Help students use AI to support their learning

An assignment in a public speaking course requires students to develop a speech over a period of weeks and then deliver the speech in class. The first step is to come up with a topic for a persuasive speech. Students have read about the key elements of a persuasive speech and the key areas that a speaker should consider when choosing a topic. 

Working in small groups, students use prompts based on key course concepts and ask AI to generate a list of 20 possible topics that might interest their peers. The students then workshop those topics, deciding which ones might be of interest and which might not be based on their previous readings and discussions. Students then individually use the conversations they have had and their own ideas to make decisions about which persuasive speech topic they may want to continue exploring. Students spend some time writing an individual journal entry about what they think they want to speak about, why that topic will be effective, and the ways that thinking with AI helped or hindered their decision-making process. Students also write critically about which topics AI suggested that they felt were particularly helpful, bizarre, or evidence of the limitations or bias that can influence AI generation of text.

Students in a statistics course are working on an assignment that requires them to make a set of decisions to help a fictional statistician. They are required to explain their suggestions in their own words to the statistician. 

The instructor tells students that as they work with concepts from the course (e.g. “degrees of freedom” or “p. value”), they can prompt AI to generate examples or explanations that will help them understand these concepts, but they should use their own words when they write up the assignment. The instructor asks them to share their conversations with AI as an appendix to the assignment and a brief written statement about how this conversation changed their thinking about specific concepts.

Students in a marketing research course are working on a semester-long marketing research assignment where they work with a community partner. One of the early steps in this assignment is the development of survey questions. 

The instructor asks students to work in small groups and draw on their individual efforts to draft 10 questions for a client survey based on survey principles and frameworks they have read about previously. Students then prompt AI to evaluate the questions for construct reliability in relation to two key constructs. Students then work in their groups to analyze the feedback they got from AI. If the suggestions were valid, they need to be able to say why. If the suggestions were not useful, they also need to explain why they are rejecting those changes. Students then spend some time reflecting on the strengths and weaknesses of AI as a “thinking partner.” This annotated new list of survey questions is submitted for feedback to the instructor along with the reflections about the use of AI at this stage of the marketing research process. 

Students in a phonetics course have a test coming up on the phonetic characteristics of consonants. The instructor gives them the following prompt as a model of how they might ask AI to quiz them on their knowledge of the test material. 

“I am an undergraduate student taking a course in phonetics and we have a test coming up on phonetic characteristics of consonants. Ask me questions about this topic to help me see what I understand and where I may need to study more. Ask me one question at a time and make each question short answer or multiple choice. After I answer a question, give me feedback on my answer, then ask another question. When I get a question wrong, make the next question a little easier. When I get a question right, make the next question a little harder. Keep asking me questions one by one until I ask to stop. When I ask you to stop, give me a summary of which topics I need to study more.” 

After this experience, students write a brief summary of what new learning took place during the quizzing session and how they can use AI in the future as a quizzing partner.

Students in an art history class are focusing on the post-impressionists. The instructor realizes that students need a lot more practice in making finer discriminations between works from this period in terms of form and content. The instructor gives students the following prompt as a model of how they might ask AI to quiz them: 

“I am an undergraduate student taking an introduction to art history class. I would like you to quiz me. I want to develop my ability to compare and contrast the styles and the work of post-impressionist artists. Please give me pairs of paintings (show me the images) and ask me difficult questions that require me to compare and contrast them in relation to both the technique the artists use as well as the subject matter of the paintings. Keep giving me pairs of paintings until I ask to stop. When I ask you to stop, give me a summary of what aspects of post-impressionism I understand well and what aspects I have confusion about.”

After this experience, students write a brief summary of what new learning took place during the quizzing session and how they can use AI in the future as a quizzing partner. 

I am an undergraduate ecology student and I am trying to improve my research analysis skills. I read this article [insert link or file]. Please ask me questions about the methods, findings, and limitations in this article. Ask me one question at a time and make each question short-answer or multiple choice. After I answer a question, give me feedback on my answer, then ask another question. When I get a question wrong, make the next question a little easier. When I get a question right, make the next question a little harder to go beyond memorization and comprehension to evaluation and synthesis. Keep asking me questions one by one until I ask to stop. When I ask you to stop, give me a summary of which parts of the article I understand best and which parts I have confusion about.

An instructor wants students to make decisions about the primary sources they will draw on to create their history projects and to find themes in those sources to help focus their interpretations. She tells students to use AI to find five relevant primary sources for a history paper and to use AI to create an annotated bibliography of those sources. She tells students that when they turn in their annotated bibliography to tell her what prompts they used. Her rationale is that historians can use AI to do this work.

The problem: If the instructor wants students to learn the skills of choosing and annotating sources, the students must build these skills. Offloading this work to AI means she is “de-skilling” her students. Historians might use AI to do some of this work, but they would also be critically exploring what AI generates. The instructor should not have students use AI to do the work for them. Alternatively, the instructor should have students review and critique the work AI does.

An instructor assigns a weekly research paper to students in an introductory genetics course that aligns with key concepts and topics to demonstrate the relevance of these concepts and topics. She wants students to know about this research, but she feels that freshman are not prepared to read and understand the research papers. She tells students to feed the papers to AI and have it summarize the papers for them.

The problem: If the instructor wants AI to do this work for her students, reading research may not be a necessary assignment in her course. Offloading this work to AI indicates that the work is unnecessary. The instructor should reconsider what students can do with the summaries AI produces or should reconsider this assignment.

An instructor wants students to learn to identify, create, and use figures of speech to good effect in public speaking. He has students prompt AI to create twenty novel metaphors and asks them to find the most unusual and unexpected metaphors.

The problem: If the instructor wants students to create metaphors, simply seeing metaphors generated by AI will not help students. The instructor must design work for students so that they develop the skills they need to use figures of speech themselves. There is little to no connection between the use of AI and the aims he has for his students’ learning.