Guidance for Artificial Intelligence (AI) and the application of AI to CGU

Before utilizing yourself, and/or encouraging the use of or asking others, such as students, to use AI tools in their work, please investigate how these tools collect and utilize personal information and data, including reading any privacy policy linked with a particular AI tool, including log-in data, tracking, and other metrics (Center for New Designs in Learning and Scholarship, 2023). You should be mindful of your own data privacy, when you input your prompts into an AI tool to see the output.

Security: Read through privacy agreements and make sure you are personally comfortable using different AI-based tools. Consider using more vetted tools in university-licensed programs. Never put private information (especially regarding students or including student data or your own data) into consumer tools.

*Note: Similarly, OpenAI has also published a page on safety best practices cautioning the output of AI tools, including having a ‘Human in the Loop,’ emphasizing having a human verifying the results of any AI interaction.

What about FERPA and other privacy compliance considerations?”]

The Family Educational Rights and Privacy Act (FERPA) are the set of laws designed to protect student information in Higher Education. Instructors who may want to make use of these tools should note that, it is considered FERPA compliant as long as users have logged in using their CGU accounts when signing up for these AI tools for CGU coursework. If you are considering requiring students to use an AI platform to complete coursework, please request the students to use their CGU email addresses. Note that with any online service that is not contracted by the university, in view of FERPA regulations, students must not be required to identify themselves to third parties.

Academic records, including assessments such as examinations and assignments, are considered a student record and protected by FERPA. For example, the free version of ChatGPT should not be used to draft initial feedback on a student’s submitted essay that included their identifying information. Asking the free version of ChatGPT to respond to general question prompts would not be a FERPA violation, as no student information is provided to ChatGPT.

• Appropriate Use of Computing Policy

Most companies that own generative AI tools reserve the right to maintain and utilize user data for various production purposes, as well as share them with third-party stakeholders. Additionally, for the majority of tools, the company will own any outputs the tool generates. This has far-reaching implications for the definition of authorship and public knowledge domain. Additionally, there is often a possibility that small batches of user data may undergo human review in the production process within the tool-owning company. For these reasons, Please do not, and teach your students to never upload unpublished work or share personal and sensitive information within generative AI tools of any kind.

For Instructors:

Develop a course policy on AI use, addressing not only plagiarism concerns and ground rules, but also issues of AI ethics, authorship, privacy, and online safety. Discuss it with students on the first days of class. Continue discussing the relevant aspects of the policy throughout the semester.

Keep in mind that students are likely facing major differences in faculty opinions about generative AI. Whether or not you allow generative AI in your classroom, be very clear about your expectations.

Write a syllabus statement that clarifies the expectations of AI use

Instructors have the discretion to allow for the use of AI or any other tools, however, are suggested to do so explicitly if they want to allow it. Stating what you expect students to use (or not use) in their work helps to answer any questions around what extra support is permittable. Depending on your pedagogical values and course expectations, consider adopting or revising one of the statements below for your syllabus. (All statements adapted from Artificial Intelligence Tools and Teaching by Iowa University’s Office of Teaching, Learning, & Technology.)

• AI is prohibited: [This course] assumes that all work submitted by students will be generated by the students themselves, working individually or in groups. Students should not have another person/entity do the writing of any substantive portion of an assignment for them, which includes hiring a person or a company to write assignments and using artificial intelligence tools like Copilot, ChatGPT and Google Bard.
• AI is allowed with attribution: Use of AI tools (Copilot is the recommended option) is permitted in this course for students who wish to use them. To adhere to our scholarly values, students must cite any AI-generated material that informed their work (this includes in-text citations and/or use of quotations, and in your reference list). Using an AI tool to generate content without proper attribution qualifies as academic dishonesty.
• AI is encouraged with certain tasks and with attribution: You can choose to use AI tools to help brainstorm assignments or projects or to revise existing work you have written. When you submit your assignment, I expect you to clearly attribute what text was generated by the AI tool (e.g., AI-generated text appears in a different colored font, quoted directly in the text, or use an in-text parenthetical citation).

For Students:

Make sure to read the course policy on AI use, addressing not only plagiarism concerns and ground rules, but also issues of AI ethics, authorship, privacy, and online safety. Discuss it with your faculty and other students on the first days of class. Continue discussing the relevant aspects of the policy throughout the semester.

Keep in mind that your faculty and peer students are likely facing major differences in faculty opinions about generative AI. Whether or not generative AI is allowed in in one course, be very clear about your understanding of the expectations within the course and AI use.

Read the syllabus statement that clarifies the expectations of AI use

Citing AI

APA Style

If ChatGPT or other AI tools have been used in research, clear description of the tool and its use should appear in the Method section of the paper (or a comparable section). For literature reviews, essays, reflective/response papers, APA suggests describing the use of tools in the introduction. You should always include the prompting language you used as well as identifying relevant text that was generated in response. Since these tools do not create content that can be replicated/retrieved by other readers, they should be treated like personal communications. The author of the tool/algorithm is treated as a source and cited in APA style. In-text citations and references follow the standard for software citation, shown in Section 10.10 of the Publication Manual.

Reference Format

• ToolAuthor. (YYYY). Name of Tool. (version) [Large language model]. URL

Parenthetical Citation Format

• (ToolAuthor, YYYY)

Narrative Citation Format

• ToolAuthor (YYYY)

Example Provided by APA Style Online 

When prompted with “Is the left brain right brain divide real or a metaphor?” the ChatGPT-generated text indicated that although the two brain hemispheres are somewhat specialized, “the notation that people can be characterized as ‘left-brained’ or ‘right-brained’ is considered to be an oversimplification and a popular myth” (OpenAI, 2023).

Reference

OpenAI. (2023). ChatGPT (Mar 14 version) [Large language model]. https://chat.openai.com/chat

MLA Style

If ChatGPT or other generative AI tool has been paraphrased, quoted, or incorporated into work in any content, the source must be cited. MLA expects writers and researchers to acknowledge all functional uses of the tool in the text of the work, as a note, or otherwise clearly stated. MLA does not recommend treating the AI tool as an author, but recommends a number of strategies for acknowledging AI tools through its existing standards and recommendations at https://style.mla.org/citing-generative-ai/.

Works Cited Format

• Description of what was generated. Name of Tool, version, ToolAuthor, date generated, URL.

Parenthetical Citation Format

• (Name of Tool, Description abbreviated)

Narrative Citation Format

• Name of Tool (Description abbreviated)

Examples Provided by MLA Style Online 

Quoted in Your Prose

When asked to describe the symbolism of the green light in The Great Gatsby, ChatGPT provided a summary about optimism, the unattainability of the American dream, greed, and covetousness. However, when further prompted to cite the source on which that summary was based, it noted that it lacked “the ability to conduct research or cite sources independently” but that it could “provide a list of scholarly sources related to the symbolism of the green light in The Great Gatsby” (“In 200 words”).

Works-Cited-List Entry

“In 200 words, describe the symbolism of the green light in The Great Gatsby” follow-up prompt to list sources. ChatGPT, 13 Feb. version, OpenAI, 9 Mar. 2023, chat.openai.com/chat.

Paraphrased in Your Prose

While the green light in The Great Gatsby might be said to chiefly symbolize four main things: optimism, the unattainability of the American dream, greed, and covetousness (“Describe the symbolism”), arguably the most important—the one that ties all four themes together—is greed.

Works-Cited-List Entry

“Describe the symbolism of the green light in the book The Great Gatsby by F. Scott Fitzgerald” prompt. ChatGPT, 13 Feb. version, OpenAI, 8 Mar. 2023, chat.openai.com/chat.

Talk about it

Additional Considerations about AI use to discuss and communicate

For Faculty:

In addition to syllabus statements, consider talking with your students about AI tools.

• Different levels of familiarity: As an emerging technology, students will have differing levels of familiarity with these tools. For instance, while Copilot, ChatGPT and Google Bard can write a grammatically correct paper or appear to solve a math problem, these tools may be unreliable and limited in scope. Discuss with students the uses and limitations of AI tools more broadly in addition to your perspective on their use in your class.
• Opt-out options: Some students may not be comfortable signing up for accounts in these tools, or may experience barriers to access if they use tools such as screen readers. Make sure you have a plan for a student who needs to opt out of using a tool.
• Connect to critical thinking skills: AI tools have many implications beyond the classroom. Consider talking with students about how to be engaged consumers of AI content (e.g., how to identify trusted sources, reading critically, privacy concerns).
• If generative AI tools are allowed in your class at all, teach your students to cite generative AI correctly.
• Ask your students to examine AI outputs critically.
  • For example: to develop your students’ critical thinking skills, ask them to generate a ChatGPT response to a question of their own choosing, and then write an analysis of the strengths and weaknesses of the ChatGPT response. Discuss the findings as a group.
• Accuracy and bias: As you encourage the use of these tools, make sure you’re talking with your students about the ways they may portray inaccurate or biased information, based on how they were made.
• Encourage students to learn more about ethical uses of AI: personalizing learning, enhancing productivity, streamlining routine tasks that do not involve concerns of privacy or authorship.

For Students:

• Connect to critical thinking skills: Many academic and professional experiences, like dissertation proposals and defenses, as well as job interviews, do not allow for the opportunity to rely on tools like AI to make up for lack of critical knowledge gained through the academic process. Please ensure you are adequately learning the critical skills and knowledge you need to in order to successfully meet the challenges of your field and the time.
• If you are collaborating with an AI tool for textual knowledge or creation of more creative content like pictures (e.g. DALL-E 3), honesty is key. Be sure to cite the use of the tools in your work.
• Outputs, accuracy and bias: As you encourage the use of these tools, please be sure to understand that these tools are only as good as the information they are trained on, and like any other research, if the data being used is bad or includes inaccurate data, so will the output. Additionally, there are ways these tools may portray inaccurate or biased information, based on how the data (or the tools) were made.
• Ethical uses of AI: Like any other tool that makes life easier, there are both positive and less positive uses for AI. The tool can powerfully improve your life, and/or take away from the meaningful parts of your experience. Please considering using the tool for personalizing your learning, enhancing your productivity, streamlining routine tasks that do not involve concerns of privacy or authorship.

Faculty can make more creative courses and mitigate risks

• Sample syllabus statements
• CAFÉ resources
• OIT suggestions for using or mitigating use of ChatGPT
• Start at course level
• Citing AI
• Talk about it
• Faculty creative courses
• AI detection
• AI and access and equity

CGU OIT does make available Turnitin as a tool for plagiarism detection, which does have a feature to include AI detection, but this add-on feature is not currently offered as part of our purchase. However, the accuracy of both plagiarism and AI detection tools are not reliable; any results from these tools should be used for nothing more than a starting point for a conversation between faculty and the student whose work is in question. Suspected use of excessive uncited content and/or Generative AI in coursework is not sufficient evidence to begin a formal Academic Integrity investigation. Instead, we recommend faculty document their expectations early and often and have open dialogs with students about the implications and responsible use of Generative AI in coursework and academia. We do recommend use of Turnitin or any other plagiarism detection tool after training on how to properly use such tools in the context of what these tools can and cannot provide.

How you interact with AI is important. As discussed in the previous section, whether you are a faculty member, a student, or a staff member, the result is only as good as the data the tool is given, both how the tool is trained and how the tool is interacted with, so long story short, what you ask AI is what you AI will give you. This is called Prompt Engineering. Prompt Engineering has become so important, the concept has sparked an entire field of research and profession, and OpenAI themselves have published a page on the topic with key considerations when prompting AI to provide you feedback.

*Note: Similarly, OpenAI has also published a page on safety best practices cautioning the output of AI tools, including having a ‘Human in the Loop,’ emphasizing having a human verifying the results of any AI interaction.

Prof. Robert Klitgaard similarly has published some work on the topic from his experience and made the work available to CGU constituents upon request