Abstract:This study investigates the creative capacities of artificial intelligence (AI), exemplified by ChatGPT-4, in comparison with human creativity, utilizing the Figural Interpretation Quest (FIQ) as the evaluative tool. The primary objective is to assess whether AI can surpass human performance in creative tasks, particularly in terms of flexibility and subjectively perceived creativity. Participants, including both ChatGPT-4 and humans, were asked to provide creative interpretations for a set of ambiguous, abstract figures. The study measured the flexibility of these interpretations and gathered subjective ratings of creativity from independent evaluators. Results indicated that while AI, on average, demonstrated higher flexibility in generating creative interpretations, human participants excelled in terms of subjectively perceived creativity. Furthermore, the most creative human responses were higher than those of AI in both flexibility and subjective creativity. These findings suggest that in these types of complex multimodal interpretation tasks, AI shows excellent potential and the ability to generate semantically diverse ideas, yet human subjectively perceived creativity remains for now unmatched. The study highlights the limitations of AI in replicating the complexity of human creativity and points to possible theoretical interpretations of the reported findings.

What problem does this paper attempt to address?

### The Problem the Paper Attempts to Solve This paper aims to explore whether artificial intelligence (AI) can outperform humans in creative tasks, particularly in the creative interpretation of multimodal visual stimuli. Specifically, the researchers used the Figural Interpretation Quest (FIQ) as an evaluation tool to compare the performance of ChatGPT-4 and humans in creative tasks. The main objectives include: 1. **Assessing AI's flexibility in creative tasks**: The researchers evaluated the flexibility of AI by measuring the diversity of creative interpretations generated by AI. 2. **Evaluating subjectively perceived creativity**: The researchers invited independent evaluators to subjectively rate the creativity of interpretations generated by AI and humans. 3. **Comparing AI and human performance**: The researchers aimed to understand whether AI can outperform humans in these complex multimodal interpretation tasks, particularly in terms of flexibility and subjectively perceived creativity. ### Research Background In recent years, the emergence of generative artificial intelligence (such as OpenAI's ChatGPT, Meta's Llama, and Google's Bard) has sparked widespread discussion about its societal impact and prompted a reevaluation of the differences in creativity between humans and AI. Although AI has demonstrated capabilities close to or exceeding human abilities in certain tasks, there is still controversy over whether AI can truly surpass humans in complex creative tasks. ### Research Methods 1. **Participants**: The study subjects included 256 native English speakers recruited through the Prolific platform and ChatGPT-4. 2. **Tasks**: Participants were asked to provide two as different as possible creative interpretations for four ambiguous abstract figures. 3. **Evaluation Metrics**: - **Flexibility**: Measured by calculating the semantic distance between the two interpretations for each image. - **Subjectively perceived creativity**: Rated by independent evaluators on a three-level scale (0: basic/non-creative; 1: creative; 2: exceptionally creative). ### Main Findings - **AI performed better in terms of flexibility**: AI outperformed humans in generating diverse creative interpretations. - **Humans performed better in subjectively perceived creativity**: Human-generated interpretations scored higher in subjective creativity ratings than AI. - **The best human performers outperformed AI**: The top human performers exceeded AI in both flexibility and subjective creativity. ### Conclusion This study reveals the potential of AI in multimodal creative tasks, particularly in generating diverse interpretations. However, humans still maintain an advantage in subjectively perceived creativity. These findings suggest that while AI excels in certain aspects, human creativity remains irreplaceable in complex multimodal interpretation tasks.

Artificial Creativity? Evaluating AI Against Human Performance in Creative Interpretation of Visual Stimuli

Best humans still outperform artificial intelligence in a creative divergent thinking task

Artificial Intelligence Creates Art? An Experimental Investigation of Value and Creativity Perceptions

Can AI Enhance its Creativity to Beat Humans ?

The current state of artificial intelligence generative language models is more creative than humans on divergent thinking tasks

Humans as Creativity Gatekeepers: Are We Biased Against AI Creativity?

Creativity in AI: Progresses and Challenges

Creativity in the era of artificial intelligence

A Comparison Study of Human and Machine-Generated Creativity

Generative artificial intelligence, human creativity, and art

AI and the creative realm: A short review of current and future applications

Boundary crossing: an experimental study of individual perceptions toward AIGC

Can AI Be as Creative as Humans?

Putting GPT-3's Creativity to the (Alternative Uses) Test

Artificial Intelligence-Augmented Brainstorming: How Humans and Ai Beat Humans Alone

Artificial intelligence in fine arts: A systematic review of empirical research

AI and the social construction of creativity

Goetterfunke: Creativity in Machinae Sapiens. About the Qualitative Shift in Generative AI with a Focus on Text-To-Image

People's Perceptions (and Bias) Toward Creative Content Generated by Ai (ChatGPT-4), Human Experts, and Human-AI Collaboration

Using Linkography to Quantitatively Analyze the Design Ideation of AI Agents