Use of generative artificial intelligence for students

Introduction

Artificial intelligence (AI) touches our lives on a daily basis, even though we may not be fully aware of it. Facial recognition to unlock your smartphone, predictive text in messaging apps, auto complete suggestions on Google search, online customer support chatbots, recommendations from Netflix or Amazon on what to watch or buy next and what you see as you swipe through your social media feeds. These are all reliant on AI. 

More recently there has been significant attention given to generative AI (GAI) tools. These tools can generate writing on pretty much any topic, generate code and presentations, artwork, graphics, video and audio. There are hundreds of GAI tools such as ChatGPT, Google Bard, Claude, Perplexity.ai, DALL-E, the list is endless. These tools are developing and evolving at a rapid rate and have the potential to be transformative. It is essential that our staff and students have the skills and knowledge to navigate the world of GAI and understand how it can be applied. 

It is against this backdrop of growing interest in the adoption of GAI that we have prepared this guidance to help our student community understand how it can be used appropriately. Our aim is to help students appreciate both the limitations of GAI and the ethical and data implications of using these technologies. It is not feasible for the University to impose a blanket restriction on the use of GAI, rather we seek to promote the discerning, transparent and ethical use of it. Further, as GAI is increasingly integrated into commonly used digital tools we will regularly review and update this guidance to ensure it remains up to date.

Using GAI to support your learning

One of the attractions of GAI technologies is their potential to support learning as they respond to specific prompts and questioning. Whilst a Google search generates a list of websites with content relevant to your search prompt, GAI will generate a written answer in response to your question that simulates natural conversation. The content it generates can help provide overviews of topics, concepts, theories etc., much as a Wikipedia article might help you develop your initial understanding of a topic. However, just as with Wikipedia it is important to appreciate the limitations of GAI as well as some of the risks associated with its use.

Potential uses of GAI

Some examples of how you might use GAI to support your learning include:

• Supporting research into a topic or concept to help develop initial understanding and insight. The answers that GAI presents are based on material which can be found on the internet.
• Help with drafting ideas or suggesting how you might structure your own written materials.
• Generating graphics, images and other visual content to support a presentation, or for inclusion in a report.
• Summarising longer texts and documents to help you check your own understanding of the key messages and concepts presented.
• Generating questions from an article/text to help check your understanding and recall of key concepts and learning points.
• Using it as a conversational or debating partner to develop your ideas and thinking.
• Helping to improve your grammar or writing structure, this may be particularly helpful if English is a second language. 
• Debugging code.
• Helping to keep you motivated and overcome writer’s block by asking it to generate initial ideas that you can reflect and build on.

Limitations of GAI

It is important to appreciate that artificial intelligence is not the same as human intelligence. AI tools do not understand the content they generate and nor can they validate the accuracy of it. These tools are trained on content and rely on pattern recognition to predict what strings of words and sentences should be generated next. Key points to be aware of include:

• Whilst the outputs of GAI can appear to be well written and credible, AI tools do get things wrong and therefore you cannot rely on them for factual accuracy.
• It is not uncommon for AI tools to make up references and citations that are completely fictitious, that don’t exist. You won’t find these articles or texts in our Library collections or the wider internet. These fake references are frequently referred to as hallucinations.
• GAI tools tend to perform better in subjects which are widely written about and do more poorly in specialist areas. The data models on which they are trained are also not up-to-date.
• AI can replicate human bias and discrimination, perpetuating particular stereotypes and perspectives associated with the Global West.

It is important to review and check the content that GAI presents. It can sound convincing and plausible, so it is essential to apply your critical thinking skills and check the accuracy of what it presents. 

Also remember that your lecturers will have curated resource lists of textbooks, journal articles, web-based articles, videos and collections of artefacts that are accessible via your modules in My Dundee. These are all reliable and have been selected to support the learning outcomes of your programme. Our Library subject guides also provide more details of how to search and access discipline specific collections and databases.

Impact of overreliance on GAI tools

One of the concerns of relying too heavily on or overusing GAI is that it will limit your own learning and the development of core skills which many employers are looking for in graduates. Skills such as problem solving, critical thinking, analytical skills, critical reflection. Such concerns have already been raised by UK-based students in Jisc’s report on ‘Student perceptions of generative AI’:

Students express concerns about the potential negative consequences of relying too heavily on generative AI tools, as it could impede their intellectual growth. They worry that an excessive reliance on such tools may lead to a decline in knowledge acquisition and hinder the development of critical thinking skills. (Jisc National centre for AI in tertiary education – Student perceptions of generative AI – 15 August 2023)

Use GAI to support your learning and personal development not do your learning.

Privacy, ethics and risks of using GAI tools

There are thousands of GAI tools, new ones are appearing every day. Before you sign up to use any of these GAI tools it is important to look at the terms and conditions, read them carefully. Many of us to skip reading these and just blindly accept them. Take time to read them and understand how these tools might use your personal data and the data you might share with the tool to prompt answers. Think carefully about any potential risks, ethical and privacy issues, as well as benefits before you sign up. This is particularly important for free GAI tools, these tools are rarely genuinely free, they often make money by using or selling your data.

Be aware that:

• Sharing personal or sensitive data in an AI platform, just as with social media, has the potential for it become publicly available. This also applies to any research data or intellectual property that you may share. 
• One of the criticisms of GAI is that it does not respect copyright and credit the sources that it has been trained on. These criticisms are particularly acute in relation to creative works in art and literature and computer code and there are a number of pending lawsuits against GAI companies for breach of copyright.
• GAI companies train their models on content that we submit, this in turn helps them generate income.
• There are concerns that some of these platforms have adopted exploitative and unethical practices to cleanse their data of violent and hateful content by employing individuals in developing countries who have been traumatised by the content they have had to vet.
• Concerns also prevail about the environmental impact of GAI. These technologies have high energy and water requirements.
• You do not have to use these tools if you do not wish to.
• Misuse of these GAI tools may result in academic misconduct.

GAI and Assessment

Academic integrity is central to the University’s core values of: valuing people; working together; integrity; making a difference; and excellence. The expectation is that the work you submit for your assessments is solely your own work, or in the case of group work, solely the work of your group. This work is a means for you to demonstrate what you have learned. Submitting work which is not your own and the unauthorised use of GAI is considered academic misconduct. Using GAI to write the text, code or generate the designs and artwork for your assessment would not be your own work.

If your lecturers have made clear in your assessment brief that you may use GAI sources in your work then you must be transparent in acknowledging, describing and referencing how you have used it. Contact your module leader if you are unsure whether you are permitted to use any AI tools in your submission.

Acknowledging the use of GAI in your work

It is important to acknowledge the use of GAI tools in your work and to be clear about how you have used them. You should therefore include a statement of acknowledgement and a description of how you have used GAI and the information generated. 

Statement of acknowledgement

Please include the statement that reflects how you have used any GAI tools:

• No content generated by AI technologies has been presented as my own work.
• I acknowledge the use of <insert AI system(s) and link> to generate materials for background research and self-study in the drafting of this assessment. 
• I acknowledge the use of <insert AI system(s) and link> to structure or plan this assessment.
• I acknowledge the use of <insert AI system(s) and link> to generate materials that were included within my final assessment in modified form. 
• I acknowledge the use of <insert AI system(s) and link> to generate materials that were included in my final assessment in unmodified form. An example of this would be a quote or image produced by the AI output.

Description of how information or material was generated

You must describe how the information or material was generated, what the output was and how the output was modified by you using the following format:

• The following prompts were input into (name of AI tool: [Provide details]) 
• The output obtained was: (Paste the full output generated by the AI tool) 
• The output was changed and adapted by me in the following ways: (explain how you adapted the output for use in your work) 

Referencing GAI

The referencing styles are still developing their guidance on how to reference GAI tools such as ChatGPT. As content generated by GAI is non-recoverable, it cannot be retrieved or linked to in the way that other resources that may be referenced in your work. We recommend that AI generated content should be cited as a personal communication as an in-text citation. Some styles such as Harvard have released guidance on how to do this which can be found in Cite them Right and is presented below as an example.

In-text citation

When prompted by the author, ChatGPT responded with a ‘definition of academic integrity’ (OpenAI ChatGPT, 2023). A copy of this response is in Appendix 1.

Reference list

OpenAI ChatGPT (2023) ChatGPT response to John Stephens, 2 April.

Practice and guidance in the use of GAI is evolving rapidly and is likely to change over time. You should regularly check back and review this guide to ensure that your practice is in line with the most up-to-date advice from the University.

15 September 2023

Acknowledgement: This guidance is informed by and builds on guidance developed by UCL, Heriot Watt University and Napier University. The work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International Licence