Using AI in Your Faculty and Staff Work at WilmU

A: It applies to all faculty and staff whenever they're using GenAI tools—things like ChatGPT, Gemini, NotebookLM, or Copilot—for WilmU work, whether that's teaching, research, or general job duties.

A: WilmU actively encourages exploring and experimenting with GenAI. The policy exists to help you do that responsibly, not to shut it down.

A: No—contact IT first, before purchasing or acquiring any AI product (paid or free) for use with University resources or data. IT and the Office of General Counsel (OGC) review vendor contracts to check for risk, and IT can point you to tools WilmU already uses to avoid duplication. Contact IT at helpdesk@wilmu.edu.

A: IT will publish a list of vetted, approved GenAI applications, along with what each is approved for. Stick to the approved scope for any tool on that list. If you have a case for a tool that isn't approved, you can request an exception through the Office of General Counsel.

A: Anything the policy defines as "Confidential Information"—a broad category that includes:

• Student records: grades, transcripts, advising notes, financial aid, papers, exams, assessments
• Health, disability, counseling, or accommodation information
• Personal Information about students, employees, applicants, alumni, donors, vendors, or research participants
• Employee personnel, compensation, benefits, discipline, or performance records
• Non-public financial, budget, payroll, tax, tuition, or vendor pricing data
• Passwords, credentials, or other cybersecurity information
• Legal, compliance, audit, or OGC communications
• Unpublished research, grant materials, or trade secrets

A: Only if a tool's contract language and security controls have been specifically reviewed and approved by IT and OGC for that purpose. Absent that, don't input Confidential Information into a GenAI tool.

A: Broadly, any information that could identify or be linked—directly or indirectly—to a specific individual, alone or combined with other available information.

A: No—always verify. GenAI output can be inaccurate, biased, or entirely made up ("hallucinated"), and it may contain copyrighted material. You're personally responsible for the accuracy of anything you publish that includes AI-generated content.

A: Watch for it, and don't rely on it. Consider whether inputs or outputs could produce a disparate impact based on protected characteristics like race, ethnicity, national origin, age, sexual orientation, or disability status. If output shows signs of bias, don't use it.

A: No. Don't hold out AI output as your own. If you quote, paraphrase, or borrow ideas from it, confirm it's accurate and doesn't plagiarize or infringe someone else's rights.

A: Be careful. Some publishers say using their licensed content with AI tools isn't allowed under their license terms. Contact the Office of Academic Affairs if you need help figuring out what's acceptable with licensed content.

A: Yes, whenever a tool offers that option, faculty and staff are expected to use it.

A: Yes. If you use GenAI to produce written materials or other work product, you need to disclose that it's based on or derived from AI.

A: Clearly disclose when written materials or other work product are based on or derived from GenAI. WilmU generally uses APA style, so follow APA guidelines when appropriate. For scholarly or research work, also follow the applicable citation, disclosure and authorship requirements of the journal, funding agency, professional society or other governing body.

APA reference list entry format: AI Company Name. (Year, Month Day). Title or description of chat [Generative AI chat]. Tool Name/Model. URL of the chat.
Example: OpenAI. (2026, August 10). Explaining supply and demand [Generative AI chat]. ChatGPT. https://chatgpt.com/share/[chat ID]. In-text, use (OpenAI, 2026) or, narratively, OpenAI (2026).

A: Yes—University-approved tools can help you create or revise lesson plans, lecture outlines, discussion prompts, rubrics, case studies, and similar materials. You're still responsible for making sure the result is accurate, accessible, pedagogically sound, and compliant with copyright and University policy.

A: State any course-specific AI rules—permitted or prohibited uses—in your syllabus or Canvas LMS course site, especially anywhere they differ from the University's student GenAI policy. Also disclose any meaningful use of AI in developing course content, materials, activities, assessments, feedback templates, or course communications. You don't need to disclose incidental things like AI-assisted editing, formatting, or grammar checks.

A: Not to make the actual judgment call. You may not use GenAI to evaluate, grade, score, rank, or make substantive academic judgments about student work—including deciding whether requirements were met, assigning grades, or making integrity determinations. You can use approved tools for limited support, but you must independently review the work, verify any AI-assisted output, and make the final call yourself.

A: Yes, everything above applies, plus some research-specific expectations below.

A: Yes—be transparent about it in your methods, acknowledgements, or wherever else is appropriate.

A: AI has been known to invent citations to papers and authors that don't exist, and to generate images for experiments that were never run. You're responsible for verifying the accuracy of anything AI-generated that ends up in your research output.

A: You still need to follow them. Many journals and funders have their own AI rules—for example, some prohibit AI-generated text, figures, or images outright. Follow those policies in addition to WilmU's.

A: No. Unpublished research, data, and other Confidential Information should not be uploaded to GenAI tools. Once entered, that material becomes part of the AI's data pool and could surface in responses to other users, and it may also jeopardize future intellectual property (IP) protection or raise privacy concerns.

A: No. That's someone else's Confidential Information, and agencies like NIH and NSF specifically prohibit using GenAI for peer review.

A: Human subjects data, interview transcripts, survey responses or identifiable information must not be entered into a GenAI tool unless the use meets the University’s confidentiality requirements and complies with Human Subject Research Committee requirements and the approved research protocol.

A: Constructive comments can be sent to ogc@wilmu.edu.

A: Reach out to the Office of General Counsel to request an exception.