Our AI writing assistant, WriteUp, can assist you in easily writing any text. Click here to experience its capabilities.

Automating creativity

View Original View Raw

Summary

AI is now able to beat humans on creativity tests, and three recent experimental papers have demonstrated that AI can generate creative ideas in real-world situations. The papers found that AI-generated ideas were more feasible and impactful than human-generated ideas, and AI-assisted humans created stories that were judged as significantly more novel and interesting. Humans still have a large role to play in innovation, but should include AI in the process due to its creative ability. The AI works best when given fewer, more specific prompts and constraints, and people should experiment to find what works best. AI acts as a powerful creative engine, and its use now gives many people access to good ideas that used to only be available to a few.

Q&As

How have AI's been able to pass flawed tests of creativity?
AI's have been able to pass flawed tests of creativity by being able to generate diverse and meaningful ideas.

What did the study by Wharton researchers show about the comparison between AI-generated ideas and student-generated ideas?
The study by Wharton researchers showed that ChatGPT-4 generated more, cheaper and better ideas than the students, and that the purchase intent from outside judges was higher for the AI-generated ideas.

What did the study by Boussioux et al. show about the comparison between AI-generated ideas and human-generated ideas?
The study by Boussioux et al. showed that the overall quality level of the AI and human-generated ideas were similar, but the AI was judged to be better on feasibility and impact, while the humans generated more novel ideas.

What conclusions can be drawn from the three studies discussed in the article?
The conclusions that can be drawn from the three studies discussed in the article are that AI can generate creative ideas in real-life, practical situations, it can help people generate better ideas, the ideas AI generates are better than what most people can come up with, but very creative people will beat the AI (at least for now), and there is more underlying similarity in the ideas that the current generation of AIs produce than among ideas generated by a large number of humans.

What advice is given in the article for how to use AI to help generate ideas?
The advice given in the article for how to use AI to help generate ideas is to not worry too much about the exact wording of the prompt, to experiment to see what might work best, to use few-shot techniques because they seem to help subjectively, to use constraints to generate less likely answers, and to use techniques that take advantage of the ways that AI can hallucinate plausible, but interesting, material.

AI Comments

👍 This article provides an insightful look into how AI can be used to generate innovative ideas. It provides practical tips on how to use AI for idea generation and encourages people to consider using AI to generate ideas even if they don't consider themselves highly creative.

👎 This article fails to provide any evidence that AI can actually come up with useful ideas in the real world. Additionally, it does not address the potential ethical concerns that could arise from using AI for idea generation.

AI Discussion

Me: It talks about how AI is increasingly able to pass tests of creativity, and how it can be used to generate ideas in practical settings. It also goes into detail about how using AI can be helpful in coming up with creative ideas.

Friend: Wow, that's really interesting. It sounds like AI could be a really powerful tool for generating ideas, even for those who don't consider themselves creative.

Me: Exactly. The article suggests that AI can act as a powerful creative engine, connecting seemingly unrelated concepts and generating ideas that humans would not have thought of. It also provides some tips on how to use AI to help generate ideas, such as using constraints and few-shot learning. So it looks like AI could be a great tool for coming up with new, creative ideas.

Action items

Experiment with different prompts when using AI to generate ideas. Try using simple prompts, few-shot techniques, and constraints to generate more original ideas.
Utilize AI as a creative engine by taking advantage of its ability to generate relationships between seemingly unrelated concepts.
Take advantage of the randomness of AI output to generate innovative concepts that beat those of most humans.

Technical terms

Generative AIs: Artificial Intelligence (AI) systems that are able to generate new information, such as ideas, from existing data.
Alternative Uses Test: A psychological test used to measure the ability of humans to come up with diverse and meaningful ideas.
Torrance Tests of Creative Thinking: A psychological test used to measure the creative thinking ability of humans.
ChatGPT-4: A type of AI system used to generate ideas.
Crowdsourcing: A process of obtaining services, ideas, or content by soliciting contributions from a large group of people, typically via the internet.
Circular Economy: An economic system that is restorative and regenerative by design, and which aims to keep products, components, and materials at their highest utility and value at all times.
Few-Shot Learning: A type of machine learning where the AI is provided with examples of the kind of results desired before it is asked to generate ideas.
Y Combinator: A famous accelerator that provides seed money, advice, and connections to startups.
LLMs: Language models, which are AI systems that are trained to generate relationships between tokens that may seem unrelated to humans but represent some deeper connections.