artificial intelligence illustration

Artificial intelligence and assessment: guidance for students

Advice to students on AI

Two students working on computers with their lecturer giving guidance

Artificial Intelligence (AI) has the potential to revolutionise our workplaces and lives, becoming as common as spellcheckers are today. 

In November 2022, the release of OpenAI’s ChatGPT and its conversational, convincing responses led to many debates about the ethics and limitations of AI. Beyond ChatGPT, there are paraphrasing, re-writing, and enhancement tools (including Quillbot and Grammarly) that make use of AI. 

Using an AI tool to gain an unfair advantage by misrepresenting its work as your own is Academic Misconduct and can have serious consequences. 

The University is committed to equipping our students with the ethical and critical skills they will need to use AI tools responsibly in their future careers.

UWL’s Academic Integrity Working Group (AIWG) has put together the following guide to help you understand the university’s position on AI tools and their use.

Main body

  • What do we mean?

    Academic Integrity 

    Behaving ethically, fairly, honestly, responsibly, and transparently when studying or undertaking and reporting research is essential. This includes but is not limited to: 

    • Acknowledging sources (not plagiarising). 

    • Originality and authorship (only submit work and results that are your own). 

    • Ownership (not uploading your work to non-UWL tools or giving it to other students). 

    Academic Misconduct 

    Submitting work that does not have Academic Integrity: the most common form of academic misconduct is plagiarism. It can also include using ghostwriters (including AI tools), research misconduct, and exam misconduct, as well as all other kinds of cheating to gain an unfair academic advantage. 

    Artificial Intelligence 

    Artificial Intelligence, in this context, refers primarily to natural language processing and Large Language Models (LLMs). These are systems that typically analyse billions of inputs to develop connections between them in ways inspired by the human brain. These models are then often then further refined through human feedback and training.

  • What do we do?

    Uphold Academic Integrity 

    In order to promote and uphold Academic Integrity, students can: 

    • Talk openly about Academic Integrity. 

    • Uphold and act according to those values. 

    • Take pride in submitting work that is truly your own. 

    • Understand the risks to Academic Integrity and their consequences. 

    Students should make use of formative assessments, subject librarians, the Study Support Team, and the academic experts at their school or college to develop Academic Integrity skills (including referencing and citation). 

    Students should not share work, either to online tools, or other students. Once it has been shared, it cannot be taken it back, and ownership is effectively lost. 

    Think about how students submitting work written by AI affects the meaningfulness of your degree and the other students who are submitting honestly. Ultimately, such behaviour could affect your career options. 

    Understand Expectations 

    Students are expected to: 

    • Have a clear understanding of what is and is not permitted in each assignment. 

    • Use rubrics to understand what assessment expectations. 

    • Ask their module leader if they feel unsure about expectations. 

    Students should set aside enough time to prepare, ideally aim to submit early, so that if anything goes wrong there is time to correct it. If students find it challenging to manage their time or putting together an assessment, they can speak to the course team or the Study Support Team. 

    Individual circumstances are never an excuse for cheating. The University is here to support you if something is affecting you or your studies: Speak to your Personal Tutor, Module or Course Leader, the Welfare Support Team, or the Students’ Union. 

    Know the limitations of Artificial Intelligence 

    The use of AI has clear limitations. For example, available AI tools can: 

    • Only simulate understanding. 

    • ‘Hallucinate’ to create plausible sounding but factually incorrect answers. 

    • Inherit bias and viewpoints from the training material. 

    • Use information that is not always up to date. 

    AI tools are not reliable sources! Think of them as equivalent to asking any random person on the street; the information they provide might be taken from any number of inaccurate, misleading, or misremembered sources, made to sound convincing. 

    Submitting work generated by an AI tool and misrepresenting it as your own is always Academic Misconduct. As a result, submission of such work may result in having to redo your work with a penalty to your mark or even withdrawal from the course.

Find out more