LibGuides: AI Student Guide: Home

AI: A Student Guide

This guide offers students an overview of generative artificial intelligence (AI) tools and basic concepts of AI to help them get started on applications of this topic in a college setting. The field is evolving quickly, however, and the information here may quickly become outdated! We do strive to provide the most accurate information at our disposal to share with students.

Always consult your instructor for AI usage policies and use these tools with caution.

Disclaimer: the library does not endorse specific AI technologies and advises against sharing personal information when using them.

What is Generative AI?

Generative AI (Artificial Intelligence) is a type of technology that can create new content, like text, images, code, audio, or video, based on what it has learned from existing content. The process of learning from existing content is called “training.”

Adapted from JISC. Generative AI - A Primer

GPT stands for “Generative Pre-trained Transformer” and is the name given to a family of natural language models developed by Open AI. It is a form of generative AI because it can produce original results and generate human-like text based on input prompts. ChatGPT is a large language model (LLM).

AI text generators, like ChatGPT, work by predicting the next word in a sentence. They don’t understand questions like humans do. These tools are trained on massive amounts of data to find patterns; they, in turn, use these patterns to predict words and then generate new content. AI image generators are trained in a similar way.

Generative AI is already part of many tools you use daily, and more applications are being developed. You’ve probably already encountered them in Chatbots (e.g., ChatGPT), writing assistance (Grammarly), language translation (Google translate), image generators (DALL-E), video edition (Adobe Premiere Pro) even some GPS systems (Waze).

Generative AI is NOT a search engine.

Search engines like Google gather and rank content from various websites, providing direct links to those sites.
Generative AI tools, such as ChatGPT, create text responses based on their training data and existing knowledge. Many of these tools (such as ChatGPT) do not actively search the internet, although some, like Bing's Copilot and Google's Gemini, do.

As you may have already experienced, generative AI is now being added to search engines. Instead of just giving a list of links, generative AI searches the web and summarizes the information it finds, presenting it directly on the answer page. This means you don’t need to visit multiple websites to get your answer. The concerns with this are that:

An AI tool's knowledge is limited to what it was trained on, and an algorithm fed inaccurate training data will produce poor results. There can be factual inconsistencies and biases.
Because these models generate text by studying patterns and calculating probabilities about what word to write next, they are actually not thinking. They cannot check their own output for inaccuracies, and can make up answers (called Hallucinations).
Where the information comes from (its provenance) can be unclear, making it difficult for you to verify the information yourself.

Go to the AI Considerations area of this guide for a further discussion of AI limitations.

Definitions

	ARTIFICIAL INTELLIGENCE The capacity of machines to mimic human cognitive functions such as learning, problem-solving, and pattern recognition, enabling them to perform tasks that normally require human intelligence. It includes various subfields, such as machine learning and natural language processing. (UNF, 2023)
	DEEP LEARNING (DL) A subset of machine learning where artificial neural networks with many layers. Like neural networks, deep learning is modeled on the way the human brain works and powers many machine learning uses, like autonomous vehicles, chatbots, and medical diagnostics. (Brown, 2021)
	LARGE LANGUAGE MODELS AI models that can understand, generate, and interpret human-like text based on the input it receives. (Allied Media, 2018)
	MACHINE LEARNING A subfield of AI where computer systems are given the ability to learn and improve from experience without being explicitly programmed. This is usually achieved by training the system with large amounts of data. (Allied Media, 2018)
	ARTIFICIAL NEURAL NETWORKS Modeled on the human brain, in which thousands or millions of processing nodes are interconnected and organized into layers. In a neural network trained to identify whether a picture contains a cat or not, the different nodes would assess the information and arrive at an output that indicates whether a picture features a cat. (Brown, 2021)
	SPEECH RECOGNITION AI technologies that transcribe spoken language into written text, enabling human-computer interactions through voice. (UNF 2023)