The activity begins with three tensions.
First, Molière Ex Machina[1] (https://www.ac-paris.fr/moliere-et-ia-des-eleves-de-l-academie-de-paris-a-l-opera-royal-du-chateau-de-versailles-134041) raises the question of artistic creation through heritage, style, a historical period and counterfactual fiction: “What play might Molière have written had he not died in 1673?” The project brings together theatre, research, classical heritage and generative AI.
Next, the scientific case presented by Cornell (https://news.cornell.edu/stories/2025/12/ai-gives-scientists-boost-cost-too-many-mediocre-papers) highlights how LLMs can increase scientific output while making it harder to distinguish between a substantive contribution and a well-written text with little scientific value.
Finally, the ESSEC article (https://knowledge.essec.edu/en/innovation/age-ai-do-we-still-need-researchers.html) offers a complementary view: AI can detect robust patterns in large datasets, but meaning-making, explanation, hypothesis formulation and judgement remain deeply human.
Activity objective
Collectively build a database of concepts, definitions, examples and tensions around four questions:
- What is an artistic, engineering or scientific output?
- What is creativity in the arts, engineering and sciences?
- What feeds this creativity: data, lived experience, knowledge, emotions, cultures, imaginaries, methods?
- How does AI support, transform, shift or replace particular creative acts?
The expected output is a structured database with the following fields: concept, domain, definition, example, human role, AI role, data used, risks, opportunities, keywords.
General format
The activity draws on physical cards. The cards prompt small-group discussion.
Each group receives a pack containing four types of cards: concept cards, situation cards, tension cards and production cards.
Schedule 30 minutes
5 minutes — Introduction and framing
Key question:
In the age of AI, is creativity a human capacity, a property of works, a collective process, a combination of data or an interaction between humans, machines and the lived world?
Introducing the three starting points:
- Molière Ex Machina: AI as a partner in creation situated within a heritage, language, style, historical period and theatrical tradition.
- Cornell: AI as an accelerator of scientific output, with the risk of a proliferation of texts of average quality or difficult to evaluate.
- ESSEC: AI as a tool for detecting patterns, while humans remain central to asking questions, interpreting, explaining and making meaning.
Instructions: each group will produce a short collective definition and complete a form.
5 minutes — Forming groups and distributing cards
Form groups of 4 to 6 people. Each group receives:
- 3 concept cards;
- 2 situation cards;
- 2 tension cards;
- 1 production card.
Each group chooses a spokesperson.
10 minutes — Guided discussion: defining creativity
Each group answers three questions using the cards:
In your case, what are you producing?
A work, a hypothesis, a paper, an interpretation, an experiment, a method, a performance, a theory, a database, a narrative?
Where does creativity lie?
In the initial idea, formulation, style, method, interpretation, lived experience, recombination, deviation, surprise, judgement, gesture, emotion or making connections?
What does AI change?
It accelerates, suggests, imitates, recombines, explores, formalises, assists, standardises, homogenises, shifts authority, increases output or blurs the boundary between human contribution and automated generation.
10 minutes — Brief feedback
Each group has 1 -2 minutes to share:
- its definition of creativity;
- what AI transforms;
- what remains irreducibly human or situated.
[1] Arte video https://www.arte.tv/fr/videos/133043-000-A/moliere-ex-machina/