When it comes to using AI tools in your research, the first thing to consider is why you are using them. If you are required to use AI as part of the assignment or project, you may be directed to a certain tool that your lecturer wants you to use. However, if you want to use it as an aid to research, think about whether you can get the same or better information from more reliable sources such as our academic databases.
If you still want to use AI, do some research into the tool you use and consider these questions:
Q – Where is the tool sourcing its information? Some tools, such as Perplexity, will link to the sources they use in their answers, making it easier to check the veracity. Others have simply been uploaded with a dataset, constructed from various sources which could include false or biased information.
Q – Will your personal information be kept safe and private? Remember to read the privacy statement before entering your email and other personal details.
Q – If articles are cited, can you double check they are not ‘hallucinations’ (i.e. made-up)? You can do this by checking them in our academic databases or a simple Google search to establish if the article exists in the form the citation suggests.
In 2020, Hervieux and Wheatley came up with a handy checklist for assessing AI tools called the ROBOT test:
R – Reliability – how reliable is the information being provided?
O – Objective – what is your purpose in using it?
B – Bias – are any bias issues acknowledged?
O – Ownership – who created the tool?
T – Type – what type of AI is it?
For more details on using AI critically in your research, please see this lib guide page.