Directed Improvisation with AI
Term
Directed Improvisation with AI
Idea level
Application: Human–AI co-creation
Definition
Directed Improvisation with AI, introduced by Yuen Yuen Ang in 2026, extends her 2016 model of Directed Improvisation to human–AI co-creation by redefining human agency—from humans producing answers on their own to directing how humans and AI produce them together.
Treating AI as a complex, co-creative system rather than a simple machine, Ang defines direction as a meta form of human agency: the ability to influence outputs without being able to predict or control them.
More precisely, Ang advances paradigm-level direction: reorienting AI’s underlying assumptions rather than merely adjusting prompts within its default paradigm.
Sources
First articulation
Ang, Y. Y. “Directed Improvisation with AI.” PhD course taught at Johns Hopkins University (Spring 2026). Syllabus posted on SSRN, with forthcoming course page.
Theoretical foundation
Ang, Y. Y. (2026). “Directed Improvisation.” Glossary on Ang’s official website.
Ang, Y. Y. (2016). How China Escaped the Poverty Trap, Chapter 2: Directed Improvisation
Genealogy
[Pillar] Mechanical thinking
→ [Concept] Complicated: Seeing AI as complicated machine
→ [Concept] Command: enter prompt and get results (e.g., write my essay)
→ [Concept] Control (e.g., ban or police AI use; return to “medieval practices”)
Contrast with
[Pillar] Adaptive Political Economy (APE)
→ [Concept] Complex ≠ Complicated: AI as complex, not complicated
→ [Concept] Influence vs. Control
→ [Model] Directed Improvisation
→ [Application] Reform-era China: central government as director + local governments as improvising agents
→ [Application] Human–AI co-creation: human as director + AI as improvising agent
Quotes
Directed Improvisation (DI) is a model for enabling collective creativity that combines top-down direction with bottom-up improvisation, in contrast to rigid control or decentralized chaos.
— Ang (2026), Glossary for Directed Improvisation.
“In a world where AI will be ubiquitous, the task is not to eliminate AI, but to redefine human agency—from producing answers to directing how humans and AI produce them together. Rather than treating AI as a passive tool that executes commands, this course treats it as a co-creative system whose outputs depend on how humans guide it.
The meta-skill explored in the course is direction: the ability to shape how problem-solving or creativity unfolds by structuring its conditions, rather than producing answers directly.
— Ang (2026), Directed Improvisation with AI.
Concept Constellation
Across these sources, Directed Improvisation with AI co-appears with the following concepts and analytic themes:
[AI-specific extensions]
AI as complex adaptive system
Human-AI cocreation
Paradigm-level direction
Direction as meta form of human agency