What is generative AI?

Tip - It is important that we learn more about artificial intelligence tools and have discussions to support students to understand the strengths and weaknesses, as well as the ethical concerns associated with the use of AI.  

Generative artificial intelligence (AI) is a type of artificial intelligence software that produces outputs such as text, images, computer code, audio or video. At a basic level, generative AI works by analysing vast sets of data, identifying patterns, and producing output based on what a human would be statistically likely to create (JISC, 2023). Typically, these Large Language Models (LLMs) such as ChatGPT or Bing produce unique responses every time (JISC, 2023) and are often interacted with through a ‘chat-bot’ interface using ‘prompts’ in a question-and-answer format. 

While there has been a recent increase in the use of generative AI tools, artificial intelligence more broadly is prevalent in the technology we use day to day. Common applications of artificial intelligence include predictive text, search engines, recommendations from music or TV streaming services or digital voice assistants.  

Though OpenAI’s products such as ChatGPT are currently the most prominent, it is important to remember that there are other types of generative AI software available. Examples include: 

In addition, Microsoft has updated its search engine, Bing, with artificial intelligence technology, and is investing billions of dollars into the development of Copilot. Google is also looking to expand their chatbot, Bard. 

It is likely that the references to specific AI software in this document will quickly become outdated as new and more powerful AI systems are released. Nevertheless, although particular platforms may change, there are some principles that we can apply to student use of any powerful generative AI platform.