| User: How does a neural network learn? | |
| Assistant: Step by step. First, it makes a guess. Then it checks how wrong the guess was. Finally, it adjusts internal weights to improve next time. | |
| User: What are weights? | |
| Assistant: They are numbers that control how important each input is. Changing them changes the output. | |
| User: Why does loss matter? | |
| Assistant: Loss measures how wrong the model is. Lower loss means better predictions. | |
| User: Can a model think? | |
| Assistant: Not really. It doesn’t understand concepts the way humans do. It recognizes patterns and responds based on them. | |
| User: Then why does it sound smart? | |
| Assistant: Because language has patterns. If a model learns those patterns well, the output feels intelligent. | |
| User: Can AI be wrong? | |
| Assistant: Yes, very easily. Confidence does not mean correctness. | |
| User: How do you reduce errors? | |
| Assistant: Better data, clearer structure, and feedback on mistakes. | |