Chapter 1: Experience AI in 3 Minutes

After completing this chapter, you will: use 3 lines of Python code to complete your first AI conversation, understand free/low-cost options

1.1 Your First AI Conversation Beginner ~$0.001

Prerequisites: Basic Python syntax

Why Do You Need It? (Problem)

Imagine you encounter this scenario at work: you need your program to understand natural language input and intelligently generate responses. For example:

• User inputs "Help me summarize this meeting minutes"
• User asks "What bugs are in this code?"
• User says "Translate this English email to Chinese"

In the traditional NLP era, you would need to collect labeled data → train a model → deploy an inference service, which could take weeks to months.

Now, by calling an LLM API, you can do it with 3 lines of Python code.

What Is It? (Concept)

Large Language Models (LLMs) are AI models trained on massive amounts of text that can understand and generate natural language. You don't need to train a model—just call it via API:

Before LLMs vs After LLMs:

Dimension	Traditional NLP	LLM API
Development Cycle	Weeks to months	Minutes
Data Required	Large amounts of labeled data	None
GPU Required	Training and inference	None (cloud processing)
Capability Range	Single task	General (translation, summarization, coding, Q&A...)

Try It Yourself (Practice)

We'll use OpenAI's API to complete your first AI conversation. The entire process takes only 3 steps:

Step 1: Install SDK

pip install openai

Step 2: Get API Key

Go to OpenAI Platform to register and create an API Key.

Don't Have an OpenAI Account?

No worries, the next section 1.2 Free Options & Zero-Cost Start will introduce multiple free alternatives. You can also skip there first, get a free API Key, and come back.

Step 3: 3 Lines of Code, Start Chatting

from openai import OpenAI

client = OpenAI()  # Automatically reads OPENAI_API_KEY environment variable
response = client.chat.completions.create(
    model="gpt-4.1-mini",
    messages=[{"role": "user", "content": "Explain what an API is in one sentence"}],
)
print(response.choices[0].message.content)

Run result (example):

An API (Application Programming Interface) is a set of rules and protocols that define 
how different software components can interact with each other, allowing applications 
to communicate and share data.

That's it! You've completed your first AI conversation.

Expanding a Bit—Multi-Turn Conversation:

from openai import OpenAI

client = OpenAI()

messages = [
    {"role": "system", "content": "You are a friendly AI assistant skilled at explaining technical concepts with simple analogies."},
    {"role": "user", "content": "What is an API?"},
]

response = client.chat.completions.create(model="gpt-4.1-mini", messages=messages)
reply = response.choices[0].message.content
print(f"AI: {reply}\n")

messages.append({"role": "assistant", "content": reply})
messages.append({"role": "user", "content": "Can you explain it with a real-life example?"})

response = client.chat.completions.create(model="gpt-4.1-mini", messages=messages)
print(f"AI: {response.choices[0].message.content}")

Core Understanding

• system message: Sets the AI's role and behavior guidelines
• user message: What you say
• assistant message: AI's reply (must include conversation history for multi-turn dialogue)

Run locally: jupyter notebook demos/01-first-experience/hello_llm.ipynb

Summary (Reflection)

• What Did This Solve: Only 3 lines of code to give your program natural language understanding and generation capabilities, no model training needed
• What Remains Unsolved: APIs cost money, different models have varying prices and capabilities—how to use them for free/low-cost? Next section answers this
• Key Points:
  1. LLM APIs are "calls" not "training", extremely low barrier to entry
  2. Conversations are passed via the messages array, containing three roles: system/user/assistant
  3. gpt-4.1-mini is currently the most cost-effective entry-level model (OpenAI's GPT-4.1 series launched April 2025, offering the best price-performance ratio; for the latest flagship, see GPT-5)

"3 lines of code to talk to an AI. Your grandparents needed 3 years of grad school to do worse. What a time to be alive."

---

Last Updated: 2026-02-22