Skip to Main Content

Artificial Intelligence

Engaging with AI: Prompting and Giving Instructions

Getting the best output from generative AI is often based on how a user communicates with the AI tool in question. To communicate with AI effectively, you need to learn about 'prompting.' However, before starting it is a good idea to ask yourself what your goal is?

  • For example, do you want to brainstorm ideas or need hep to get started with a project? 
  • Or do you want AI to consult its knowledge banks and summarise a topic and the arguments therein? 

The clearer you are about your questions, the better the AI’s response will be. You need to  understand your project needs before starting. 

Be specific - the more precise and concise the question you ask, the more likely it is that the AI tool will produce a useful answer. For example, rather than prompting AI to talk about mental health, you could prompt it to talk about "wellness programmes in the modern hybrid workplace and mental health advocacy." 

Give AI scope and frame - think about the type of language  you want the AI to use, i.e., formal or informal, do you want the output to be in simple terms or something more complex? Is there a specific word count or format? You can issue these instructions to the AI. 

Offer feedback by asking follow-up questions, and engaging with the content it produces. Let the AI know therefore, if the first answer you get isn't right or if you need to add more boundaries. Engaging successfully with AI is about offering it direction - remember the tool works for you!

 


A reminder about appropriate use

Always refer to your supervisor or module lead about the expectations surrounding AI use and academic work.

Be critical about the content that AI produces - you still need to verify its value and accuracy.

Use AI as an assistant, and as inspiration. It should not be used to craft an entire assignment nor can it replace the human ability to critical engage with theory and concept.