesson, students will learn about how the inputs they give a machine learning model directly impact the predictions that the model makes. First, they will learn about the difference between random predictions and AI predictions, before training their own AI model to make predictions on whether an item is a fish or trash in the ocean. Finally, students will discuss the implications of biased training data by discussing hypothetical scenarios.

In this lesson, students will begin thinking about how to classify real world objects by their features. They will build out two feature vectors (foods and emojis) in order to learn how AI makes decisions based on the probability of an outcome. Students will use “Dance Party: AI Edition” to see first-hand how AI can make tasks more efficient, but that it might come at the cost of customization. Students will consider what it means to collaborate and co-create with AI.

In this lesson, students will evaluate sample student data. They will choose a hypothetical AI tool and consider what training data the tool will need to successfully make a decision. Then, students will write a letter to the school principal explaining their opinion on using AI tools. At the end of the lesson, students will reflect on what they have learned about AI over the course of the 4 lessons.

Students will understand that a predictive AI model is only as good as the information it was trained on. Therefore it is important to train the model on a large amount of data that is relevant to the task at hand. Also, students should recognize that there needs to be human oversight when it comes to working with AI to ensure that any mistakes made by AI get corrected.

In this lesson, students will create an artifact explaining their opinion on using AI tools. They have the option to write a letter to the principal, make a speech or song, make a poster, create a Sprite Lab project, or code a poem to share their opinion. At the end of the lesson, students will reflect on what they have learned about AI over the course of the module.