Teaching & Learning
- Use of generative AI in coursework is governed primarily by instructor discretion
- Submitting AI-generated material as one's own work is treated as academic dishonesty unless the instructor has specifically authorized that use
- The college states that some faculty may prohibit AI entirely, while others may permit it with limits or for specific purposes; students must understand and follow the policy for each course or assignment
Faculty members are encouraged to define and communicate with students the extent to which use of generative AI is permissible in the classroom and on assignments. In some classes, the use of generative AI may be prohibited, while in others, use may be permitted or even encouraged in specific ways.
Students should understand and follow the policy for each class or assignment and seek clarification from the instructor if they are unsure of the acceptable use of generative AI in a particular context.
Submitting the work of another person or resource as one’s own, allowing another person or resource to alter or revise one’s work, or using artificial intelligence or machine learning tools beyond what is allowed by the instructor in a given course all count as academic dishonesty.
Use of AI should be in service of your writing and learning, but the writing and learning should still be your own.
GenAI can be a useful tool for writers throughout the writing process if used effectively and ethically. However, students should communicate with instructors about when and how they may use AI in their work.
- The university does not provide a separate college-wide AI rule specifically for exams or quizzes in the provided sources
- Instead, exam-related AI use falls under course-level rules and academic integrity expectations, meaning use beyond what an instructor allows would be misconduct
Faculty members are encouraged to define and communicate with students the extent to which use of generative AI is permissible in the classroom and on assignments. In some classes, the use of generative AI may be prohibited, while in others, use may be permitted or even encouraged in specific ways.
Students should understand and follow the policy for each class or assignment and seek clarification from the instructor if they are unsure of the acceptable use of generative AI in a particular context.
Submitting the work of another person or resource as one’s own, allowing another person or resource to alter or revise one’s work, or using artificial intelligence or machine learning tools beyond what is allowed by the instructor in a given course all count as academic dishonesty.
- Colby permits AI use for learning support when used appropriately
- Students are also told to consult instructors about acceptable use in their courses
- The guidance recommends using AI to ask questions, summarize concepts, identify points needing clarification, generate examples, and support brainstorming or revision, while keeping the student's own learning and authorship central
Chat with AI while taking notes from reading or class discussion and ask questions. If a point from lecture isn’t making sense, ask AI to answer the question in simpler terms or from a different angle. You can ask AI to summarize and to identify the point where understanding breaks down.
As with the other strategies, it’s often best to reserve use of AI for after a draft has already been developed and just for points where you are really stuck. It can also be useful for examples and sentence-level revision concerns.
Use AI in the brainstorming process, asking for ideas and examples.
If you want to use AI at any stage of the writing process, ask your instructor about what would be considered acceptable use.
- The provided sources do not set a distinct university-wide policy for AI-assisted coding or programming assignments
- Any use of AI tools for coding would therefore be governed by instructor permission and the general academic integrity rule against AI use beyond what the instructor allows
Faculty members are encouraged to define and communicate with students the extent to which use of generative AI is permissible in the classroom and on assignments. In some classes, the use of generative AI may be prohibited, while in others, use may be permitted or even encouraged in specific ways.
Submitting the work of another person or resource as one’s own, allowing another person or resource to alter or revise one’s work, or using artificial intelligence or machine learning tools beyond what is allowed by the instructor in a given course all count as academic dishonesty.
Research
- Colby does not publish a dedicated policy specifically for AI use in research manuscripts or dissertations
- Students producing research writing are subject to the same instructor-permission framework and academic integrity rules that govern all coursework
- However, the General Counsel's AI guidance applies broadly to faculty and staff work, requiring that AI outputs be independently verified before being relied upon, and prohibiting input of regulated or high-risk data into AI systems without prior approval
In any use of generative AI, it is imperative that users independently verify all outputs before relying on them.
Without prior, express approval from the Office of General Counsel or Information Security and the use of required safeguards, users may not input regulated, confidential or high-risk data, as designated by Colby's data classification policy, into generative AI.
Faculty and staff should not use generative AI to make discretionary decisions, or in any instance where professional judgment is required, without significant independent review.
- The guidance also warns that AI-generated outputs must be validated by a human user before being relied upon
- Colby's legal guidance permits faculty and staff to use certain AI tools with institutional data only when the data are classified as low or moderate risk and the tool has approved safeguards
- High-risk, confidential, or regulated data may not be entered into AI systems without explicit approval from the Office of General Counsel or Information Security and protection of the data to the university standard
Colby faculty and staff should only use generative AI with institutional data where the data is low to moderate risk, and where the generative AI has proper safeguards in place.
Without prior, express approval from the Office of General Counsel or Information Security and the use of required safeguards, users may not input regulated, confidential or high-risk data, as designated by Colby’s data classification policy, into generative AI.
In any use of generative AI, it is imperative that users independently verify all outputs before relying on them.
- Colby does not publish a dedicated AI policy for grant proposals or IRB submissions
- Researchers working with human-subjects or other regulated data must therefore ensure compliance with the data classification policy before using any AI tool in their research workflows
- The General Counsel's guidance is the closest applicable institutional framework, requiring independent verification of all AI outputs before reliance and prohibiting input of regulated or confidential data into AI systems without express approval from the Office of General Counsel or Information Security
In any use of generative AI, it is imperative that users independently verify all outputs before relying on them.
Without prior, express approval from the Office of General Counsel or Information Security and the use of required safeguards, users may not input regulated, confidential or high-risk data, as designated by Colby's data classification policy, into generative AI.
Colby faculty and staff should only use generative AI with institutional data where the data is low to moderate risk, and where the generative AI has proper safeguards in place.
Academic Integrity
- More broadly, students must follow course-specific AI rules and clarify expectations with the instructor
- Colby requires transparency about AI use when a student incorporates AI-generated text into submitted work
- The Writers' Center says such use should be explicitly acknowledged, and students are directed to ask instructors about preferred citation or documentation practices
Do note, however, that if you use AI-generated language in your text, this must be acknowledged clearly in your final product.
You should also ask your instructor if they have specific requirements for AI citation or documentation.
Students should understand and follow the policy for each class or assignment and seek clarification from the instructor if they are unsure of the acceptable use of generative AI in a particular context.
- Unauthorized or undisclosed AI use is enforceable as academic dishonesty under Colby's academic integrity policy
- The Faculty Handbook (August 2025) is the primary institutional reference for faculty enforcement procedures, and no specific AI detection software policy is defined in the available sources
- Using AI tools beyond what an instructor permits counts as academic dishonesty, with sanctions ranging from a warning to failure in the course; severe or repeated violations may be referred to the dean of the faculty for additional disciplinary action
Submitting the work of another person or resource as one's own, allowing another person or resource to alter or revise one's work, or using artificial intelligence or machine learning tools beyond what is allowed by the instructor in a given course all count as academic dishonesty.
A student found responsible for violating the academic integrity policy can expect a sanction from the faculty member that can range from a warning to failure in the course. In the case of severe or repeated violations, a faculty member may also choose to refer the student's case to the dean of the faculty or designee, who can choose to take additional disciplinary action.
Institutional & Administrative
- Faculty are also encouraged to communicate their classroom AI rules clearly to students
- Colby permits faculty and staff to use generative AI in their work, but requires human oversight and care with professional judgment
- The legal guidance says AI outputs must always be independently verified, and faculty are specifically told not to rely on AI to assess student work or assign grades without review
In any use of generative AI, it is imperative that users independently verify all outputs before relying on them.
Faculty and staff should not use generative AI to make discretionary decisions, or in any instance where professional judgment is required, such as evaluation of student work or assigning grades, without significant independent review.
Faculty members are encouraged to define and communicate with students the extent to which use of generative AI is permissible in the classroom and on assignments.
- Colby restricts AI use based on data classification and tool safeguards, as detailed in the General Counsel's AI Guidance Document
- Users are also prohibited from accepting legal terms or certifying compliance on behalf of the college unless specifically delegated contracting authority
- Faculty and staff may use AI only with low- to moderate-risk institutional data and only when the AI system has proper safeguards in place; regulated, confidential, or high-risk data require prior express approval from the Office of General Counsel or Information Security along with required protections
Colby faculty and staff should only use generative AI with institutional data where the data is low to moderate risk, and where the generative AI has proper safeguards in place.
Without prior, express approval from the Office of General Counsel or Information Security and the use of required safeguards, users may not input regulated, confidential or high-risk data, as designated by Colby's data classification policy, into generative AI.
Finally, some generative AI requires review and acceptance of legal terms. Unless they are specifically delegated contracting authority by Colby, users should not accept legal terms or certify compliance with legal obligations on behalf of the college or themselves in relation to institutional use of the generative AI.
- Colby has established an institutional AI governance framework anchored by the General Counsel's AI Guidance Document and the Academic Integrity Faculty Handbook (both updated August 2025)
- No dedicated AI task force or long-range AI strategy roadmap is explicitly described in the available sources, but the framework reflects a cautious, human-oversight-centered approach to generative AI across teaching, research, and administrative functions
- Oversight responsibilities are distributed across the Office of General Counsel and Information Security for data governance and legal compliance, the Center for Teaching and Learning for pedagogical guidance, and individual faculty for course-level rule-setting
In any use of generative AI, it is imperative that users independently verify all outputs before relying on them.
Faculty members are encouraged to define and communicate with students the extent to which use of generative AI is permissible in the classroom and on assignments.
Without prior, express approval from the Office of General Counsel or Information Security and the use of required safeguards, users may not input regulated, confidential or high-risk data, as designated by Colby's data classification policy, into generative AI.
Faculty and staff should not use generative AI to make discretionary decisions, or in any instance where professional judgment is required, such as evaluation of student work or assigning grades, without significant independent review.
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 Colby College's AI Policies
Colby College has defined AI policies in 12 of 12 categories, with an overall coverage score of 100%.
Colby requires transparency about AI use when a student incorporates AI-generated text into submitted work. The Writers' Center says such use should be explicitly acknowledged, and students are directed to ask instructors about preferred citation or documentation practices. More broadly, students must follow course-specific AI rules and clarify expectations with the instructor.
Unauthorized or undisclosed AI use is enforceable as academic dishonesty under Colby's academic integrity policy. Using AI tools beyond what an instructor permits counts as academic dishonesty, with sanctions ranging from a warning to failure in the course; severe or repeated violations may be referred to the dean of the faculty for additional disciplinary action. The Faculty Handbook (August 2025) is the primary institutional reference for faculty enforcement procedures, and no specific AI detection software policy is defined in the available sources.
Colby restricts AI use based on data classification and tool safeguards, as detailed in the General Counsel's AI Guidance Document. Faculty and staff may use AI only with low- to moderate-risk institutional data and only when the AI system has proper safeguards in place; regulated, confidential, or high-risk data require prior express approval from the Office of General Counsel or Information Security along with required protections. Users are also prohibited from accepting legal terms or certifying compliance on behalf of the college unless specifically delegated contracting authority.