Working with friends on the AI Assistant project teaches several important lessons. Students learn about collaboration and teamwork, including how to divide responsibilities for collecting different types of data. They learn technical skills from each other, such as how to use Teachable Machine effectively, how to organize data into proper categories and sub-categories, and how to label images correctly for training. Friends may share tips on improving model accuracy, such as collecting more diverse images or adjusting training parameters. Students also learn about problem-solving when the AI model makes errors, and how adding more training data can improve recognition accuracy. Additionally, friends may share creative ideas for making the AI Assistant more interactive using Scratch, such as adding characters, animations, and personality to the assistant. The collaborative experience helps students understand that creating an AI Assistant requires planning, organization, testing, and continuous improvement.