Generative AI Classroom Guidelines: An Example
Indicate allowed and non-allowed uses
I have been experimenting in class with ungraded GenAI exercises in class since Spring of 2023 and have also talked with a variety of students and faculty about GenAI inside and outside my university during this time. The question I am asked most often by faculty is what guidelines I use for AI in the classroom.
Here for your information is my current set of guidelines, with some university- and class-specific terms XXXX’s out. Steal any or all text that is helpful. No attribution is necessary.
Generative AI Guidelines
A wide variety of generative AI tools is available to professionals including both text- (e.g., ChatGPT) and image-generating (e.g., Midjourney) tools that will create output in response to inputs (“prompts”). Some assignments will require the use of GenAI as indicated earlier in this syllabus. Beyond those specific assignments, here are allowed and non-allowed uses of GenAI for this class. Note these guidelines are for our class only; other professors (and employers) may have different guidelines.
Allowed Uses:
· Generating ideas. AI tools can be helpful in suggesting ideas, especially for your XXXX Project. Treat this as one starting point among others. You need to evaluate whether the ideas are good.
· As an alternative search engine. Some AI tools such as Copilot, Gemini, and Perplexity are connected to the internet and thus may be helpful in identifying sources that you may then review.
· Generating images. You may use image-generating tools to produce images to illustrate a deliverable; as with ideas, this is likely most relevant to your XXXX Project. The image source should be noted in the deliverable (e.g., “produced by Dall-E 3”)
· Providing feedback on writing. In all written assignments you may ask a text model such as GPT or Claude for feedback on your writing. A prompt that may be helpful: “Imagine you are an expert in US business English, which favors simple, direct language and short paragraphs. Improve the writing in this draft.”
Note that at time of writing, none of the AI tools are good at word counts.
· Providing feedback on ideas. Text-based AI can sometimes be a useful critic of ideas. For example, you may submit an idea to an AI and ask it to identify strengths, weaknesses, areas for improvement or concern, etc. This is another starting point. You need to evaluate whether the feedback is good.
If you use AI in your XXXX Project, please include a paragraph in your assignment indicating how you used it (e.g., tools, prompts). This will not be part of the grade for your assignment, but will help me understand how you are using the tools.
If you use AI to improve writing on your case reports, I would like you to submit both your pre-AI draft and your post-AI final deliverable. I will only grade the latter, but I would like to see where you start in writing, as it may suggest advice I can give you.
Non-Allowed Uses:
· Direct use of AI output. With the exception of the image generating example earlier, you may not simply copy and paste an AI output into a deliverable. This is a violation of academic integrity guidelines on plagiarism, which indicate: “using as one’s own the words, ideas, data, code, or other original academic material of another without providing proper citation or attribution.” Quotation marks around AI text and citing it are not allowed in this class. All references in this class must be ones that I can check. AI text does not exist independent of the prompt, tool, and moment that you use it, and so fails this test.
· As a source of facts and analysis. All text-based AI generators will quickly generate plausible-sounding facts, numbers, and analysis. That does not mean these are correct. You must check everything it produces. Where relevant, ask for sources, review those sources and then incorporate those sources (with citations) as appropriate in your assignment.
· As a source of XXXX coursepack analysis. Do not upload cases or readings from our XXXX coursepack to an AI program and then ask for an analysis. This is a violation of copyright. (It’s also a terrible way to learn.)
Note in all cases, “the AI told me to” is not a valid argument in grading. I will evaluate the quality of your deliverables using the usual criteria for the course.
In all aspects of our course, the University’s policy on academic integrity governs our use of any tools or sources.
Bruce Clark is an Associate Professor of Marketing at the D’Amore-McKim School of Business at Northeastern University where he has been a teaching mentor for both online and on-ground teaching. He researches, writes, speaks, and consults on managerial decision-making, especially regarding marketing and branding strategy, and how managers learn about their markets, though increasingly he is engaged with GenAI in business and higher ed. You can find him on LinkedIn at https://www.linkedin.com/in/bruceclarkprof/.
Beyond the headline image, no AI was used in the writing of this article!
