While Large Language Models (LLM) like ChatGPT can provide valuable assistance, they are no replacement for human guidance and support.
Further, there are reliability and ethical issues surrounding these models and their outputs which are explained on subsequent pages of this guide. Always evaluate the information in front of you. ChatGPT has no truth filter!
ChatGPT can help with the following:
In order to get useful results you will need to build the right questions and set parameters for the reply or dialogue you are seeking. This concept is also called prompting, prompt design or prompt engineering.
Knowing how a Large Language Model (LLM) like ChatGPT works can be useful for creating prompts and understanding its limitations. LLMs use their vast amounts of training data to try and predict words. These predictions are based on learnt patterns and content around words. In the examples below, the words 'colourful' and 'garden' in the prompt help ChatGPT predict the missing word. You could almost think of it as a very sophisticated autocomplete. If confronted with less patterns or content in the prompt the output is less specific.
The examples above demonstrate that the prompt you enter will influence the reply you will get.
A good prompt generates basic text that needs tweaking; a great prompt is specific and useful to the user. Great prompts have some or all of the following elements:
Try the following two prompts in ChatGPT and see the difference:
(by QUT, licensed under CC BY-NC-SA)
Southern Cross University acknowledges and pays respect to the ancestors, Elders and descendants of the Lands upon which we meet and study.
We are mindful that within and without the buildings, these Lands always were and always will be Aboriginal Land.