⚙️ Under the hood
RLHF
Reinforcement learning from human feedback: people rate answers, and the model learns to be more helpful and safer.
💭 Imagine …
Like an intern getting feedback after every task: “Good. Shorter, please. That's a no-go.”
🔎 Example
That's why assistants answer politely and structured, and decline some requests.
🎯 Why it matters
Explains an assistant's “character”, and why models are sometimes too agreeable.
🎓 Prefer step by step?
The AI Literacy course: 6 short modules with quizzes, a solid foundation in about 60 minutes.