Lesson 1.1: What is AI?
Module: 1 - Introduction to AI & Prompting
Duration: 20 minutes
Type: Reading + Quiz
Introduction
Welcome to your AI journey! Before we dive into prompt engineering, let's understand what artificial intelligence actually is and why it matters.
By the end of this lesson, you'll understand the fundamentals of AI and how we arrived at the powerful language models we use today.
What is Artificial Intelligence?
Artificial Intelligence (AI) is the simulation of human intelligence by machines. Instead of being explicitly programmed for every task, AI systems learn from data and improve over time.
The Evolution of AI
1950s: Alan Turing asks "Can machines think?"
1960s: Expert systems (rule-based AI)
1980s: Machine learning emerges
2010s: Deep learning revolution
2017: Transformer architecture invented
2022: ChatGPT makes AI mainstream
2025: AI is everywhere (including this course!)
From Programming to Learning
Traditional Programming
Input + Rules → Output
Example:
IF temperature > 30°C
THEN output "It's hot"
ELSE
THEN output "It's cold"
Machine Learning
Input + Output → Rules (learned from data)
Example:
Show 1000 examples of hot/cold weather
AI learns the pattern
Can predict temperature feeling for new data
What are Large Language Models (LLMs)?
Large Language Models are AI systems trained on massive amounts of text to understand and generate human-like language.
Key Characteristics:
- Large - Billions of parameters (like brain neurons)
- Language - Understand and generate text
- Model - Mathematical representation of patterns
Popular LLMs (2025):
| Model | Company | Parameters | Cost | |-------|---------|------------|------| | GPT-4 | OpenAI | ~1.7T | $$$ (paid) | | Claude 3 | Anthropic | ~1T | $$$ (paid) | | Llama 3.1 | Meta | 8B-405B | 🆓 (FREE!) | | Mistral | Mistral AI | 7B-123B | 🆓 (FREE!) |
We use Llama 3.1 (via Ollama) - completely FREE and unlimited!
How LLMs Changed Everything
Before LLMs (2021):
- Complex rule-based systems
- Requires programming for each task
- Hard to adapt
- Expensive to maintain
With LLMs (2025):
- Natural language instructions
- One model, infinite tasks
- Adapts through prompting
- Accessible to everyone
Example:
Old way (2021):
Write 500 lines of code for sentiment analysis
Train custom model
Deploy and maintain
New way (2025):
Prompt: "Analyze the sentiment of this text: [text]"
Done! ✅
Real-World AI Applications
1. Content Creation
- Writing articles, code, emails
- Image and video generation
- Music composition
- Translation
2. Business Operations
- Customer support chatbots
- Document analysis
- Data extraction
- Process automation
3. Software Development
- Code generation and review
- Bug detection
- Documentation writing
- Test case creation
4. Decision Support
- Market analysis
- Medical diagnosis assistance
- Legal research
- Financial forecasting
The AI Landscape in 2025
┌─────────────────────────────────────┐
│ AI Ecosystem │
├─────────────────────────────────────┤
│ │
│ Cloud APIs (OpenAI, Anthropic) │
│ └─ Easy but expensive │
│ │
│ Local Models (Ollama, LMStudio) │
│ └─ FREE but needs setup │
│ │
│ Enterprise (NexusCore, Azure) │
│ └─ Secure and scalable │
│ │
│ Open Source (Hugging Face) │
│ └─ Flexible but technical │
│ │
└─────────────────────────────────────┘
We focus on: Local models (Ollama) + Enterprise patterns
Why Prompt Engineering Matters
As AI becomes ubiquitous, knowing how to communicate with AI is becoming as important as knowing how to code.
The New Skill:
Old: "I need to learn Python to build this"
New: "I need to write the right prompt to build this"
Career Impact:
- Prompt Engineers: $120K-$250K salary
- AI Architects: $150K-$300K salary
- Demand: Growing 50%+ year over year
Key Takeaways
✅ AI enables machines to learn and improve from data
✅ LLMs are trained on massive text to understand language
✅ We can control AI behavior through prompting
✅ Ollama lets us run LLMs for FREE locally
✅ Prompt engineering is a valuable career skill
Knowledge Check
Test your understanding with these questions:
-
What is the key difference between traditional programming and machine learning?
- A) Machine learning is faster
- B) Machine learning learns patterns from data
- C) Machine learning doesn't use code
- D) Machine learning is more expensive
-
What does LLM stand for?
- A) Large Language Model
- B) Learning Language Machine
- C) Limited Language Model
- D) Local Language Machine
-
Which of these is a FREE local LLM we use in this course?
- A) GPT-4
- B) Claude
- C) Llama 3.1
- D) Gemini
-
Why is prompt engineering becoming important?
- A) It's easier than coding
- B) AI requires clear instructions to perform well
- C) It pays less than programming
- D) It's replacing all other jobs
-
What year did ChatGPT make AI mainstream?
- A) 2017
- B) 2020
- C) 2022
- D) 2025
Answers: 1-B, 2-A, 3-C, 4-B, 5-C
Further Reading
Want to dive deeper?
- 📖 Attention Is All You Need (Transformer paper)
- 📖 The Illustrated Transformer
- 📖 Llama 3.1 Technical Paper
- 🎥 Video: How LLMs Work (3Blue1Brown)
What's Next?
In Lesson 1.2, we'll explore how LLMs actually work under the hood - understanding tokens, context windows, and parameters that affect AI behavior.
Continue to Lesson 1.2: How LLMs Work →
Progress: 🟩⬜⬜⬜ (25% of Module 1)