Dissertation or thesis 

Image generated with Flux (Pro)

A written manuscript presenting the findings of a substantial original research project. Includes projects completed as part of an honours program.

Type = Dissertation or thesis

Category = Production and creation - open

Description

A dissertation or thesis is a substantial and formal piece of academic writing that is based on original research. It is typically a requirement for the completion of an honours thesis or a master’s degree. Students typically undertake this assessment over an extended period of time, depending on their degree program. The primary purpose of a dissertation or thesis is to conduct an in-depth exploration of a specific topic within a chosen field of study, with the expectation that the student will contribute new knowledge, perspectives, or insights to the academic discourse. The work should showcases the student’s expertise in their subject area, their ability to conduct independent research, engage in critical thinking, and communicate complex ideas effectively.  Students will usually complete this assessment under the guidance of academic supervisors.

Special notice for Honours, Masters, and PhD students and all researchers

Research at the University of Sydney must adhere to the guidelines set out by the University. Researchers (including research students) need to complete the specialised training developed by the Research Integrity and Ethics Administration Office, which overrides other advice below.

Research students: Enrol for the Responsible Use of Generative AI for Researchers module

IMPORTANT: Honours students, HDR students, other students who are conducting research activities, and University researchers must comply with expectations of responsible research practice. The University's Research Code of Conduct 2023 sets out the University's policy around responsible research practice, including integrity in research and our obligations under various codes. The Australian Code for the Responsible Conduct of Research 2018 Links to an external site. also establishes a framework for the responsible conduct of high quality research. Additionally, bodies such as the NHMRC and the ARC have published policies and guidelines on the use of generative AI in grant applications and peer review that must be complied with (for example, see the NHMRC's policy on generative AI in grant applications and peer review Links to an external site., and the ARC's policy on generative AI in grant programs Links to an external site.). Journals, funding bodies, and other entities may also have additional expectations and policies around the use of generative AI and these must be adhered to.

IMPORTANT: Researchers at the University must be aware of our guidelines for researchers Links to an external site.. This includes risks around sensitive information (including research data, identifiable information), research integrity (such as accuracy of AI outputs, accountability, and attribution), and research ethics (including bias and fairness in AI models). The guidelines contain important examples of safe use of generative AI in research, including not uploading unpublished research findings and sensitive data, verifying AI outputs, observing confidentiality, and checking policies of journals, funding bodies, commercial contracts, etc.

Meaningful engagement with AI

Keeping all of the above in mind and ensuring compliance, AI might be integrated into several stages of a dissertation or thesis, offering students an opportunity to learn to use AI tools to enhance their research capabilities rather than replace them. Below are some examples.

• Literature review: AI can facilitate the summarisation of existing research and help identify key scholarly papers relevant to the study.
• Writing: AI tools may help students with structuring and articulating their findings clearly and concisely, ensuring that key results are effectively communicated using appropriate technical language.
• Methodology: AI tools can provide valuable feedback on students' research methodology by evaluating its feasibility, validity, and overall robustness. Additionally, AI can prompt students to consider crucial factors such as variable control, ethical considerations, bias minimisation, and error reduction.
• Data analysis and visualisation: AI can assist students by generating code to automate data cleaning, conduct analyses, and create visualisations. However, students must apply their technical expertise to critically assess AI-generated results, ensuring the analysis is accurate and appropriately conducted.

Supervisors should discuss with students relevant ethical and legal implications of sharing data with AI tools, emphasising the importance of data privacy, confidentiality, and compliance with existing regulations such as grant and journal policies.

Educators should then provide students guidance on the effective use of AI tools in research, emphasising that AI can introduce errors, produce invalid analyses, or lead to inaccurate conclusions, especially in emerging research fields where AI knowledge may be incomplete or outdated. They should make clear that students are expected to apply their own reasoning, discipline-specific knowledge, and contextual understanding when using AI-generated outputs. Supervisors should also stress that students bear full responsibility for their final work and must ensure the accuracy and reliability of their results.