Teaching & Learning
- Use of AI in graded academic submissions is treated as an academic integrity issue when it is prohibited by the instructor/department or otherwise considered inappropriate for the academic submission
- The Honor Code explicitly includes “prohibited or inappropriate use” of Artificial Intelligence (AI) within “Unauthorized Use.” Expectations and permissibility can depend on course and/or department/program policies
Unauthorized Use: Any actual or attempted use or possession of resources prohibited by the instructor(s) or those that a reasonable person would
consider inappropriate under the circumstances for academic submissions,
including prohibited or inappropriate use of the internet or Artificial
Intelligence (AI).
Depending on how the student responds, you can decide how to proceed based on your course and/or department/program policies.
- The Honor Code’s definition of cheating also includes “copying answers on a test” and claiming credit for another’s work contrary to instructor/department instructions
- Using AI in exams/assessments is covered under academic integrity rules when such resources are prohibited by the instructor or are otherwise inappropriate for the circumstances of the academic submission
Cheating: Any actual or attempted effort to engage in falsification of
academic materials, claiming credit for another’s work contrary to
instructor/department instructions, such as copying answers on a test, and/or
Unauthorized Use: Any actual or attempted use or possession of resources
prohibited by the instructor(s) or those that a reasonable person would
consider inappropriate under the circumstances for academic submissions,
including prohibited or inappropriate use of the internet or Artificial
Intelligence (AI).
Research
- It also notes risks that generative AI may create non-existent citations and may include copyrighted material from other authors
- University data governance guidance warns that AI-generated content may be misleading or inaccurate and that users are responsible for reviewing AI-generated content for accuracy and ownership
Don't assume public data is free of intellectual property rights. AI-generated content may be misleading or inaccurate. Generative AI technology may create citations to content that does not exist. Responses from generative AI tools may contain content and materials from other authors and may be copyrighted. The tool user is responsible for reviewing the accuracy and ownership of any AI-generated content.
Academic Integrity
- The Honor Code defines plagiarism to include failing to include appropriate citations in academic and educational submissions
- The provided sources do not explicitly define an AI-specific disclosure/citation requirement (e.g., requiring AI use statements), but they do state that failing to include appropriate citations constitutes plagiarism
Plagiarism: Any representation of another’s work or ideas as one’s own in
academic and educational submissions, including, failure to include
appropriate citation(s).
- It also describes Turnitin’s AI Writing Detector and notes issues such as false positives and that internal OTL tests confirmed problems
- The Honor Code establishes AI misuse (when prohibited or inappropriate) as part of “Unauthorized Use,” making it subject to academic integrity enforcement processes
- The Office of Teaching & Learning advises faculty not to rely on AI checking software to confirm suspicions of AI-related academic dishonesty, noting that such tools can be unreliable
Don’t rely on AI checking software to confirm your suspicions. Although Turnitin and other companies have programs that check for AI-generated writing, many of these programs are, at best, in their earliest stages, or at worse, unreliable.
And speaking of the uses and limitations of technology, let’s turn, as it were, to Turnitin’s latest feature: its AI Writing Detector.
On its website, Turnitin claims that its false positive rate is “less than 1% for documents with 20% or more AI writing,” which is a change from their previous claim of a less than 1% false positive rate. Furthermore, they go on to say that “there is a 4% likelihood that a specific sentence highlighted as AI-written might be human-written.” Our own tests within the OTL have confirmed this issue.
Unauthorized Use: Any actual or attempted use or possession of resources prohibited by the instructor(s) or those that a reasonable person would
consider inappropriate under the circumstances for academic submissions,
including prohibited or inappropriate use of the internet or Artificial
Intelligence (AI).
Institutional & Administrative
- The Office of Teaching & Learning advises instructors to define expectations in syllabus policies regarding whether and how students can use AI tools
- It also provides guidance for faculty use of Turnitin and cautions against relying on AI detection to confirm suspicions, but does not define a university-wide policy on faculty use of AI for grading/feedback/administrative work in the provided sources
Update your syllabus policies
Be very clear about your expectations regarding students’ work. Consider syllabus statements indicating whether and how AI tools can be used.
Don’t rely on AI checking software to confirm your suspicions.
- DU policy requires that all new software acquisitions (including free and trial) be initiated through the Technology Solution Center
- DU data security standards and data classification policies define classification levels (Public, Internal, Confidential, Restricted) and outline protection and sharing requirements for confidential data
- DU IT identifies Microsoft Copilot Chat as the recommended generative AI chat tool when prompts contain institutional information and states it is strongly recommended over other AI tools if a prompt includes DU data, citing commercial data protection and that prompts/responses are not saved and are not used to train underlying models
Microsoft Copilot Chat allows for authenticated access to Generative AI ChatGPT technologies. This is the recommended tool when users have questions or prompts to ChatGPT that contain institutional information.
Keeps DU Data Secure. It is strongly recommended to use Microsoft Copilot instead of other AI tools if your prompt includes any DU data. Copilot includes commercial data protection when users sign in with their DU account. This means user and organizational data are protected, prompts and responses are not saved, Microsoft has no eyes-on access and chat data isn’t used to train the underlying large language models.
This policy outlines the security measures for protecting data classified as Public, Internal, Confidential, and Restricted. See Policy IT 13.10.051 – Data Classification.
University Data is classified as Public, Internal, Confidential and Restricted.
Sharing: Confidential data may be shared among university employees according to a well-defined business process approved by the data steward. It may be released publicly only according to well-defined business processes and with the permission of the data steward.
All new software acquisitions (including free and trial) and purchase of
additional modules for existing software shall be initiated through the
Technology Solution Center.
- The University of Denver, through its Provost's Office, is actively developing its AI strategy and encouraging campus-wide engagement
- This is demonstrated through initiatives like the "Winter 2025 Provost Conference: AI and the Public Good" and a "Provost Conversation Series" focused on critical topics facing higher education (including AI)
- While a formal AI strategic plan or dedicated governance committee is not defined in the provided sources, these Provost-led initiatives indicate a university-wide approach to exploring and shaping AI policy and practice
Join us for the Winter 2025 Provost Conference: AI and the Public Good! Join us to explore and share ideas for how we teach and conduct our research and other work in the age of generative artificial intelligence.
The Provost Conversation Series is one of the ways that the provost engages with the campus community on critical topics facing higher education.
DocuMark: Responsible AI Use for Academic Integrity
Knowing your institution's AI policy is step one. DocuMark helps enforce it fairly by empowering universities to manage AI-generated content, prevent cheating, and support student writing through responsible AI use.
Common Questions About University of Denver's AI Policies
University of Denver has defined AI policies in 8 of 12 categories, with an overall coverage score of 67%.
The Honor Code defines plagiarism to include failing to include appropriate citations in academic and educational submissions. The provided sources do not explicitly define an AI-specific disclosure/citation requirement (e.g., requiring AI use statements), but they do state that failing to include appropriate citations constitutes plagiarism.
The Office of Teaching & Learning advises faculty not to rely on AI checking software to confirm suspicions of AI-related academic dishonesty, noting that such tools can be unreliable. It also describes Turnitin’s AI Writing Detector and notes issues such as false positives and that internal OTL tests confirmed problems. The Honor Code establishes AI misuse (when prohibited or inappropriate) as part of “Unauthorized Use,” making it subject to academic integrity enforcement processes.
DU IT identifies Microsoft Copilot Chat as the recommended generative AI chat tool when prompts contain institutional information and states it is strongly recommended over other AI tools if a prompt includes DU data, citing commercial data protection and that prompts/responses are not saved and are not used to train underlying models. DU data security standards and data classification policies define classification levels (Public, Internal, Confidential, Restricted) and outline protection and sharing requirements for confidential data. DU policy requires that all new software acquisitions (including free and trial) be initiated through the Technology Solution Center.