University policy
Based on the policy of GenAI usage, here's a detailed summary of the main policies and approaches to generative AI across top 100 universities (based on the ARWU ranking).
Policies
General Stance
-
Most universities recognize the potential of AI to enhance learning, teaching, and research.
-
They encourage responsible and ethical use of AI tools.
-
Many emphasize the need for transparency when using AI.
Use in Coursework
-
Policies often vary by course, discipline, and instructor.
-
Common approaches include:
a) Prohibiting AI use without explicit permission
b) Allowing limited use with proper acknowledgment
c) Encouraging AI use as a learning tool
Academic Integrity
-
Submitting AI-generated work without acknowledgment is generally considered academic misconduct.
-
Students are typically held responsible for the accuracy and integrity of their work, even if AI-assisted.
-
Many universities require students to disclose AI use in their assignments.
Citation and Acknowledgmen
-
When AI use is permitted, students are often required to cite or acknowledge it.
-
Some universities provide specific guidelines on how to cite AI tools.
Data Privacy and Security
-
Many universities caution against inputting sensitive or confidential information into public AI tools.
-
Some provide approved AI platforms for handling university data.
AI Literacy
-
Many universities are developing programs to educate students and faculty on AI use.
-
This includes understanding AI's capabilities, limitations, and potential biases.
Assessment Design
-
Instructors are encouraged to design assignments that can't be fully completed by AI alone.
-
Focus on higher-order thinking skills and process-based assessments.
AI Detection
-
Many universities caution against over-reliance on AI detection tools due to high false-positive rates.
-
Some advise against using these tools as the sole basis for academic misconduct accusations.
Research Use
-
Generally, AI use in research is allowed but must be disclosed.
-
Researchers are responsible for verifying AI-generated content.
-
AI tools are not considered authors on academic papers.
Institutional AI Tools
-
​Some universities provide their own AI tools or approved platforms for student and faculty use.
Equity Considerations
-
Some policies address the need to ensure equal access to AI tools for all students.
Continuous Policy Evolution
-
Many universities acknowledge that their AI policies will need regular updates as the technology evolves.
Discipline-Specific Approaches
-
Some universities allow individual departments or faculties to set more specific AI use guidelines.
AI in Grading
-
It can be used for exams with objective answers (multiple choice exams).
-
The use of AI in grading, generally requires human oversight and final approval.
-
Instructors take the responsibility.
Copyright and Intellectual Property
-
Many policies remind users to respect copyright when using AI, especially when inputting course materials.
Examination Policies
-
AI use is typically prohibited in examinations unless explicitly allowed.
Note
Example university policies
The summary above synthesizes common patterns across top universities. Here are a few individual schools' public policies, if you want to see how a specific institution frames its own rules:
-
MIT — AI and Education: Ad Hoc Committee report (August 2026). Focused on teaching, learning and assessment; recommends course-level AI policies and redesigned assessments rather than one uniform rule.
-
Stanford — Creating your course policy on AI. Default is that AI use counts as unauthorized assistance unless an instructor explicitly permits it.
-
Harvard — Generative AI Guidelines. University-wide guidance centers on data privacy and confidentiality; detailed classroom rules are set by individual schools.
-
University of Oxford — Generative AI at Oxford. First UK university to give all staff and students free ChatGPT Edu access (September 2025), paired with training on safe use.
-
University of Cambridge — Using Generative AI. Students may use GenAI for personal study and research but must disclose significant, unrevised use in assessed work.
-
Imperial College London — Generative AI Principles. Requires AI-generated content to be attributed and cited like any third-party source.
-
Copenhagen Business School — AI Hub @CBS. Moved from prohibition to structured integration; students must declare GenAI use in written assignments.
Please refer to each university's own policy page for the current, full text — these summaries may be updated by the schools at any time.