"

4. AI applications and task examples of AI

Simplifying complex concepts

AI tools can be used to help simplify complex topics. Use the tool as an expert to explain concepts in simpler terms. Set a context or an age group to help determine the level.

Input:

You are a skilled teacher with a talent for breaking down complex concepts into simpler ideas. Describe the basics of string theory in a way that a first-year university student would easily grasp.

Output:

Understanding String Theory: A Simple Introduction

Imagine the universe as a giant orchestra, where every particle is like a musical note played by tiny, vibrating strings. These strings are the fundamental building blocks of everything we see around us. Instead of particles being tiny dots, as described in traditional physics, string theory suggests they are more like tiny loops or threads.

Here’s how it works step by step: …

[Prompted 29 November 2024, OpenAI (2024), ChatGPT 4.0 abbreviated response on topic]

Sometimes using analogies can help with comprehension of complex topics. Try asking the tool to use an analogy to explain complex ideas.

  • Explain hawking radiation core concepts using cooking analogies.
  • Explain the bicameral mind using nautical examples.

Summarising content

AI tools can summarise text and even full papers and reports.

important icon Be careful not to upload licensed or copyright-protected materials into AI tools. Refer to restrictions on the use of online collections to help you use Gen AI ethically and legally.

You can upload items that have been created under a Creative Commons licence, such as this module. LLMs often have limits on how much text can be uploaded, so using a link to an open resource may make a simpler prompt.

Input:

Act as a graduate student and summarise the following article “The rate, causes and predictors of ambulance call outs to residential aged care in the Australian Capital Territory: A retrospective observational cohort study”. Be concise by using multiple paragraphs with dot points and highlighting key points in the data and conclusions. [Upload open access article] or link to it: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0311019

Output:

The article “The rate, causes and predictors of ambulance call outs to residential aged care in the Australian Capital Territory” analyzes patterns of ambulance usage in aged care settings. Below is a structured summary: …

Input:  

Based on the summary responses, create an X/Twitter thread targeted to a general audience on general outcome of this study. Start with an introductory tweet which includes title of the paper, authors and DOI and what the study was about in general terms. Divide the thread into 7 tweets with up to 280 characters. Stay concise but engaging. End your thread with a concluding tweet with the main highlight of the study.

Output:

1/7 🧵 “The rate, causes and predictors of ambulance call-outs to residential aged care in the Australian Capital Territory” explores why ambulances are frequently called to aged care facilities. Authors: [insert author names]. DOI: https://doi.org/10.1371/journal.pone.0311019.

2/7 🚑 Key finding: Ambulances are called to aged care facilities (RACFs) often, especially for preventable issues like falls and respiratory infections. This highlights a strong reliance on emergency services for manageable health problems. …

[Prompted 29 November 2024, OpenAI (2024), ChatGPT 4.0 abbreviated response on topic]

AI tools can organise and improve your notes.

Input:

Act as a helpful assistant by creating clear, well-organized, and informative meeting notes from the provided information:

Dec Meet

  • Planning a unit retreat. Discuss future research needs and to act with external collaborators. Funding issues and grants need to be included.
  • Unit awards to be discussed and ideas who added
  • Joe Bloggs will be away for the next month
  • The information chart and structure on Y drive needs everyone to update it

Brainstorming and ideas generation

AI tools can be used to help prompt ideas for projects and research. Try asking for lists of topics, ideas and subject gaps or to suggest ideas. If there is a topic or area that you find difficult, ask what you can do to improve your understanding or approach.

  • We are exploring ways to address smoking cessation in rural Australia. Can you generate 3-5 “How Might We” questions that could help us reframe the issue and inspire innovative solutions?
  • How would a child, a scientist, and an artist each solve the problem of plastic waste?
  • Give me a concise summary of the major strengths and weaknesses of carbon credits in Australia.

Image generation

Many AI tools can generate images. The more descriptive your prompt is, the more relevant your result will be.

Consider adding the following feature outputs to your prompt:

  • Image type — RAW, JPEG
  • View — close-up, a wide shot, a bird’s-eye view, or a specific angle.
  • Subject — What the image is
  • Context — The place
  • Details — Higher resolution, softer focus, brighter, glowing, dark, dynamic, swirling, Ultra-HD
  • Lighting — Natural, soft
  • Framing — Landscape, macro, AR (Aspect ratio)
  • Lens — Depth of field, bokeh
  • Style — Anime, realistic, painting, claymation, pencil sketch, cyberpunk, “art deco,” or minimalist.

Input:

A raw photo of a cat reshelving books in a library, using a wide-angled lens with cinematic morning lighting –ar 16:9.

Output:

Image of a cat shelving books in a library created with AI

Note: Microsoft Copilot has not provided the required aspect ratio of 16:9. You should always verify the accuracy and quality of the output.

[Prompted 13 November 2024, Microsoft Copilot (2024)]

Refine or get feedback

In your prompts, add a context or a role and ask for the specific type of feedback you want. Narrow it to one or two areas if it is too general and be clear about the amount of feedback. AI tools can also be used to check on grammar, spelling and structure in your writing. You can use it to rephrase a sentence or paragraph and to modify tone and audience.

Input:

As a journal editor, evaluate the following paper, highlighting both its strengths and areas for improvement. Your candid and objective feedback is appreciated. Use the table format to show the evaluated paper and feedback comparison.

Output:

Here’s a structured evaluation of the paper using a strengths and areas-for-improvement comparison:
Aspect Strengths Areas for Improvement
Objective & Relevance – The study addresses an important and under-researched issue: [example] – The paper could further explore the broader implications of its findings on policy and [example] systems.
Methodology – A retrospective observational cohort study provides valuable data on real-world practices. – The study could benefit from a more robust analysis of causality versus correlation in [example].
Data & Analysis – Clear data presentation, with solid statistical analysis backing key findings. – The study could improve by including a comparison with similar regions or populations to generalize the results.
Recommendations – Practical and actionable recommendations aimed at reducing ambulance reliance, such as improving on-site care. – More specific interventions or pilot studies should be proposed to test these recommendations in practice.
Clarity & Structure – The paper is well-organized, making it easy to follow the argument and evidence. – Some sections could benefit from clearer linkage between data points and conclusions.
Implications for Practice – Offers significant insights for improving aged care, with implications for resource allocation and care delivery. – More detailed exploration of implementation barriers in aged care facilities could be added.

Overall Feedback:

This paper provides valuable insights into [example], offering actionable recommendations. However, strengthening the discussion on causality, broadening comparisons, and exploring implementation barriers would improve its practical value.

[Prompted 29 November 2024, OpenAI (2024), ChatGPT 4.0 generated this table of the article]

Performing repetitive tasks

AI tools can help to create code, excel formulas, and scripts for data visualisation and to refresh text for social media sources.

Act as a Python programmer and simplify the following code, ensuring it stays easy to read and more efficient. [Insert code]

Organise information from unstructured data

AI tools can create tables from text.

  1. Paste in your notes or use a tool with internet access to access information from webpages or reports
  2. Ask the tool to generate a table with this information and copy and paste or export it.

Input:

What is the population of each state in Australia in 2023 compared to 2001? Generate one table containing the 2023 and 2001 information and the percentage change from [example source].

Output:

Table with population data for each state

[Prompted on 12 January 2024, Microsoft (2024), CoPilot generated this table in response to the topic. It also includes options to follow up the sources and export the table to Excel.]

Language translation

AI tools can be used for language translation. When the translation is complete, review it for any errors. Read our Machine Translation Library Guide for more information.

Translate the following text to English. Keep the translation as close to the original in tone and style as you can. [Article URL]

Mouse iconThe University of Sydney’s AI in Education offers many more task examples of how you can apply AI tools to help you learn.

Licence

Icon for the Creative Commons Attribution-NonCommercial 4.0 International License

Artificial Intelligence Copyright © 2023 by The University of Queensland is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License, except where otherwise noted.

Share This Book