Large Language Model (LLM)

Artificial intelligence systems trained on vast amounts of text data to understand, generate, and manipulate natural language. LLMs like GPT, Claude, and Gemini use neural networks with billions or trillions of parameters to process language patterns, enabling them to write essays, answer questions, generate code, translate languages, summarise documents, and engage in nuanced conversations. These models are 'large' both in training data volume and model parameter count. LLMs don't truly 'understand' like people do but excel at pattern matching and generating statistically probable responses based on training. They've revolutionised how people interact with computers and automated countless knowledge tasks.

Why it matters

LLMs represent the most transformative technology since the internet, enabling individuals to automate cognitive work that previously required trained experts. Understanding LLM capabilities and limitations is essential for building competitive AI-powered businesses.