GENAI 101 - Module 2: Theater
Lesson 2.3: Digital Choreography and Drama-Dance
Estimated time: 90 minutes. Competencies: problem solving, creativity and innovation.
Video: Lesson 2.3: Digital Choreography and Drama-Dance. Four-minute introduction: the elements of dance as a prompt frame, expecting and diagnosing failure, and composition as your job. (Instructor: build with make_videos.py or record and embed your own.)
Learning objectives
By the end of this lesson you will be able to:
- Translate a dramatic idea into a movement score using the elements of dance: body, action, space, time, energy.
- Generate short movement clips with a text-to-video tool and diagnose failures using the three-fix method.
- Assemble a 30 to 60 second drama-dance piece that combines generated video with your own movement direction.
- Explain how the piece would translate back to live bodies in a rehearsal room.
Warm-up: move the sentence (5 minutes)
Stand up. Take the sentence "I am not going back" and perform it three times: once with only your hands, once with only your feet, once with your whole body but no sound. Notice that each version changed the meaning. Movement is a language with its own grammar, and the five elements below are its parts of speech.
Key ideas
The elements of dance as a prompt frame
Text-to-video models respond to the same kind of specificity that music and text models do. Dance educators already have the vocabulary:
| Element | Question | Prompt language |
|---|
| Body | Which parts lead? What shape? | "arms lead, torso curved, one dancer" |
| Action | What movement verbs? | "reach, collapse, spiral, freeze" |
| Space | Level, direction, pathway, size | "low level, moving backward in a wide arc" |
| Time | Tempo, rhythm, duration | "slow, sustained, one held pause" |
| Energy | Quality of movement | "heavy and bound, then sudden and light" |
A prompt that names all five ("one dancer in an empty studio, arms leading a slow backward spiral at low level, heavy and bound, then a sudden light reach upward and freeze, 5 seconds") gives the model something to render. "A sad dance" does not.
Expect failure, then diagnose
Text-to-video is the least reliable tool in this course. Bodies drift, hands multiply, and the model may ignore your tempo. This makes it the best place to practice problem solving. For every unusable clip, name the failure and pick the fix:
- Revise the prompt. Reduce to one dancer, one action, one quality. Add "single continuous shot" and a duration.
- Change the tool or mode. Try an image-to-video mode with a still you make in Canva, or switch to Animaker's character animation for precise control.
- Edit the output. Trim to the two usable seconds, slow it, mirror it, or cut on the beat. Editing turns a failed generation into a usable phrase.
Composition is yours
The model makes phrases; you make the piece. Choreographic devices such as repetition, canon, contrast, and accumulation are decisions about sequence and relationship, and they belong to you. A piece built from four generated phrases arranged with intention is a composition. The same four phrases dumped in order are a slideshow.
Activity: build a drama-dance piece (65 minutes)
- Choose a dramatic idea. One sentence, such as "A student decides to tell the truth." Your Lesson 2.1 scene may supply one.
- Write a movement score of four phrases. For each, write one line using all five elements. Decide on the choreographic device that will link them (repetition, canon, contrast, or accumulation).
- Generate phrase 1 in invideo AI (or another text-to-video tool). Generate twice. Record what failed and which fix you applied.
- Generate phrases 2 to 4. Apply what you learned from phrase 1. Keep a log of failures and fixes; the log is part of the submission.
- Assemble. In invideo AI's editor, Canva, or CapCut, sequence the phrases using your device. Add a soundtrack (your Lesson 1.3 cue if you completed the music module, or a generated instrumental).
- Write the transfer note. In three sentences, describe how you would teach these four phrases to live dancers or actors in a rehearsal room, and what the video showed you that improvisation would not have.
- Export as MP4 with captions or title cards naming each phrase. Name it
Lesson-2-3_LastName_Title.mp4.
Failure and fix log
| Phrase | Generation | What failed | Fix applied | Result |
|---|
| 1 | 1 | extra limbs, tempo ignored | revised prompt: one dancer, "slow, single shot" | usable 3 seconds |
| 1 | 2 | | | |
Create and share
Post in the Lesson 2.3 Gallery:
- The MP4 or share link.
- Your movement score (four phrases with all five elements) and the choreographic device.
- Your failure and fix log.
- Your three-sentence transfer note.
- A rationale of no more than 75 words: the fix that taught you the most.
- Your one-line AI credit.
Reply to two peers: name the choreographic device you see in their piece, and suggest one way the piece could be performed by live bodies.
Resources
- invideo AI; Canva video and Animaker as alternatives.
- Elements of dance reference: the National Dance Education Organization standards.
- Your Lesson 1.3 cue, if available, for the soundtrack.
Reflection prompts
- Which element of dance did the model handle worst, and how did you compensate?
- Did generating movement change how you think about directing movement?
- What safety, consent, or body-image conversations would this activity require in a classroom?