The University of Edinburgh

Staff and researchers must not input identifiable patient/clinical data or any identifiable research participant data into any AI platform, including the University's institutional ELM platform.

University staff are strictly prohibited from uploading personal data or confidential university information to external AI tools, and must assume third-party providers share inputs externally.

The full Microsoft 365 Copilot suite is not provided under the University's current institutional license agreement, and ISG currently has no plans to make the full Copilot for MS365 license available, though free Copilot chat within Edge is supported.

Postgraduate research students who choose to use generative AI for aspects of assessed research (such as doctoral theses, MScR dissertations, or annual review submissions) should include a brief acknowledgment in the Declaration of Own Work at the start of the thesis or submission.

Late appeals will only be considered in extraordinary circumstances explaining and evidencing why the deadline could not be met; appeals submitted over one year after the ratified or publication date of the original decision will not be considered under any circumstances.

An appeal cannot be lodged until the Board of Examiners or decision-maker publishes its final decision in EUCLID Student View (as results on LEARN may not yet be ratified or moderated); appeals against academic decisions must be submitted promptly within 10 working days of the result being issued for undergraduate non-final years, and within 30 working days of the result being issued for undergraduate final year and all postgraduate taught and research students.

Appeals against decisions of Student Discipline Officers or the Student Discipline Committee, and appeals against decisions of Fitness to Practise Panels, must be submitted within 10 working days of the receipt of the decision.

Suspected academic misconduct at Edinburgh is investigated by School and College Academic Misconduct Officers (SAMOs / CAMOs), while formal grade penalties are determined by Boards of Examiners; allegations relating to research undertaken by students are investigated using the Academic Misconduct Investigation Procedure rather than staff research misconduct procedures.

Presenting generative AI outputs as your own original work in assessed work constitutes academic misconduct at the University of Edinburgh.

Using an AI translator to convert assessed work to English before submission is treated as false authorship and constitutes academic misconduct because English is the language of teaching and assessment.

Submitting an assessment which includes elements of AI-generated text without acknowledgment constitutes academic misconduct.

Submitting an assessment which includes AI-generated images, audio or video without acknowledgment constitutes academic misconduct.

Submitting an assessment which includes AI-generated mathematical formulae, reasoning, or computer code without acknowledgment constitutes academic misconduct.

Citing and referencing AI-found sources in assessed work without reading and verifying them constitutes academic misconduct.

For systematic or large-scale non-research adoption of AI or introduction of AI into existing platforms, staff must follow the approval process via the ISG Ethics Board, frame decisions using AI Impact Assessment guidance, and complete required impact assessments, including DPIA (Data Protection Impact Assessment) and EQIA (Equality Impact Assessment) where appropriate / as required (while research AI uses follow College Research Ethics Boards).

Presenting AI-generated content as one's own work is regarded as academic misconduct at Edinburgh; all assessed work must be the student's own original work.

Students must acknowledge any use of GenAI before submitting assessed work.

GenAI permissions at Edinburgh are set at assessment and course level; restrictions should be stated in the assessment task and students must consult their Course Organiser when unclear.

The University of Edinburgh's central AI guidance page presents ELM as the University's central gateway to safer access to generative AI for staff and students.

Edinburgh's Registry guidance says students will not be penalised for using GenAI provided they stay within the stated acceptable uses and adhere to assessment-specific restrictions.

ELM (Edinburgh access to Language Models) is listed for The University of Edinburgh in an official university AI tools source. Derived availability: allowed. Derived endorsement type: officially endorsed.

Llama in ELM is listed for The University of Edinburgh in an official university AI tools source. Derived availability: allowed. Derived endorsement type: self hosted system.

Microsoft Copilot is listed for The University of Edinburgh in an official university AI tools source. Derived availability: conditionally allowed. Derived endorsement type: institutionally licensed or procured.

ELM (Edinburgh access to Language Models) is listed for The University of Edinburgh in an official university AI tools source. Derived availability: allowed. Derived endorsement type: self hosted system.

The University of Edinburgh says students must not use third-party AI-based translation apps in class.

The University of Edinburgh says using an AI agent within Learn or any university learning or assessment platform requiring login is not permitted.