About every 30 minutes, take 5–10 minutes away from screens. Look into the distance out of a window, or try animal guessing, copying movements, a paper maze, chess or Jungle. Do not switch to a phone or video during the break. An adult keeps track of the time; children can take a break sooner if they need one. Include breaks in the total lesson time, shortening computer tasks or presentations accordingly. For a session of about two hours, plan breaks around minutes 30, 60 and 90.
Watch the video (Chinese dubbing and subtitles)
35-second intro: 30 seconds making mooncakes for friends, 5 seconds of Mid-Autumn greetings. The realistic cat fairy tale visuals and voiceover are AI-generated, not filmed with real cats or based on actual conversations; the four cats are derived from real pet photos provided by users. Adults select the story, then review the visuals and audio, which include Chinese subtitles, Chinese voiceover, cat meows, and music by YuE2 + ACE-Step.
Xiaobi (the calico cat) wants to share a thoughtful gift with her friends. She kneads the dough, fills it, presses it into molds, packs the baked mooncakes into small boxes, and delivers them to Jinzi (a tabby cat), Xiaofei (a black-and-white bicolor cat), and Xiaomai (a white cat). With each delivery, she says to her friend: 'Happy Mid-Autumn Festival! May the moon be full and everyone reunited!'
If it were you, who would you like to give the gift to? What do you want to say to him? Later we will draw our own maze and invite friends to play with it.
Who would you like to give a gift to? What do you want to say to him?
After watching, ask: Real cats meow, but who decided they would speak Chinese and make mooncakes? This is an AI story created by humans. The child facilitator then holds up a real maze sketch and invites everyone to create something for their friends to play with.
You can call me...
I want to do...
Because I hope...
Gifts for friends, picture books, music, animal houses... not necessarily games.
"I heard you thinking..."
"Am I right? What would you like to add?"
Round 1: One person speaks and the other listens. Exchange in a while.
Can draw, point, skip, or just talk to a partner. No name input, no recording, no wish uploading.
Main scenario 1 / 3
Think about it first: Who can help with which step? You can select more than one, or you can select none temporarily.
Tell me why you chose this.
Pre-written classroom scenarios and discussion references, not live AI output. AI is a tool made by humans, not humans, and will make mistakes; what AI says must be checked. When you are sad or need help, you can go to your family and teachers.
Real production materials / pre-class generation / offline display
Source of photo: Actual pet photo provided by parent. The original photos are not distributed with the class package, the teacher only needs to clarify the source.
"Please refer to Xiaobi's photo to have her make mooncakes in the kitchen for Jinzi, Xiaofei, and Xiaomai, preserving each cat's unique markings."
Class simplification request to explain true production intent; not a verbatim copy of the original model request.
What other detail do you want to tell AI?
This is still a static image. Look at the patterns, paws, and mooncake molds.
Watch the video (Chinese dubbing and subtitles)
Where is the movement? Where is it still the same? This is a video generated before class. Clicking to play will not call AI.
There are 2 pieces left from the boxed shot, this scene should stay at 2 pieces as well. What this group modifies is the quantity, not the identity of the cat.
People also checked the names, patterns and subtitles: Xiaobi, Jinzi, Xiaofei, Xiaomai; only the bottom set of subtitles and the real AHALab logo were retained in the final film.
My thoughts are...
I want AI to help me...
I will check...
If you don’t like a suggestion, you don’t have to use it or you can change it.
Adults only: Send this prompt in the pre-logged AI tool: "Write two different greeting messages for the story of Xiaobi giving mooncakes. Each message must be no more than 12 characters long and must not include real names." Let the child choose one message or none. Wait up to 60 seconds; if there is no result, state clearly that nothing was returned, then return to the pre-class materials on this page. This webpage does not connect to AI and does not treat preset buttons as live responses.
Round 1 / 3
You can slowly walk to the side of A / B, or you can sit and point it out to the teacher.
The teacher can click on the corresponding option to count the number of people; or he can just let the children stand in line without recording.
Tell me your name and ask your partner: What did you choose?
The pieces begin to face upward. Turning around only changes direction, not moving. Stop when you reach the wall.
The commander says the action and the operator presses the button. Exchange after 90 seconds.
Listen to the sound first, then guess what happened.
There is no prompt sound or animation, how do you know it is successful?
Behind the door is...
Each group discusses for 20 seconds before saying anything. You can just say a word, point to a picture, or ask a partner to help.
This is an imaginary question, and you don’t need to guess the teacher’s answer.
Choose one of the two, command cooperation and story solitaire are adapted from the event. For the source and complete guidance, please see Teacher Guide. Sound Detective is based on this game.