Managing AI in the Classroom

Last Update: Nov 22nd, 2024
Written by: Neil Chong-Kit, Authorship AI

Introduction

Generative AI is reshaping the classroom, presenting both opportunities and challenges for educators. As a secondary school teacher or undergraduate instructor, you have the flexibility to determine how AI fits into your course. However, setting clear boundaries and guidelines is crucial to ensure that AI tools enhance learning rather than undermine academic integrity.

This guide offers practical advice on integrating generative AI into your teaching. It covers course planning, managing AI misuse, and setting clear expectations, helping you create a thoughtful approach to its use in education.

Disclaimer

This guide is provided by Authorship AI, as part of our mission is to ensure fairness in the classroom. While care has been taken to present this topic without bias, implicit bias may still exist. Learn more about us, or try our free tool that verifies authorship by quizzing students on their own papers at https:/authorshipai.com.

Overview

Managing AI in your classroom begins with understanding what restrictions are set by your school's academic integrity policy and guidelines.

Your school's policy may vary, but in general:

• inclusion of any AI generated text is considered plagiarism, unless cited
• general use of AI is prohibited as it is considered an unauthorized aid
• teachers may permit the use of AI at their discretion
• Permitted AI use should be included in the course syllabus

See Appendix A for more information.

Therefore, as a teacher, you have a new responsibility to determine the appropriate student use of AI in your courses. While an outright ban of AI is an option, the threat of punishment alone has not proven to be a sufficient deterrent.

While not definitive, the remainder of this guide is meant to provide some helpful information in creating a more effective strategy for managing AI in the classroom.

Understanding Student Motivation describes why students use AI.

Permitted Use Cases outlines different activities where the use of AI may be appropriate.

Dealing with Unauthorized AI Writing provides advice when students submit work written by AI.

Understanding Student Motivation

To mitigate the unauthorized use of AI, it is helpful to think beyond "students use AI to write their assignments because they are lazy".

In our opinion, well-intentioned students are more likely to use AI to write their assignments when the choice is between cheating or not meeting the submission deadline. A possible mitigation strategy is to prevent students from reaching this point.

The Student Perspectives on AI Tools in Academia in 2024 survey highlights key challenges students face when preparing for writing assignments:

Difficult to start/gather up the mind	31%
Difficult to formulate an idea	24%
Difficult to gather sources	22%
Lack of time	18%
Non-native speakers	3%

Somewhat ironically, allowing students to use generative AI to help formulate ideas, gather sources, and work more efficiently, can be part of the strategy in stopping students from using AI to write their paper for them. The next section, Permitted Use Cases provides further details.

Of course, existing techniques in helping students with time management should also be part of the mitigation strategy.

Permitted Use Cases

Being specific about when AI is allowed to be used, helps students make educated choices. The desire to use AI is strong. In addition to reasons outlined in the previous section, sanctioning the use of AI for tasks that don't compromise learning objectives prevents the slippery slope effect.

"I've already used AI to generate ideas for my paper when I wasn't supposed to. Since I've already crossed the line, I might as well use AI to complete the assignment."

Potential Permitted Use Cases

1. academic help
2. idea generation
3. gathering sources
4. document structure and formatting

Below are some example instructions for students when some use of AI is beneficial to their learning.

Academic Help
"The use of AI is permitted when you need help to understand a problem or assignment details, or to review an assignment that has already been marked. You may not use AI to give you the answer to a problem or generate the content of an assignment in whole or in part."

Idea Generation

"When brainstorming ideas for your paper, feel free to use AI to spark inspiration. For example, you could ask, "What are some common themes in Romeo and Juliet?" Afterward, try to take a less common approach for your paper—unique perspectives tend to stand out and earn higher scores."

Gathering Sources

"You may use AI in your initial search for source material. However, you must find and subsequently cite and quote from the original source. AI is known to sometimes misrepresent the original material."

Document Structure and Formatting

"You may use AI to provide feedback on your paper's structure and formatting. For instance, asking AI `Does my paper meet the formatting requirements of this assignment, and how can I improve its layout?` Asking AI to rewrite or revise your paper is not permitted."

For some permitted use cases, you may want to require students to disclose how they used generative AI. This can be as simple as students providing a copy of their conversation, or require more formal citation. Here are three articles describing how students can cite the use of generative AI.

How to cite generative AI output (MLA)

https://style.mla.org/citing-generative-ai

How to cite generative AI output (APA)

https://apastyle.apa.org/blog/how-to-cite-chatgpt

How to cite generative AI output (Chicago)

https://www.chicagomanualofstyle.org/qanda/data/faq/topics/Documentation/faq0422.html

Dealing with Unauthorized AI Writing

Even with clear guidance, it’s almost inevitable that some students will submit AI-generated work as their own.

Based on a survey of educational policies, the most common approach for undergraduate and lower-level students is to address the issue through a discussion before pursuing any formal disciplinary action. For graduate-level work and above—where originality and potential publication are expected—AI-generated content is treated as plagiarism, with stricter consequences.

In general, it is recommended that teachers:

Use AI Detection Tools with Caution: Recognize the limitations and potential inaccuracies of AI detection tools. Treat their results as a starting point for inquiry rather than conclusive evidence.

Review Writing Evidence: Encourage students to maintain drafts and timestamps of their work (e.g., in Google Docs). This can serve as verifiable proof of their writing process if questions about authorship arise.

Engage Students Directly: Instead of immediately accusing a student of misconduct, initiate a conversation about their work. Ask about their research methods or require clarification on specific aspects of their submission to gauge understanding.

Collaborate with Colleagues: If suspicions arise, seek input from peers or department heads before taking action. This ensures fairness, protects students, and helps educators align with institutional policies.

See these articles for real-life examples:

My Students are Submitting AI Papers - Erik Ofgang

He Was Falsely Accused of Using AI. Here’s What He Wishes His Professor Did - Erik Ofgang

Summary

Managing the use of generative AI in the classroom is a complex but necessary task for educators. Clear communication, proactive course planning, and thoughtful policies tailored to your students' needs can help ensure AI tools are used responsibly to enhance learning rather than undermine it.

Setting clear expectations, providing guidance on permitted use cases, and addressing common student challenges like time management and idea generation are effective ways to discourage misuse. By fostering open discussions and offering constructive solutions, educators can create an environment that prioritizes integrity and learning.

Reliability of AI Writing Detection Scores

The opinion of this author is that AI writing detection tools like Turnitin should not be used to decide which students face disciplinary actions. The unreliable accuracy rates creates unfair situations where innocent students are accused while actual offenders face no consequences.

Factors which affect the reliability of AI writing detection scores:

• Vendor algorithm changes can raise or lower the score
• Advancements in generative AI can raise or lower the score
• Using auto-corrections in spelling or grammar tools can raise the score
• Students who naturally write in a style similar to AI, have higher scores
• Students can tweak AI output to seem human, artificially lowering the score

Please try our free alternative at https://authorshipai.com. It quizzes students on their own papers, to test for authorship based on understanding.

Appendix A: Generative AI Academic Integrity and Policy

The following appendix includes excerpts from educational institutions who have made their generative AI policy public.

Key takeaways from a survey of these and other policies:

• existing plagiarism policy statements generally prohibit the submission of AI output
• existing statements regarding the use of unapproved aids generally prohibits the use of AI
• student use of AI is commonly left to the discretion of the teacher / course instructor, with recommendation to disclose permitted use in the course syllabus
• there are some privacy and intellectual property concerns when students submit their own work or course materials to AI tools

The University of Texas at Austin

Resource Link: https://ctl.utexas.edu/chatgpt-and-generative-ai-tools-sample-syllabus-policy-statements

Their overall position is that use of generative AI is already covered under their existing university policy regarding plagiarism. "It is already a violation of policy for students to represent work they did not do as their own, and work generated by an AI system that is not credited to that system falls under that policy." They do not provide a default or blanket policy regarding the use of generative AI beyond this statement.

Allowing students to use generative AI is left to the discretion of faculty and instructors. They recommend this is made clear to students at the beginning of a course, and provide sample policy statements instructors can include in the course syllabus.

No Generative AI Policy Statement

"This course assumes that work submitted by students – all process work, drafts, brainstorming artifacts, final works – will be generated by the students themselves, working individually or in groups as directed by class assignment instructions. This policy indicates the following constitute violations of academic honesty: a student has another person/entity do the work of any substantive portion of a graded assignment for them, which includes purchasing work from a company, hiring a person or company to complete an assignment or exam, and/or using generative AI tools (such as ChatGPT)." - CTL at UT Austin

Limited Generative AI Policy Statement

"Understanding how and when to use generative AI tools (such as ChatGPT, DALL-E) is quickly emerging as an important skill for future professions. To that end, you are welcome to use generative AI tools in this class as long as it aligns with the learning outcomes or goals associated with assignments. You are fully responsible for the information you submit based on a generative AI query (such that it does not violate academic honesty standards, intellectual property laws, or standards of non-public research you are conducting through coursework). Your use of generative AI tools must be properly documented and cited for any work submitted in this course." - CTL at UT Austin

Encouraged Generative AI Policy Statement

"Students are invited to use AI platforms to help prepare for assignments and projects (e.g., to help with brainstorming or to see what a completed essay might look like). I also welcome you to use AI tools to help revise and edit your work (e.g., to help identify flaws in reasoning, spot confusing or underdeveloped paragraphs, or to simply fix citations). When submitting work, students must clearly identify any writing, text, or media generated by AI. This can be done in a variety of ways. In this course, parts of essays generated by AI should appear in a different colored font, and the relationship between those sections and student contributions should be discussed in cover letters that accompany the essay submission." - CTL at UT Austin

See the link above for additional sample generative AI policy statements.

Harvard Graduate School of Education

https://registrar.gse.harvard.edu/AI-policy

They have a blank policy statement not allowing the usage of generative AI by students to complete an assignment. If permitted by an instructor for a course, usage must both be cited and explained.

Excerpt from their policy:

"1. Unless otherwise specified by your instructor, it is a violation of the HGSE Academic Integrity Policy to use generative AI to create all or part of an assignment for a course (e.g., a paper, memo, presentation, or short response) and submit it as your own. …

2. Permissible uses of generative AI in HGSE coursework include seeking clarification on concepts, brainstorming ideas, or generating scenarios that help contextualize what you are learning. For instance, it is fine to use AI-powered web search and to have “conversations” with tools like ChatGPT to help you explore ideas, refine your thinking, identify examples, and better understand course material.

3. For any permitted use of GenAI tools, you must acknowledge and document that use in your assignment submission by explaining what tool(s) you used, prompts you provided (if applicable), and how you integrated the output into your work

…" - HGSE

University of Toronto

University Policy
https://www.academicintegrity.utoronto.ca/perils-and-pitfalls/using-chatgpt-or-other-ai-tool-on-a-marked-assessment/

They have a blanket policy statement for the university as a whole, and specific guidance for graduate theses. Their existing policy prohibits the use of generative AI.

"It shall be an offense for a student knowingly … to use or possess an unauthorized aid or aids or obtain unauthorized assistance in any academic examination or term test or in connection with any other form of academic work." - UT

While this statement does not explicitly state generative AI is not allowed for brainstorming, they clarify that if an instructor has explicitly prohibited the use of generative AI tools, it's a violation to create even a part of an assignment using such tools.

Graduate Thesis

https://www.sgs.utoronto.ca/about/guidance-on-the-use-of-generative-artificial-intelligence/

The graduate thesis guidelines are more prescriptive. Any use of generative AI must be planned and approved in writing. Failure to do so may be considered an offense.