Chapter 1: Note to Readers
Key concepts: Note to Readers
1. Note to Readers
Defining AI
- McCarthy's 1955 proposal first used 'artificial intelligence'
- AI is the science of making intelligent machines
- Author's definition: machines that think, see, learn, act like humans
- Understanding this definition is key to reading the book
Abbreviated History of AI
- Timeline from Turing test (1950) to ChatGPT (2022)
- Key milestones: ELIZA, Deep Blue, Roomba, Watson, Alexa, AlphaGo
- Transformer paper (2017) changed everything
- Progress has accelerated dramatically in recent years
The AI Zoo Concept
- AI is a zoo of different species, not one technology
- Systems differ: recommendation engines, self-driving cars, chatbots
- Companies misuse 'AI' as a single magical brain
- Understanding differences reveals what AI can and can't do
Core AI Glossary
- Nested circles: AI > Machine Learning > Neural Networks > Deep Learning
- Machine learning learns from data, not if-then rules
- Deep learning handles messy real-world data
- Key terms: model, training, frontier models
Key Takeaways
- AI is a zoo of systems with different strengths
- Definition: intelligent machines that may exceed humans
- Machine and deep learning are modern AI engines
- History shows rapid acceleration in progress
