Skip to Main Content

Generative AI for Faculty

What are Large Language Models (like ChatGPT) and How Do they Work?

What is a Large Language Model (LLM)?

ChatGPT is one of several artificial intelligence Large Language Models (LLM) being developed. ChatGPT (generative pre-trained transformer) is owned by Open AI, based in San Francisco. Elon Musk and Microsoft are both major investors, along with a variety of venture capitalists. The company is privately owned and is rapidly monetizing. Other LLMs are currently being developed by Google and Meta as well as several other companies. 

While LLMs have been around since the 1960s, ChatGPT-3.5's introduction in November, 2022 made a very big splash. This is due to a variety of factors, including:

  • its easy accessibility (GPT-3.5 is free)
  • it's fast, easy to query and is capable of human-like discussion
  • social media was quick to spread the word

Nevertheless, it is not without its concerns and principal among those concerns are its impact on teaching and education, which this guide attempts to address. Other uses, such as its ability to provide help with coding, for example, are not addressed. And, as more and more people become familiar with the tool, more and more ways it can be useful are becoming identified. 

How Do You Use It?

For a better understanding of how LLMs are created and how they operate, see the links included in the page "Associated Sources for Understanding LLMs." Information explaining how to develop effective prompts that will result in more robust responses from ChatGPT are also found there.

For instance, ChatGPT works by responding to prompts given by the user. It is designed to be similar to having a conversation with someone who is giving you assistance. So, the conversation can expand, go in different directions, adapt to a different tone, go into more detail, etc., depending on the prompts it receives from you.