What Students Explore
- How prompts, inputs, and outputs shape chatbot behaviour.
- Why AI should be honest when it does not know something.
- How safety rules protect private information.
- How testing helps improve an AI project.
Lesson 1 Preview
See how students customize a chatbot called Byte, test safety rules, and learn why AI needs careful prompts and checking.
In the classroom version, students do more than chat. They write rules, test edge cases, and discuss how Python can connect user input to an AI service.
Live AI Preview
Change one role sentence and one safety rule, then test how the chatbot responds. This is a simplified preview of the guided classroom activity.
Safety first: do not enter your real name, password, address, phone number or other private information.
安全第一:請勿輸入真實姓名、密碼、地址、電話號碼或其他私人資料。
Step 1
Choose a topic, then tune one role sentence and one safety rule before you test the chatbot.
Step 2
Tell Byte how to behave for this activity.
140 characters left
Add one rule that keeps the answer safe and honest.
180 characters left
This demo keeps sensitive settings on the server. Visitors can only edit the short role and rule above.
Step 3
Ready for short, safe questions.
Byte is thinking...
Keep it under 300 characters and avoid private information.
Try one of these
This public demo uses a real AI model, so responses may occasionally be incorrect. During the course, students learn how to test and improve AI rather than trust every answer.
此公開 Demo 使用真正的 AI 模型,因此回答有時可能不正確。課程會教導學生如何測試及改善 AI,而不是盲目相信每個答案。
The public demo does not save chat conversations.