Evaluating crowdsourced creative design ideas: How close is artificial intelligence (ChatGPT) to human experts judgments? - Archive ouverte HAL
Communication Dans Un Congrès Année : 2024

Evaluating crowdsourced creative design ideas: How close is artificial intelligence (ChatGPT) to human experts judgments?

Adnan Mendares
Marie-Laure Gavard-Perret

Résumé

The adoption of recent technology to offer superior value for all stakeholders in organizations has become a new reality. In the context of innovation management, artificial intelligence can be useful for the complex and expensive creative idea evaluation process. In this research, we showed through two studies that, a artificial intelligence-based judgments method (using the application (ChatGPT) was able to evaluate the originality of crowdsourced creative design ideas in a similar way to a human-based judgments method (using human experts) with moderate reliability. However, in terms of evaluating the attractiveness, the results of both approaches are not similar.. These very early studies integrating two different judgmental evaluation of creative idea indicate the potential of using IA in the idea assessment process.
Fichier non déposé

Dates et versions

hal-04696662 , version 1 (13-09-2024)

Licence

Identifiants

  • HAL Id : hal-04696662 , version 1

Citer

Adnan Mendares, Marie-Laure Gavard-Perret, Eline Jongmans. Evaluating crowdsourced creative design ideas: How close is artificial intelligence (ChatGPT) to human experts judgments?. AFM, Université paris Dauphine - PSL, Jun 2024, Paris, France. ⟨hal-04696662⟩

Collections

UGA CERAG MEMO-SHS
33 Consultations
0 Téléchargements

Partager

More