Note: All use of GenAI tools should be for personal learning or for assessments where use of GenAI is explicitly included in the assessment instructions or the unit outline.
Generative AI tools may improve efficiency and increase productivity by streamlining critical analysis, synthesis, design and writing processes, and for researchers, streamlining the preparation of grant, fellowship and project proposals and publications.
GenAI tools may be useful in certain scenarios to stimulate critical or creative thinking by providing new insights and perspectives. Under appropriate circumstances, GenAI tools can also help with analysis of large amounts of non-sensitive data and help to highlight important findings, saving hours of manual data analysis.
GenAI can be a useful tool to assist you with your studies in the following ways (please ensure your use of GenAI meets UC guidelines and academic integrity principles as per the section GenAI at UC):
Feedback literacy
Providing clarification and suggestions for improvement from unit convener feedback.
Accessibility
Creating tools or content to make learning more accessible, such as generating sign language interpretations or readable text for the visually impaired.
Career readiness:
Create practice interview questions
Write resumes and cover letters.
(Deakin University. (2024). Generative artificial intelligence (AI): Thinking about using AI in your research? Read this first. Research site. https://www.deakin.edu.au/research/support-for-researchers/research-integrity/generative-artificial-intelligence-ai)
Prompts are elements such as words, phrases, questions, or keywords that users enter into GenAI tools to generate results. Prompts refer to the instruction or discussion topic that users provide for the GenAI tool to respond to. A prompt can take the form of a question, statement, or any stimulus aimed at fostering creativity, reflection, or engagement. These prompts are designed to be open ended allowing customisation to align with the user's preferences and interests.
The better the prompt, the better the results obtained will be. Techniques include direct prompting (with or without examples) and role prompting (e.g. asking the AI to play a patient for practising diagnostics). Best practices include clear communication, using specific examples, and applying constraints to guide the model’s output. Creativity and persistence play a significant role in achieving better results/be creative and keep practising to achieve the best results.
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:
Give the AI a persona.
Are specific, clear, and concise. Define what you are looking for.
Provide content and avoid ambiguity. Open-ended prompts are not narrow enough to generate a meaningful response.
Provide specific constraints, such as a target length or tone, to guide the model's response.
Are conversational and use complete sentences.
Additionally, a good prompt should have a clear and achievable goal that the model can work towards.
Repeat the process to iteratively improve your prompt's effectiveness. As you chat with GenAI, the tool will learn more about what you need. Use the conversational interaction to help the AI improve its outputs.
The acronym ACCENT outlines six strategies to make your prompts more effective.
Alignment: Ensuring the model’s capabilities are in sync with your task.
Clarity: Writing clear and concise prompts.
Context: Providing relevant context so that the output content is relevant to your needs.
Examples: Using examples to demonstrate the format of response that you require.
Neutral language: Designing prompts to minimise biases (for example, to minimise assumptions about
gender or race).
Trial and error: Refining prompts based on the tool's performance and feedback.
(This framework for improving prompt engineering was adapted from the 26 guiding principles proposed by Sondos, M. B., Myrzakhan, A., & Shen, Z. (2024): Principled instructions are all you need for questioning LLaMA-1/2, GPT-3.5/4 )
When prompting GenAI (such as Microsoft Copilot) it is effective to break your prompt down into three parts:
Outline the domain or topic
Any relevant background information
Any specific terminology
Constraints and limitations of the specific context
Task: Outline the topic or activity you want it to write about (for example, asking for a menu with macro-nutrient make up.)
State your objective or goal
Break down the steps involved
Put in any details such as information or text that the GenAI needs
Output: How do you want it to respond? What kind of format? What kind of tone (i.e. formal or informal, etc.)
Specify the format you want the output in
Outline how much detail you want
Any specific things you want included or excluded from the output.
Prompting examples
Tell me in simple terms how the theory of relativity works. Use no more than 50 words.
Create a 10-question quiz about the theory of relativity. Provide feedback for my answers.
Act as a speech pathologist. Provide an assessment of a three-year-old child based on the speech sample: Mummy get in ka.
Summarise this article in one sentence, preserving information about dates and locations.
Prompt writing resources
Many of these sites mention ChatGPT, however the University would prefer you use Microsoft Copilot. The principles will be the same.
University of Newcastle's ChatGPT Prompt Writing Resources
Prompt Engineering from OpenAI (including "Six strategies for getting better results")
The Masterpreneur provides some useful tips in their guide for beginners
Uplevel your prompt craft in ChatGPT with the CREATE framework by Tom Barrett, Edte.ch, 6 February 2023
How to write better ChatGPT prompts for the best generative AI results by David Gewirtz, ZDNet, 11 August 2023
(Sources: Using GenAI - GenAI basics - LibGuides at Deakin University ; ChatGPT and prompting - GenAI - Library guides at Southern Cross University (scu.edu.au); Bing Copilot )
Microsoft Copilot is an AI-powered digital assistant that aims to provide personalised assistance to users. It incorporates GenAI with the MS Office environment.
The University of Canberra (UC) provides free access for staff and students to the premium version of Microsoft Copilot and is available to UC staff and students. Sign in using your university login credentials.
Microsoft Copilot is based on GPT 4 and can be used in many of the same ways as ChatGPT. It includes up-to-date web search capabilities, so its responses have more chance of using up-to-date sources. Copilot also provides links to where it obtained information, which can aid in checking for authority and accuracy. Image generation is also possible via Copilot 's integration of DALL-E 3.
Copilot is an effective way to safely explore the capabilities of GenAI. Your prompts and data will not be saved or used outside of UC - so your intellectual property remains safe:
Your queries will not be used to train the underlying large language model
No history is retained of your queries or the responses
No reporting or auditing capability is provided.
Access Copilot here.