Welcome to the introduction to Generative Artificial Intelligence (GenAI) page. By reading this, you will gain a high-level understanding of what GenAI is, what is a Large Language Model, and how it functions.

Please explore the links below to find specific information you are looking for on this page.

What is Generative Artificial Intelligence?

Generative artificial intelligence (GenAI) technologies are a subset of artificial intelligence (AI) that use models based on patterns learned from large quantities of training data to generate new text, audio, or other media.

What is a Large Language Model?

A large language model (LLM) is a type of GenAI designed to understand and generate human language. It achieves this by processing and analysing large volumes of text data from diverse sources, such as books, articles, and websites. The model employs complex mathematical algorithms and neural network architectures to learn patterns, relationships, and structures within the language. LLMs often take the form of a chatbot like ChatGPT.

Understanding the Functioning of GenAI

Data Learning: GenAI learns by processing large amounts of text and visual data from across the internet, including books, articles, and websites. This helps it grasp grammar and language nuances that it can use in its responses.

Pattern Recognition: By analysing these large volumes of data, GenAI learns language patterns: how words fit together in sentences and the contextual meaning of different phrases. This enables it to understand user queries and generate relevant responses.

Response Generation: When a user interacts with GenAI by asking a question or giving a prompt, the program employs its learned patterns to construct a response. It aims to provide coherent and informative answers, drawing from its extensive knowledge.

Creativity and Prediction: Beyond factual responses, GenAI can display creativity. It can generate imaginative stories, anticipate likely outcomes based on context, and simulate human-like thinking to enhance interactions.

Bias & Hallucination: Depending on the data it was trained on, GenAI can exhibit unintended favouritism or unfairness in its decisions or outputs. Also, GenAI can generate content that is unrealistic, inaccurate, or doesn't correspond to the patterns it has learned from the training data. This is hallucination.

Continuous Improvement: GenAI is not static; it evolves over time. Its developers continually update it with fresh information and improved algorithms to enhance its comprehension and conversation capabilities. 

User Feedback and Learning: User interactions play a pivotal role in refining GenAI. It learns and updates from each user interaction.