Teaching & Learning
- For student coursework and other assignments, UGA’s academic honesty guidance frames AI use as unauthorized assistance unless an instructor explicitly authorizes it in advance
- UGA also encourages instructors to clearly state acceptable and unacceptable AI uses in syllabi and assignment statements (allowing, limiting, or prohibiting AI use at the course level)
Never use Artificial Intelligence on an assignment unless it is explicitly authorized by your instructor before the assignment is turned in.
At UGA, the default rule is that students are not permitted to use AI in their coursework unless it is explicitly authorized by the course instructor.
Instructors are strongly encouraged to clearly communicate with students about acceptable and unacceptable uses of generative AI tools through syllabus and assignment acceptable use statements, as well as through in-class discussion. Whether you allow, limit, or prohibit AI use, clarity helps prevent confusion and supports academic honesty.
- UGA’s academic honesty guidance prohibits using electronic/digital devices during an exam or closed assignment unless the instructor permits it, and it specifically warns against using AI on an assignment unless explicitly authorized in advance
- UGA also provides instructor-facing guidance on using Turnitin’s AI Writing Detector, emphasizing it should not be treated as definitive proof of misconduct and that instructors must follow academic honesty reporting processes when they suspect violations
Using any cellular device, electronic device, digital device, or programmable calculator without permission during an exam or closed assignment.
Don’t access any electronic devices or notes for any reason unless your instructor explicitly says it’s allowed during an exam.
Never use Artificial Intelligence on an assignment unless it is explicitly authorized by your instructor before the assignment is turned in.
Turnitin’s AI Writing Detector is the only AI Writing Detector approved for use at UGA.
It is important that Turnitin scores not be used as a definitive measure of misconduct. Instructors should review sentence-level data from Turnitin and determine the appropriate next steps.
If you believe a violation of A Culture of Honesty may have occurred, instructors have a responsibility to report to the Office of Academic Honesty.
- UGA also cautions that AI tools may produce fabricated, incorrect, or biased content and that users must independently verify outputs
- UGA guidance describes AI as acceptable for educational support use cases such as tutoring assistance and personalized learning experiences, but for coursework the default rule is that students are not permitted to use AI unless explicitly authorized by the instructor
Educational Support: AI can be used to support educational activities, such as providing personalized learning experiences, offering tutoring assistance, and enhancing accessibility for students with disabilities.
At UGA, the default rule is that students are not permitted to use AI in their coursework unless it is explicitly authorized by the course instructor.
Please note that large language models (LLMs) may occasionally incorporate fabricated, incorrect or biased information in its output. Gemini and NotebookLM users must be prepared to independently verify all output for validity and applicability and be aware that hallucinations and bias may be present in the tool’s output.
- UGA also lists code writing as a capability/use case of institutionally provided AI tools (e.g., Copilot)
- UGA does not define a programming-assignment-specific AI rule in the provided sources; instead, it treats AI use in coursework under the general academic honesty expectation that AI use is not permitted unless explicitly authorized by the instructor
At UGA, the default rule is that students are not permitted to use AI in their coursework unless it is explicitly authorized by the course instructor.
Copilot is a conversational chat interface with several capabilities, including search and information retrieval, text generation, image creation and code writing.
Research
- For faculty/staff research-adjacent use, UGA’s general AI guidance emphasizes personal responsibility for the integrity and accuracy of work and requires users to verify AI-generated information
- For graduate theses and dissertations, UGA states that use of generative AI is prohibited as unauthorized assistance unless specifically authorized by the advisory committee within an approved scope, and if approved, the extent of AI use should be disclosed in a statement within the thesis/dissertation
As such, the use of generative AI in theses and dissertations is considered unauthorized assistance per the Academic Code of Honesty and is prohibited unless specifically authorized by members of the advisory committee for use within the approved scope. If approved by the advisory committee, the extent of generative AI usage should be disclosed in a statement within the thesis or dissertation.
Users are accountable for their work, irrespective of the AI tools used. This includes verifying the accuracy of information generated by AI and ensuring it is free from errors and biases.
- It also states that permission to access UGA data does not automatically grant permission to use that data in AI models and directs users to seek permission and guidance (starting with a supervisor)
- UGA’s acceptable-use guidance describes AI as usable for research activities including data analysis, pattern recognition, and predictive modeling, while requiring compliance with data privacy laws/policies and careful stewardship of UGA data
Research and Development: AI can be used to support research activities, including data analysis, pattern recognition, and predictive modeling.
Permission to access UGA data does not constitute permission to use that data in AI models (or for other unauthorized purposes). The user is responsible for seeking permission and guidance in using UGA data in novel ways such as in an AI tool and should begin by discussing it with their supervisor.
All data used in AI projects must be handled in compliance with the University's data privacy policies (https://eits.uga.edu/access_and_security/infosec/pols_regs/policies/dcps/ ).
- UGA’s acceptable-use guidance also points researchers to human-subjects protections where applicable and emphasizes informed consent and privacy impact assessments when using human-related data
- UGA’s research misconduct policy explicitly includes misconduct committed through the use or assistance of AI-based tools within fabrication, falsification, or plagiarism in proposing, performing, reviewing, or reporting research results
Research Misconduct means Fabrication, Falsification, or Plagiarism in proposing, performing, or reviewing Research, or in reporting Research results, whether committed by the Respondent directly or through the use or assistance of other persons, entities, or tools, including artificial intelligence (AI)-based tools.
Users of AI should adhere to applicable international, foreign, federal, and state laws and regulations, as well as USG and UGA data policies. This includes ensuring that the collection and usage of human-related data is, where applicable, based on informed consent.
Refer to UGA’s Human Research Protection Program for details when conducting research with human subjects.
Academic Integrity
- UGA’s graduate thesis/dissertation policy states that if generative AI use is authorized by the advisory committee, the extent of AI usage should be disclosed in a statement within the thesis or dissertation
- UGA Libraries guidance also directs students to cite or acknowledge AI tool use and to disclose all AI use, and the acceptable-use guidance states AI-generated content for UGA-related outputs should clearly indicate AI was used in its creation
If approved by the advisory committee, the extent of generative AI usage should be disclosed in a statement within the thesis or dissertation.
You should cite or acknowledge the use of an AI tool if it contributed to your work in any of the following ways:
Disclose all use of AI tools, even for idea generation, translation, or editing.
Users that generate content through AI (e.g., for UGA websites; advertisements; brochures; invitations; etc.) should clearly indicate the use of AI in the creation of the content.
- It also states Turnitin scores should not be used as definitive proof of misconduct and directs instructors to analyze work and, when a violation is suspected, report to the Office of Academic Honesty for a facilitated discussion process
- UGA instructor guidance states that Turnitin’s AI Writing Detector is the only approved AI writing detector for use at UGA and that instructors should not use non-supported detectors due to lack of vetting for security, FERPA compliance, and protection of student intellectual property
Turnitin’s AI Writing Detector is the only AI Writing Detector approved for use at UGA.
UGA instructors should not use AI detectors that are not supported by UGA, as these other tools have not been vetted by UGA’s information security team, for FERPA compliance, or for protection of student’s intellectual property.
It is important that Turnitin scores not be used as a definitive measure of misconduct.
If you believe a violation of A Culture of Honesty may have occurred, instructors have a responsibility to report to the Office of Academic Honesty. A facilitated discussion will be scheduled for you to talk with the student about their process for completing the assignment and whether a violation occurred.
Institutional & Administrative
- UGA prohibits using generative AI to directly assign/determine/calculate grades without final instructor review, and it prohibits using non-enterprise/public AI tools in ways that may allow student work or IP to be retained/reused or used for training without student consent
- UGA allows faculty use of generative AI for high-level grading support tasks like developing rubrics or assessment criteria, provided no student-specific or personally identifiable information is entered, and requires that AI outputs be reviewed and finalized with professional judgment
Generative AI tools may be used to support the development of grading rubrics, assessment criteria, or other high-level grading guidelines, provided that no student-specific or personally identifiable information is input into the tool.
All AI-generated outputs must be carefully reviewed, validated, and finalized using the faculty member’s own professional judgment.
The following uses of generative AI tools are not permitted:
Directly assigning, determining, or calculating grades for student work without final review by an instructor.
Using non-enterprise or publicly available generative AI tools in ways that may allow student work or intellectual property to be retained, reused, or incorporated into training data without the student’s consent.
- UGA also states that its enterprise instances of Microsoft Copilot and Google Gemini/NotebookLM do not share UGA data with outside entities or use it to train external models when accessed through a UGA account
- For UGA-related work, it states users should only use AI tools properly acquired through procurement (including EITS vetting for cybersecurity and legal compliance such as FERPA) and should not use a private copy/license with UGA data
- UGA’s acceptable-use guidance requires compliance with data privacy policies, emphasizes that access to UGA data does not equal permission to use it in AI models, and directs users to seek permission and guidance (starting with a supervisor) for novel uses such as AI tools
Permission to access UGA data does not constitute permission to use that data in AI models (or for other unauthorized purposes).
Use of AI Tools Outside of Those Officially Adopted by UGA: ... UGA users who engage with AI tools for UGA-related work should only use AI tools that have been properly acquired through the Procurement process, which includes vetting by EITS for cybersecurity concerns and compliance with state and federal laws (e.g., FERPA). A user should not obtain a private copy or private license for an AI tool and use it with UGA data.
Unlike other generative AI tools found online, Copilot is housed within UGA’s instance of Microsoft, meaning UGA data is not shared with outside entities or used to train external AI models.
Similar to UGA’s instance of Microsoft Copilot, when these tools are accessed through a UGA account, data is not shared with outside entities or used to train external models.
- Beyond this acceptable-use guidance, the provided sources do not define a broader UGA-wide AI governance structure, formal strategy, or roadmap within the extracted text
- UGA’s AI acceptable-use guidance indicates it was approved by the UGA Leadership Council on AI and by the Provost, and it frames the guidance as supporting responsible AI adoption aligned with the university’s mission/values and compliant with laws and USG/UGA policies
Approved by the UGA Leadership Council on AI on February 21, 2025
Approved by the Provost on March 5, 2025
The UGA community is encouraged to explore these important tools and to use them responsibly. However, their use must (i) align with the University's mission and values to ensure integrity and fairness in all academic and administrative processes, and (ii) comply with applicable federal and state laws and regulations, as well as with University System of Georgia (USG) and UGA policies.
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 Georgia's AI Policies
University of Georgia has defined AI policies in 12 of 12 categories, with an overall coverage score of 100%.
UGA’s graduate thesis/dissertation policy states that if generative AI use is authorized by the advisory committee, the extent of AI usage should be disclosed in a statement within the thesis or dissertation. UGA Libraries guidance also directs students to cite or acknowledge AI tool use and to disclose all AI use, and the acceptable-use guidance states AI-generated content for UGA-related outputs should clearly indicate AI was used in its creation.
UGA instructor guidance states that Turnitin’s AI Writing Detector is the only approved AI writing detector for use at UGA and that instructors should not use non-supported detectors due to lack of vetting for security, FERPA compliance, and protection of student intellectual property. It also states Turnitin scores should not be used as definitive proof of misconduct and directs instructors to analyze work and, when a violation is suspected, report to the Office of Academic Honesty for a facilitated discussion process.
UGA’s acceptable-use guidance requires compliance with data privacy policies, emphasizes that access to UGA data does not equal permission to use it in AI models, and directs users to seek permission and guidance (starting with a supervisor) for novel uses such as AI tools. For UGA-related work, it states users should only use AI tools properly acquired through procurement (including EITS vetting for cybersecurity and legal compliance such as FERPA) and should not use a private copy/license with UGA data. UGA also states that its enterprise instances of Microsoft Copilot and Google Gemini/NotebookLM do not share UGA data with outside entities or use it to train external models when accessed through a UGA account.