1. AI Fundamentals

Learning objectives

By the end of this lesson you should understand:

Artificial Intelligence (AI) is the broad field of building systems that perform tasks associated with intelligent behavior.

Examples:

A simple abstraction:

Input
  |
  v
AI System
  |
  v
Output

AI does not necessarily mean a chatbot. A recommendation system, fraud detector, computer-vision system and coding agent can all be AI systems.

Machine Learning (ML) is a major approach to AI in which models learn patterns from data.

Training data
     |
     v
Learning algorithm
     |
     v
Trained model
     |
     v
New input
     |
     v
Prediction

Examples:

Deep Learning uses neural networks with multiple layers.

It is widely used in:

Generative AI creates new content.

Examples:

Traditional predictive ML might answer:

Is this transaction fraudulent?

A generative model might answer:

Write a Python function that validates this transaction.

Training and inference are different.

A model learns patterns from training data.

A trained model processes a new input and produces an output.

For an LLM:

Prompt
  |
  v
Model inference
  |
  v
Generated tokens

AI output is not automatically correct.

A model can:

Therefore an engineering workflow must include verification.

Generate
   |
   v
Test
   |
   v
Review
   |
   v
Verify