Authors: Lennart Meincke, Ethan R. Mollick, Christian Terwiesch

Published on: January 27, 2024

Impact Score: 8.07

Arxiv code: Arxiv:2402.01727

Summary

• What is new: This paper introduces an advanced approach to enhancing the diversity of AI-generated ideas through tailored prompt engineering, specifically using GPT-4 for new product development for college students.
• Why this is important: Previous AI brainstorming tools failed to generate a sufficiently diverse set of ideas, limiting novelty and overall idea quality.
• What the research proposes: Utilizing different prompting methods with GPT-4, including Chain-of-Thought (CoT) prompting, to increase idea dispersion and novelty.
• Results: CoT prompting with GPT-4 significantly increased the diversity of ideas, closely matching the diversity seen in human subject groups, and produced the highest number of unique ideas.

Technical Details

Technological frameworks used: GPT-4

Models used: Chain-of-Thought (CoT) prompting

Data used: New product development ideas for college students priced under $50

Potential Impact

Creative industries, marketing departments, product development firms, and educational technology companies might either be disrupted or stand to benefit from these insights.

Want to implement this idea in a business?

We have generated a startup concept here: IdeaGenAI.