Machine-learning algorithms: Generative artificial intelligence (GenAI) tools use machine-learning algorithms to create “new” digital content (e.g., text, images, audio, multimedia, computer code, etc.).
Patterns, not analysis: GenAI tools are not currently capable of evaluating, critically combining, or verifying information. Instead, they predict and rearrange the most likely outcome based on the requested output, feedback, and patterns in the data models they are trained on.
Natural language processing (NLP): NLP is a subset of GenAI tools (e.g., ChatGPT) that allows computers to analyze and then mimic human language. Paired with a user-friendly interface, these tools are commonly called chatbots and allow users of all skill levels to place requests and interact with the language models in a conversational manner.
Example: A chatbot for a course could answer student questions related to the course subject matter, offer definitions, summarize or identify main ideas in documents or materials, generate review questions, or identify support resources. For the model to learn, it needs to be exposed to relevant data. The chatbots learn what the best answers are and how to deal with questions that don’t fit the framework of the questions asked.