Understanding AI Generative AI and LLMs

1. What is Artificial Intelligence (AI)?

Artificial Intelligence is the field of creating computer systems that perform tasks normally associated with human intelligence.

Examples include:

• Recognizing faces in photos.

• Understanding human language.

• Predicting future demand.

• Detecting fraudulent transactions.

• Answering questions and writing code.

Think of AI as a large umbrella containing several different technologies.

For example, a demand forecasting system might predict how much material will be required next quarter. A Generative AI system could explain that forecast in plain English. An LLM could also answer questions about the forecast using relevant data supplied to it.

2. What is Generative AI?

Generative AI is AI that learns patterns from data and uses those patterns to create new content.

Examples:

The key difference is that traditional predictive systems often estimate a value or category, whereas Generative AI produces content.

3. What is an LLM?

LLM stands for Large Language Model.

Let's break down the name:

• Large: It is trained using substantial amounts of data and computational resources.

• Language: It learns patterns in text, code, and other language-like sequences.

• Model: It is a mathematical system with learned parameters that processes input and produces output.

Examples include GPT, Claude, and Gemini.

How does an LLM generate a response?

Imagine asking:

What is the capital of France?

A simplified explanation of the process is:

An important point: an LLM is not simply searching a database of ready-made answers. It generates text based on its learned patterns and the context available to it. It can still make mistakes, so important answers need verification. 

In next lesson, we'll explore tokens and tokenization in detail: why a word is not always one token, how tokens affect input and output limits, and why token counts matter when building AI applications.