Creativity in AI: Progresses and Challenges

2024-10-22

arXiv.org

AI Summary

Generative AI exhibits fundamental bottlenecks in creative problem-solving—weak abstract reasoning, poor compositional generalization, fragmented long-range logical coherence, stylistic drift, and frequent hallucinations—hindering simultaneous optimization of originality, diversity, and consistency, while raising copyright and authorship concerns. Method: We propose the first process-driven, multi-dimensional collaborative framework for evaluating AI creativity, explicitly formalizing the intrinsic tension among originality, diversity, and hallucination; integrating cognitive science principles to design a scalable 7-dimensional evaluation metric suite; and conducting cross-domain empirical analysis across language, visual art, and scientific reasoning using LLMs and diffusion models. Contribution/Results: Our evaluation reveals that state-of-the-art models achieve creative problem-solving success rates below 35%. The framework establishes a rigorous theoretical foundation and methodological toolkit for measurable, assessable, and augmentable AI creativity.

Technology Category

• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Philosophy and Ethics of AI: Artificial General Intelligence
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

Creativity is the ability to produce novel, useful, and surprising ideas, and has been widely studied as a crucial aspect of human cognition. Machine creativity on the other hand has been a long-standing challenge. With the rise of advanced generative AI, there has been renewed interest and debate regarding AI's creative capabilities. Therefore, it is imperative to revisit the state of creativity in AI and identify key progresses and remaining challenges. In this work, we survey leading works studying the creative capabilities of AI systems, focusing on creative problem-solving, linguistic, artistic, and scientific creativity. Our review suggests that while the latest AI models are largely capable of producing linguistically and artistically creative outputs such as poems, images, and musical pieces, they struggle with tasks that require creative problem-solving, abstract thinking and compositionality and their generations suffer from a lack of diversity, originality, long-range incoherence and hallucinations. We also discuss key questions concerning copyright and authorship issues with generative models. Furthermore, we highlight the need for a comprehensive evaluation of creativity that is process-driven and considers several dimensions of creativity. Finally, we propose future research directions to improve the creativity of AI outputs, drawing inspiration from cognitive science and psychology.

Problem

• Assessing AI's creative capabilities in problem-solving and art
• Identifying limitations in AI's diversity and originality
• Addressing copyright and authorship issues in generative models

Innovation

• Surveying AI's creative problem-solving and artistic outputs
• Evaluating creativity via process-driven multi-dimensional metrics
• Improving AI creativity using cognitive science insights