L1
·
Quiz
·
Lab
L2
·
Quiz
·
Lab
L3
·
Quiz
·
Lab
L4
·
Quiz
·
Lab
Module Test
Module 3 · Lesson 1

Motion Capture and Machine Learning

How AI systems learn to read, analyze, and generate human movement
Can a machine understand what a body is saying — before the dancer does?

In 2019, choreographer Wayne McGregor collaborated with Google Arts & Culture Lab on a project called Living Archive. The system was trained on 25 years of McGregor's choreographic footage — thousands of hours of rehearsal and performance video. The AI learned the statistical patterns of his movement vocabulary: which gestures followed which, how his dancers transitioned from grounded to aerial, the characteristic way extension moved through a spine in his work.

When McGregor's dancers fed the system a movement seed — a single gesture performed live via motion capture — it generated novel movement sequences in his recognizable style. The dancers then chose which AI-generated sequences to embody, creating a feedback loop between human intuition and algorithmic suggestion.

How Motion Capture Works with AI

Motion capture (mocap) translates physical movement into numerical data. In optical systems like those used in film and advanced dance research, reflective markers are placed on joints — hips, knees, wrists, shoulders — and infrared cameras triangulate their position in 3D space up to 240 times per second. The result is a skeleton of coordinates moving through time.

Inertial mocap suits (such as those made by Xsens, used at institutions like the Norwegian National Ballet and The Royal Ballet) embed accelerometers and gyroscopes in the suit itself, enabling wireless capture without camera rigs. These produce joint angle data rather than positional data — a different but complementary representation of movement.

AI models — typically recurrent neural networks (RNNs) or transformer architectures — are trained on large libraries of mocap data. They learn the conditional probability of movement: given a sequence of poses, what pose is statistically likely next? This is structurally identical to how large language models predict the next word.

Key Distinction

A motion-prediction AI does not "understand" dance aesthetically. It identifies statistical regularities in movement data. The artistic interpretation of which suggestions are meaningful remains entirely with the choreographer and dancer.

Vocabulary of the Field
Mocap SkeletonThe hierarchical representation of a body as a set of joints and bones, each with position and rotation values across time — the raw data that AI models train on.
Movement VocabularyThe characteristic set of gestures, transitions, and spatial patterns that define a choreographer's or company's identifiable style.
Pose EstimationAI techniques (such as Google's MediaPipe or OpenPose) that extract skeletal data from ordinary video without specialized markers — democratizing movement analysis.
Latent SpaceThe compressed internal representation a neural network uses to encode movement patterns. Points close together in latent space represent movements that are stylistically similar.
Real-World Applications Beyond McGregor

The Chunky Move company in Melbourne used computer vision-driven reactive environments as early as 2006 in their production Glow, where a dancer's movement directly triggered projected light patterns through real-time tracking — an early demonstration of movement-responsive AI environments in performance.

At MIT Media Lab, researchers developed Choreographic Language Agent (CLA), a system that could translate high-level choreographic instructions ("move as if resisting invisible water") into movement sequences, bridging the gap between verbal artistic intent and physical output.

The Batsheva Dance Company in Tel Aviv has integrated motion analysis tools to study how founder Ohad Naharin's Gaga movement language propagates through a company — using AI to map which physical principles are most consistently reproduced across dancers and rehearsals.

Ethical Dimension

When an AI is trained on a specific choreographer's movement archive, questions of ownership arise: if the system generates a "Wayne McGregor-style" phrase, who holds authorship? These questions remain legally unresolved in most jurisdictions, though McGregor's project was framed explicitly as a collaborative research tool rather than an autonomous creator.

Lesson 1 Quiz

Motion Capture and Machine Learning · 4 questions
1. What kind of data does optical motion capture primarily produce?
Correct. Optical mocap triangulates the 3D position of reflective markers using infrared cameras, producing coordinate data sampled at high frequency.
Not quite. Optical mocap produces 3D positional coordinate data for joint markers — not audio or color annotation.
2. Wayne McGregor's Living Archive project trained an AI on how many years of his choreographic footage?
Correct. The system was trained on 25 years of McGregor's rehearsal and performance footage, giving it a rich corpus of his movement vocabulary.
Not quite. The Living Archive project trained on 25 years of McGregor's choreographic footage, accumulated since the founding of his company.
3. What is a "latent space" in the context of movement AI?
Correct. In a neural network's latent space, movements that share stylistic or structural qualities are encoded as nearby points — enabling navigation and interpolation between movement styles.
Not quite. Latent space is a compressed internal representation within the neural network, where proximity reflects stylistic similarity between movements.
4. How is an AI movement-prediction model structurally similar to a large language model?
Correct. Movement prediction models and language models share the same core logic: given a preceding sequence, what is the statistically probable next element — whether that element is a word or a pose.
Not quite. The structural similarity is in next-element prediction: given a sequence, both models predict the most likely continuation — words for LLMs, poses for movement models.

Lab 1 · Movement Data Analysis

Explore how AI reads and interprets body movement data

Your Task

You are advising a mid-size contemporary dance company that wants to digitize their artistic director's 30-year movement archive and build an AI choreographic assistant. Explore the possibilities, technical requirements, and artistic implications with your AI consultant.

Starter prompt: "We have 30 years of rehearsal video in various formats — VHS, MiniDV, and HD digital. What would we actually need to do to train a movement AI on this archive, and what could such a system realistically do for our choreographers?"
AI Choreography Consultant
Lab 1
Welcome. I'm your AI choreography technology consultant. I can help you think through the process of building a movement AI from a dance archive — covering digitization, data preparation, model training, and what the resulting system could realistically achieve. What would you like to explore first?
Module 3 · Lesson 2

Generative Choreography Systems

AI tools that propose, extend, and transform movement sequences
When an algorithm generates a dance phrase, is it creating — or calculating?

In 2022, choreographer Phoebe Sengers and the research collective moves.ai presented work using a system called Choreographic Language Agent at the Barbican's AI exhibition. The system could receive abstract choreographic instructions in plain language — "contract as if absorbing a blow," "expand through the fingertips as though releasing heat" — and generate corresponding movement sequences for a virtual dancer, which human performers then used as a compositional starting point.

Separately, the Motion Diffusion Model (MDM), published by researchers at Tel Aviv University in 2022, demonstrated that diffusion-based generation — the same approach behind image generators like Stable Diffusion — could produce realistic, coherent full-body movement sequences from text descriptions. MDM generated 196-frame sequences (roughly 6.5 seconds at 30fps) that human evaluators rated as more natural than previous GAN-based approaches.

Architecture of Generative Movement Models

Several distinct architectural approaches have emerged for AI-generated choreography:

Recurrent Neural Networks (RNNs / LSTMs) were the dominant approach through roughly 2019. They generate movement autoregressively — one pose at a time, each conditioned on all previous poses. This works well for short sequences but tends to drift into repetitive or physically implausible postures over time.

Generative Adversarial Networks (GANs), specifically those trained on motion data (like motion-GAN), use a generator/discriminator architecture. The generator produces movement sequences; the discriminator tries to distinguish generated from real. This produces diverse output but can be unstable and difficult to control directionally.

Diffusion Models (the current frontier, as of 2024) work by learning to denoise movement sequences from pure noise. They accept conditioning signals — text, a starting pose, a musical signal — and iteratively refine noise toward movement that satisfies the condition. MDM, MotionDiffuse, and MLD (Latent Diffusion for Motion) all use variants of this approach.

Transformer-based models treat a sequence of poses the way GPT treats a sequence of tokens — attending to relationships across the entire sequence simultaneously. Transflower and TEACH use transformers to generate temporally coherent movement with better long-range consistency.

Training Data Reality

Most publicly available movement AI models are trained on HumanML3D (14,616 motion clips with text annotations) or the HumanAct12 dataset. These datasets are heavily weighted toward pedestrian and athletic movement, meaning systems often struggle with stylized or abstract dance vocabulary unless fine-tuned on domain-specific data.

Commercial and Institutional Deployments

The animation industry has been an aggressive adopter. NVIDIA Omniverse integrates motion-synthesis tools that game developers and virtual production teams use to generate secondary character movement. Adobe Mixamo uses AI to auto-rig 3D models and blend motion-captured animations — tools increasingly used by choreographers prototyping work in virtual environments.

In 2021, the Paris Opera Ballet collaborated with researchers on a system that could generate ballet vocabulary — specifically classical port de bras and battement sequences — from stylistic parameters, used internally by the company's pedagogy division to explore what AI-generated classical technique looked like and where it broke down.

The National Ballet of Canada partnered with AI research teams to create interactive installations where audience movement, captured by overhead cameras, influenced real-time choreographic projections — not full generative choreography but demonstrating the pipeline from live human movement to AI-mediated performance response.

Key Terms
Autoregressive GenerationProducing output one element at a time, each conditioned on all prior elements — the approach used by RNN-based choreography systems and by most language models.
Diffusion ModelA generative model trained to reverse a gradual noising process, iteratively refining noise toward coherent output conditioned on a text prompt or other signal.
Conditioning SignalThe input that guides what a generative model produces — could be a text description, a musical excerpt, a starting pose, or a style label.
Fine-TuningAdditional training of a pre-trained model on domain-specific data — essential for making general movement models perform well on specific dance styles.
Artistic Insight

Choreographers who have worked with generative systems consistently report that the most valuable output is not "good" movement but unexpected movement — phrases that would not have emerged from their own embodied intuition. The AI functions as a productive stranger in the creative process, suggesting directions the choreographer would not naturally take.

Lesson 2 Quiz

Generative Choreography Systems · 4 questions
5. What is the primary limitation of RNN-based movement generation over long sequences?
Correct. RNNs generate movement autoregressively and accumulate errors over time, leading to repetition or physically implausible configurations in longer sequences.
Not quite. The key limitation is temporal drift — as the sequence lengthens, RNNs tend toward repetitive or physically impossible poses.
6. The Motion Diffusion Model (MDM) from Tel Aviv University was notable for applying which generative approach to movement?
Correct. MDM applied diffusion model architecture to movement generation, producing sequences from text descriptions rated more natural than previous GAN approaches.
Not quite. MDM used diffusion-based generation — iteratively denoising movement sequences conditioned on text input, analogous to how Stable Diffusion generates images.
7. Why do most general movement AI models struggle with stylized dance vocabulary?
Correct. Public movement datasets are weighted toward everyday and athletic motion. Without fine-tuning on specific dance corpora, models produce movement that looks like a walking human rather than a trained dancer.
Not quite. The core issue is training data distribution — publicly available datasets are dominated by everyday movement, not dance vocabulary.
8. What do choreographers most frequently identify as the creatively valuable output from generative AI movement systems?
Correct. Choreographers consistently report that the greatest value is not "good" movement but movement they would never have generated themselves — the AI as productive stranger in the creative process.
Not quite. The consistent finding from practitioners is that the value lies in unexpected, non-intuitive movement suggestions — directions the choreographer's own embodied history would not naturally generate.

Lab 2 · Generative Movement Design

Design a generative choreography system for a specific artistic context

Your Task

You are a choreographer who wants to use a text-to-movement AI system to develop a new evening-length work about geological time — the slow movement of continents, glaciers, and tectonic plates translated into human-scale dance. Work with your AI collaborator to design your approach.

Starter prompt: "I want to use a text-to-movement AI to help me create a 70-minute work about geological time. What kinds of text prompts would generate the most useful movement material, and how would I structure the creative process between AI generation and human embodiment?"
AI Choreographic Collaborator
Lab 2
Geological time as a choreographic subject is a profound choice — the challenge of making imperceptible movement perceptible through the body. I can help you design prompting strategies for text-to-movement AI and think through how to structure the human-AI creative dialogue. Where would you like to start — with the prompting language itself, or with the broader workflow?
Module 3 · Lesson 3

Real-Time Responsive Performance

AI systems that watch, interpret, and respond to live dance in performance
What changes when the audience and the algorithm both watch — and one of them responds?

At Ars Electronica 2018, the company Random International presented an evolution of their ongoing research into behaviour-following objects. While not exclusively focused on dance, their systems — which track and mirror human movement through robotic and projected elements — demonstrated how real-time pose estimation could create a convincing sense of reciprocal attention between a human and a system.

More directly choreographic: Garth Paine's long-running series of interactive performance works used real-time audio-visual systems that responded to dancer movement, creating environments where the performer genuinely did not know in advance what sound or light would respond to a given gesture — requiring a fundamentally different rehearsal methodology, one built on improvisation with an unknown partner.

The Technical Architecture of Real-Time Response

A real-time responsive performance system requires three components working in sub-100ms latency for the response to feel present rather than delayed:

1. Input Layer — A sensor system tracking the performer. Options range from professional mocap (high accuracy, high infrastructure cost, markers required) to depth cameras like the Microsoft Kinect or Intel RealSense (markerless, moderate accuracy) to ordinary RGB cameras processed through pose-estimation models like MediaPipe BlazePose or OpenPose. The choice determines what movement data is available and how precisely.

2. Interpretation Layer — AI models that classify, recognize, or respond to movement. This might be a classifier that identifies which of a set of choreographic states a dancer is in (enabling branching narrative structures); a regression model that maps movement qualities to parameter values; or a generative model that proposes responses in real time.

3. Output Layer — The medium through which the system responds. Sound (through systems like Max/MSP or SuperCollider), light (through DMX-controlled rigs), projection (through TouchDesigner or MadMapper), robotics, or — in the most experimental cases — generation of movement instructions displayed to other performers.

Latency Problem

Human perception of synchrony requires responses within roughly 100ms of a movement stimulus. At 30fps, a single frame takes 33ms. Processing a full pose estimation, running a model inference, and triggering an output must happen within 3 frames for the response to feel immediate. This is why many real-time performance systems use simpler models (classifiers, rule-based triggers) rather than full generative AI, which requires more computation.

Documented Performance Works

Glow (Gideon Obarzanek / Chunky Move, 2006) — A dancer's movement triggered a generative projected light environment computed by a custom computer vision system. Notably created in 2006, this work predated modern deep learning but demonstrated the aesthetic potential of real-time movement-responsive AI environments. The work toured internationally and influenced subsequent generations of choreographers working with technology.

Zero Gravity (Adrien M / Claire B, 2014) — French company AM/CB created a series of works where performers interact with projected particle systems that behave as physically simulated entities responding to the dancer's presence and movement. The system uses depth cameras and custom physics simulation, creating the appearance that the dancer is physically manipulating light.

Body Navigation (Palindrome Inter-media Performance Group) — Over more than two decades, Palindrome developed progressively more sophisticated systems for movement-responsive performance, including early work with eye-tracking and full-body motion analysis feeding into sound and video synthesis.

The IDC Research Group at Goldsmiths, University of London, has produced extensive documentation of interactive dance performance works, analyzing how performers adapt their movement vocabulary when they know it is producing audible or visible consequences in real time — finding consistent evidence of what researchers call kinesthetic listening: heightened somatic awareness induced by perceived environmental responsiveness.

Rehearsal and Artistic Methodology
System CalibrationThe process of tuning a responsive system to a specific performer's movement range and qualities — analogous to tuning an instrument to a musician's touch.
Choreographic StatesDiscrete conditions the AI can classify a performance as being in — allowing branching narrative structures where the performer's choices determine what the audience experiences.
Kinesthetic ListeningHeightened somatic awareness in performers who know their movement is producing environmental consequences — documented by researchers as a distinct phenomenological state.
Lag CompensationArtistic strategies that account for system latency — composing movement with built-in pauses, using sound or light that legitimately begins after the trigger, or accepting latency as an aesthetic element.
The Authorship Reversal

In traditional performance, the choreographer authors movement in rehearsal and the audience receives it in performance. In real-time responsive AI performance, the system authors its response in real time, the performer makes choices that they know will be responded to but cannot fully predict, and the audience witnesses an encounter rather than a presentation. The choreographer's role shifts from author of movement to designer of a system of relationships.

Lesson 3 Quiz

Real-Time Responsive Performance · 4 questions
9. What latency threshold must a real-time performance system generally meet for its response to feel present rather than delayed?
Correct. Human perception of synchrony requires responses within roughly 100ms — at 30fps, this means the full pipeline (sensing, inference, output) must complete within approximately 3 frames.
Not quite. The threshold is approximately 100ms. Beyond this, the response feels delayed rather than simultaneous — breaking the perceptual contract of liveness.
10. Which Chunky Move work, created in 2006, is considered a landmark in real-time movement-responsive AI performance environments?
Correct. Glow (2006) by Gideon Obarzanek / Chunky Move used computer vision to trigger a generative projected light environment from a dancer's movement — a landmark work that influenced subsequent practitioners.
Not quite. Glow (2006) by Chunky Move is the landmark work — using computer vision to create a dancer-responsive projected environment, pre-dating modern deep learning but demonstrating the aesthetic possibilities.
11. What is "kinesthetic listening" as documented by researchers studying interactive dance performance?
Correct. Researchers at institutions including Goldsmiths documented that when performers know their movement triggers environmental responses, they enter a distinct phenomenological state of heightened somatic attention.
Not quite. Kinesthetic listening describes the heightened somatic awareness performers experience when they know their movement is consequential — producing sound, light, or other environmental effects.
12. Why do many real-time performance systems use simple classifiers rather than full generative AI?
Correct. The latency constraint is demanding — full generative inference often takes too long for real-time response, while a pre-trained classifier can return a result in milliseconds.
Not quite. The reason is computational: generative AI inference takes longer than classification, making it difficult to meet the ~100ms perceptual synchrony threshold in live performance.

Lab 3 · Real-Time System Design

Architect a responsive AI performance environment

Your Task

You are designing a solo performance work where a single dancer performs in an environment that responds to their movement in real time. The work explores themes of grief and memory — the environment should feel like it is remembering, distorting, and echoing the dancer's movement. Work with your consultant to design the technical and artistic system.

Starter prompt: "I want to design a real-time responsive environment for a solo about grief. The environment should feel like it has memory — echoing, distorting, accumulating the dancer's movement over time. What technical components would I need, and how would the AI layer work?"
Interactive Performance Consultant
Lab 3
Memory and accumulation in a real-time responsive environment — this is one of the richest areas of interactive performance design. Systems that accumulate, echo, and distort create a genuinely different dramatic relationship than systems that simply respond and reset. Let's think through how to build this technically and artistically. What's your current technical infrastructure and budget range?
Module 3 · Lesson 4

Authorship, Labor, and the Choreographer's Role

Ethical and professional questions as AI enters the choreographic studio
If an AI generates the movement, who made the dance?

In 2018, New York City Ballet partnered with machine learning researchers from Google to create an algorithmic ballet. The project used machine learning to analyze NYCB's existing repertoire and generate new choreographic sequences. Choreographer Justin Peck then worked with the AI-generated sequences, selecting, modifying, and staging them with the company's dancers.

The project drew immediate critical attention — not primarily for its artistic outcomes, but for the questions it raised. The American Guild of Musical Artists (AGMA), which represents ballet dancers, raised concerns about whether AI-generated choreography could eventually reduce demand for human choreographers and, by extension, for rehearsal time, rehearsal directors, and the ancillary creative labor that surrounds traditional choreographic processes.

The Authorship Question: Legal and Artistic Dimensions

Copyright law in the United States and most jurisdictions requires human authorship for copyright protection to attach. In 2023, the US Copyright Office confirmed this position explicitly, denying copyright to AI-generated images and text that lacked sufficient human creative input. Choreography is explicitly listed as a copyrightable category under US law (17 U.S.C. § 102) — but only when it involves human creative expression fixed in a tangible medium.

This creates a practical ambiguity: if a choreographer prompts an AI for movement sequences and performs them, is the work copyrightable? Legal consensus is forming around the idea that the human's creative choices — which prompts to use, which outputs to select, how to arrange and stage the material — constitute sufficient authorship. But the AI-generated sequences themselves, in isolation, likely are not protectable.

The Society of Stage Directors and Choreographers (SSDC) in the US has not yet issued formal guidance on AI-generated choreography but has begun internal discussions mirroring those of the writers' guilds following the 2023 WGA and SAG-AFTRA strikes, which secured protections against AI replacing human writers and performers.

The Training Data Problem

AI movement models trained on professional dance footage raise an additional concern: the movement of professional dancers — their technique, their training, their artistic development — may be embedded in the model without consent or compensation. Unlike text, where individual sentences may be difficult to attribute, a professional dancer's distinctive movement quality can be measurable and, in principle, attributable. No legal framework currently addresses this.

Labor Implications

The choreographic labor ecosystem extends well beyond the choreographer: rehearsal directors, notators, repetiteurs, movement coaches, dramaturgs, and rehearsal pianists all support the creation and transmission of dance works. AI tools that compress the ideation and drafting phase of choreography could reduce the time — and therefore the paid hours — for some of these roles.

Conversely, early adopters within the field argue that AI tools function more like notation software than like replacement labor: they expand what is possible within a given budget rather than directly substituting for human roles. The analogy is to how DAW (digital audio workstation) software changed music production — it displaced some roles while enabling solo artists to produce work that previously required large teams, and creating entirely new professional categories.

In 2023, the International Federation of Arts Councils and Culture Agencies (IFACCA) published a working paper on AI in the performing arts, noting that impacts on choreographic labor were highly context-dependent: smaller companies with limited budgets saw potential democratization, while established companies with unionized workforces faced more complex contractual questions.

Emerging Frameworks for Ethical Practice
Informed Consent in TrainingThe principle that performers and choreographers whose work is used to train AI movement models should be informed and should have the option to consent or refuse — analogous to informed consent in biomedical research.
Attribution TransparencyThe practice of disclosing when AI tools contributed to a choreographic work — in program notes, promotional materials, and grant applications — as a developing professional norm.
Residual and Reuse RightsContractual mechanisms (modeled on those in film and television) that could entitle dancers and choreographers to compensation when their captured movement data is reused in AI training.
Creative Control ProvisionsContractual language specifying that no AI-generated material may be staged in a production without the choreographer's explicit selection and modification — preserving human creative authority.
The Longer View

Every significant technological shift in choreographic tools — from Laban notation to video recording to digital editing to virtual reality — has raised analogous concerns about authenticity, authorship, and labor. The consistent historical pattern is that the technology neither destroys the art form nor is absorbed without change: the field adapts, new aesthetics emerge, and the definition of what choreography means expands. The ethical work is in ensuring that adaptation is equitable — that the benefits of new tools do not accrue only to those who already hold power in the field.

Lesson 4 Quiz

Authorship, Labor, and the Choreographer's Role · 4 questions
13. Which union raised concerns about the NYCB Algorithm Project's potential labor implications for choreographers and creative workers?
Correct. AGMA, which represents ballet dancers, raised concerns about AI-generated choreography potentially reducing demand for human choreographers and the associated creative labor infrastructure.
Not quite. It was AGMA — the American Guild of Musical Artists, representing ballet dancers — that raised concerns about the labor implications of the NYCB Algorithm Project.
14. Under current US copyright law, what is the status of AI-generated choreographic sequences produced without sufficient human creative input?
Correct. The US Copyright Office confirmed in 2023 that copyright requires human authorship. AI-generated material without sufficient human creative input is not protectable, including choreographic sequences.
Not quite. US copyright requires human authorship. The Copyright Office confirmed in 2023 that AI-generated material without meaningful human creative input cannot receive copyright protection.
15. What historical analogy do proponents of AI choreography tools most frequently use to argue these tools will not simply replace human labor?
Correct. The DAW analogy is the dominant one in this discourse: music production software expanded what was possible within a budget and created new categories of work, rather than simply eliminating existing roles.
Not quite. The most common analogy is to DAW software in music — which changed who could produce music and how, displaced some roles, created others, and enabled solo artists to produce work previously requiring large teams.
16. What does "attribution transparency" mean as an emerging professional norm in AI-assisted choreography?
Correct. Attribution transparency as a professional norm means choreographers disclose when AI tools contributed to a work — in programs, marketing, and funding applications — as an emerging practice of honesty about creative process.
Not quite. Attribution transparency is a professional practice norm: the choreographer discloses AI tool contributions in programs and other materials, being honest about the creative process with audiences and funders.

Lab 4 · Ethics and Policy Design

Develop ethical guidelines for AI use in a dance organization

Your Task

You are the artistic director of a mid-size contemporary dance company. Your board has asked you to draft a policy on AI tools in choreographic work — covering authorship disclosure, training data consent, labor protections, and when AI tools may and may not be used. Work with your AI advisor to develop this policy.

Starter prompt: "I need to draft an AI policy for my dance company. We want to be open to using AI choreographic tools but need to protect our dancers, choreographers, and artistic integrity. What should a comprehensive policy cover, and what are the most urgent provisions I should address first?"
Arts Policy Consultant
Lab 4
Developing an AI policy for a dance company is genuinely important work — you'll be operating ahead of most legal frameworks and industry standards, which means you have both an opportunity and a responsibility to set thoughtful precedent. I can help you think through the key provisions: authorship and attribution, training data consent, labor protections, and permitted versus restricted uses. Where would you like to start, and what's the most pressing concern your board raised?

Module 3 Test

AI in Choreography · 15 questions · Pass at 80%
1. What does optical motion capture use to track joint positions?
Correct. Optical mocap uses infrared cameras to triangulate the 3D positions of reflective markers placed at key joints.
Optical mocap uses infrared cameras to triangulate reflective markers at high sample rates — not ultrasound, electromagnetic, or audio sensors.
2. Wayne McGregor's Living Archive was developed in collaboration with which technology partner?
Correct. Living Archive was a collaboration between Wayne McGregor and Google Arts & Culture Lab, completed in 2019.
The collaboration was between Wayne McGregor and Google Arts & Culture Lab — not Microsoft, MIT, or IBM.
3. Inertial mocap suits produce which type of data, as distinct from optical mocap?
Correct. Inertial mocap suits use accelerometers and gyroscopes to measure joint angles — a different representation from the positional coordinates of optical systems.
Inertial mocap measures joint angles via accelerometers and gyroscopes — not world-space positions, biometrics, or thermal data.
4. Which architecture does the Motion Diffusion Model (MDM) use to generate movement?
Correct. MDM uses diffusion-based generation — iteratively denoising movement sequences from noise, conditioned on text descriptions.
MDM uses diffusion architecture — the same class as Stable Diffusion for images — not LSTMs, GANs, or rule-based expert systems.
5. What is the primary reason most public movement AI datasets produce models that struggle with specialized dance vocabulary?
Correct. Datasets like HumanML3D are weighted heavily toward everyday and athletic movement — fine-tuning on dance-specific data is required for stylized dance vocabulary.
The core issue is data distribution — public datasets contain mostly walking, running, and athletic movement, not specialized dance styles.
6. In the context of generative choreography, what do choreographers most consistently report as AI's unique creative value?
Correct. Practitioners consistently identify the value as the unexpected — movement the AI suggests that the choreographer's own movement history would not naturally generate.
The consistent finding is that AI's value lies in productive surprise — movement outside the choreographer's habitual creative territory.
7. The three components of a real-time responsive performance system are input layer, interpretation layer, and what third component?
Correct. The output layer is the medium of response — sound, light, projection, or robotic systems — that translates the AI's interpretation into a perceptible performance element.
The third component is the output layer — the medium (sound via Max/MSP, light via DMX, projection via TouchDesigner) through which the system responds to the dancer.
8. Glow (2006) by Chunky Move is significant in the history of responsive AI performance primarily because:
Correct. Glow was a landmark demonstration of dancer-responsive projected environments using computer vision — pre-dating deep learning but establishing aesthetic possibilities that influenced subsequent practitioners.
Glow was significant for demonstrating the aesthetic potential of movement-responsive environments through computer vision, establishing possibilities that shaped subsequent work in the field.
9. What perceptual threshold must a real-time performance system generally meet for its responses to feel simultaneous rather than delayed?
Correct. Human perception of synchrony requires responses within approximately 100ms — a demanding constraint that shapes which AI architectures are feasible for live performance.
The perceptual synchrony threshold is approximately 100ms. Beyond this, the response is perceived as a consequence rather than a simultaneous reaction.
10. The concept of "kinesthetic listening" describes:
Correct. Kinesthetic listening is the documented phenomenological state of heightened somatic attention that performers enter when they know their movement is consequential in the environment.
Kinesthetic listening describes the heightened somatic awareness performers experience when they know their movement triggers environmental responses — a distinct state documented by performance researchers.
11. The NYCB Algorithm Project (2018) involved which choreographer working with AI-generated sequences?
Correct. Justin Peck worked with AI-generated choreographic sequences in NYCB's partnership with Google machine learning researchers.
The NYCB Algorithm Project involved Justin Peck selecting and staging AI-generated movement sequences in collaboration with Google researchers.
12. Under current US copyright law, which of these scenarios is most likely to result in a protectable choreographic work?
Correct. The human's creative choices — selecting, arranging, and staging — constitute sufficient authorship under current legal interpretation, even when the raw material is AI-generated.
Copyright attaches to human creative choices. A choreographer's selection and arrangement of AI-generated material likely qualifies; unmodified AI output, credited to the AI, does not.
13. What does "fine-tuning" mean in the context of preparing AI movement models for dance-specific applications?
Correct. Fine-tuning involves continued training of a pre-trained model on a specialized dataset — in this case, dance-specific movement data — to produce outputs that reflect that domain's vocabulary.
Fine-tuning is additional training on domain-specific data. A general movement model fine-tuned on ballet or contemporary dance data will produce outputs that better reflect those specific vocabularies.
14. Which of the following is an example of "attribution transparency" as an emerging norm in AI-assisted choreography?
Correct. Attribution transparency as a professional norm means choreographers openly disclose AI contributions in programs, marketing, and funding applications.
Attribution transparency is the professional practice of choreographers disclosing AI tool contributions in program notes, promotional materials, and grant applications.
15. What historical analogy do AI choreography advocates most commonly use when arguing that these tools will expand rather than simply replace human creative labor?
Correct. The DAW analogy is the dominant frame: music production software democratized production, created new professional categories, and changed the field without simply replacing human musicians.
The most common analogy is to DAW (digital audio workstation) software — which transformed music production, enabled new work, and created new roles rather than simply replacing existing labor.