UC Library Guides: GenAI for Students: What is Gen AI?

What is GenAI?

Generative Artificial Intelligence, or GenAI, is a type of artificial intelligence that allows machines to create new content like text, images, music, videos, code, and more, based on inputs or prompts. These GenAI systems are trained on massive data sets, learning patterns and features from existing data. When you use a GenAI tool, it analyses these data sets, and generates new content that resembles what it has learned, much like the predictive text on your phone.

The following video explains how GenAI works and some of the uses and limitations of using GenAI at university.

What you need to know about GenAI tools

While GenAI can be a powerful tool, there are important considerations to keep in mind. Being able to responsibly and ethically use Generative AI is dependent on your critical evaluation skills, much like when you evaluate the results from a Google search or academic database. This means critical thinking skills are essential to evaluate what has been generated.

Accuracy and Authenticity: GenAI tools often produce content that can look or sound convincing, but may be inaccurate, misleading or completely false (often known has AI hallucinations). Always verify the generated content against reliable sources.
Bias and fairness: GenAI tools reflect the biases present in the data they were trained on. This often leads to biased or unfair outputs. It is important to be critical of the results and consider diverse perspectives.
Privacy and Security: Using GenAI tools often involves sharing data. Be mindful of what information you input as it could be added to the tool’s training data. It is also important to consider privacy or data security policies.
Copyright and Plagiarism: Generated content can often infringe on existing copyrights. You are responsible for any AI-generated content that you submit. Content not approved by the unit convener, and/or appropriately referenced, may violate the University’s Assessment Policy.

For more information, refer to GenAI at UC and GenAI and Ethical Considerations.

Types of GenAI tools

The tools listed below are examples of the types of tools available and are not endorsed by the University.

Type	Purpose	Potential student uses	Considerations	Examples
Text GenAI	Composes content from a conversational language model.	Brainstorm and idea generation  Text translation  Create practice quizzes and questions  Summarise difficult texts	Plagiarism risk  Has been known to make up content  Correct citation and referencing needed  Academic integrity is essential - all use must be "permitted and admitted"	ChatGPT  Perplexity AI  Microsoft Copilot
Image GenAI	Creates images based on learned patterns and styles from text descriptions and prompts.	Source of inspiration and ideas  Create images for presentations  Assist in design elements and choices  Explore and understand visual concepts  Storyboards	Is the content biased or unethical?  Copyright and attribution. Who owns the copyright of the image?	Dall.E.2  Stable Diffusion  Midjourney  Adobe  Canva Microsoft Copilot
Code GenAI	Generate or assist in the creation of computer code.	Help with syntax  Assistance with complex coding  Learn different coding patterns and structures  Debug	Stay mindful of AI limitations  Review and test the code before using it  Does it meet standards and security requirements?	ChatGPT  CodeT5  Tabnine
Sound GenAI	Generates sounds, music and other auditory content.	Create background music for presentations, podcasts, videos and games  Inspiration and creativity  Explore sound design principles  Expression	Copyright issues of content  Appropriateness of content	AIVA  Soundful  Murf.ai
Video GenAI	Uses images and text to generate video content.	Add special effects to video  Create graphics for presentations  Edit  Create prototypes	Quality and authenticity  Understand the limitations  Ethical use	Invideo  Gen-1 Runway
Research discovery and explanation GenAI	Generates content to assist the research process.	Summarise large number of research papers  Analyse data sets  Use as a starting point	Validate and verify the generated results through traditional research methods  Information could be inaccurate  The training data may have biases and ethical issues	Elicit  Raxter  Research Rabbit  Scite

(Copied from What is generative AI - Generative AI and Assignments - Library Guides at James Cook University (jcu.edu.au) Licensed under Creative Commons Attribution-ShareAlike (CC BY-SA) 4.0 International License)