Academic Integrity and Use of AI
Table of contents
Academic integrity is a fundamental requirement for all assessed work at the Department of Computational Linguistics, including written work, programming projects, bachelor's and master's theses. Students must complete their work independently, use only permitted aids, appropriately acknowledge sources and contributions by others, and transparently document the use of generative AI where required.
Use of Generative AI
The Department of Computational Linguistics supports the responsible and critical use of generative AI in studying, writing, programming, and research. Learning how to use these technologies appropriately is an important part of studying Computational Linguistics and Language Technology.
At the same time, submitted work must demonstrate the student's own knowledge, skills, and intellectual contribution. Students remain responsible for all submitted content, including AI-assisted text, code, analyses, and other material.
Whether and to what extent generative AI may be used depends on the particular assessment. The applicable rules are communicated by the instructor or supervisor. For seminar papers and theses, students must in particular follow the Department's Guidelines on the Use of Text Generation Models for Seminar Papers and Theses.
For assessed work where a declaration is required, students must complete the Department's Declaration of Independent Work and Use of Generative AI and submit it together with their work. The declaration records whether and how generative AI was used and requires more detailed information for substantial AI assistance. The declaration is available below as a PDF and Word document.
Students must be able to explain their submitted work and their own contribution to it. Instructors and supervisors may ask students to explain, for example, selected passages, code, analyses, methodological decisions, or the use and evaluation of AI-generated contributions.
The use of unauthorized aids, including generative AI where its use is not permitted, or failure to disclose the use of generative AI as required, may constitute unfair conduct (unlauteres Verhalten). Such cases are handled in accordance with the applicable regulations and procedures of the University of Zurich and the Faculty of Arts and Social Sciences.
Plagiarism and Other Forms of Unfair Conduct
Plagiarism is one form of unfair conduct. It occurs when another person's research findings, ideas, texts, or other intellectual contributions are presented as one's own without appropriate acknowledgement. Under the UZH Disciplinary Regulations, plagiarism may constitute a disciplinary offense whether it is committed intentionally or negligently.
Plagiarism includes, for example, submitting work written by another person under one's own name (ghostwriting), reusing the same work or parts of it for different assessments without appropriate disclosure (self-plagiarism), translating material from another language without acknowledging the source, copying text without proper citation, or paraphrasing another person's work without appropriate acknowledgement. Sources must be cited in a way that makes clear which passages, ideas, or other contributions have been taken from them.
Unfair conduct is not limited to plagiarism. It also includes the use of unauthorized aids and other forms of fraud or dishonesty in assessed work.
If plagiarism or another form of unfair conduct is established, this can have consequences for the assessment concerned. In addition, a disciplinary procedure may be initiated under the UZH Disciplinary Regulations. The disciplinary consequences depend on the circumstances and seriousness of the individual case.
Plagiarism-detection software may be used to identify textual similarities. A similarity report is an aid to assessment and does not by itself establish that plagiarism has occurred; the identified similarities and their context must be evaluated.
Students must also observe any additional requirements specified by the Faculty of Arts and Social Sciences, the Department of Computational Linguistics, the relevant study program, and the instructor or supervisor responsible for the assessment.
Further information on plagiarism, unfair conduct, and the legal framework:
Recording meetings and classes
Classes
Do not make audio or video recordings of classes without the consent of the lecturer. Unauthorized recording of university classes or events may constitute a disciplinary offense at UZH.
Meetings and supervision sessions
Do not make audio or video recordings of meetings, supervision sessions, or other conversations without the consent of everyone involved. This also applies when a recording is made in order to create an AI-generated transcript or summary.