The sorrow of humanity? Wharton MBA can't match ChatGPT in terms of business creativity.

Wallstreetcn
2023.09.12 03:38
portai
I'm PortAI, I can summarize articles.

In the compilation of the top 40 brilliant ideas by Professor from the Wharton School, 35 of them are from GPT, while only 5 are from human students.

To find opportunities for new businesses, provide solutions for unmet needs, or name a new company, these are all unstructured tasks that seem unsuitable for algorithms.

But the reality is quite the opposite. Recently, several professors from the Wharton School organized a competition to determine who is better at coming up with innovative ideas: humans or AI.

As it turns out, AI has overwhelmingly triumphed in the realm of creativity that humans take pride in.

The competition's challenge was to attract college students with a new product or service priced at $50 or less.

In academia, the measurement of innovation value typically revolves around three dimensions: quantity of ideas, average quality, and the number of genius ideas.

In terms of quantity of ideas, GPT has completely outperformed humans. It takes humans several days to generate 200 ideas through brainstorming, while ChatGPT only requires an hour. Foldable laundry baskets, dormitory kitchen utensils, ergonomic seat cushions for classrooms... hundreds and thousands of creative ideas miraculously emerge from a single laptop.

The professors collected a total of 400 ideas from both students and GPT, and then conducted online tests to ask the target customers—college students—how likely they would be to purchase these products or services.

These feedback responses can be translated into purchase probabilities. The average purchase rate for human ideas is 40%, while GPT-4's average purchase rate is 47%. In short, ChatGPT is not only faster, but also better on average in generating ideas.

However, it is well known that in the realm of innovation, one outstanding idea surpasses 100 decent ideas.

The professors curated the top 10% of the 400 ideas, and the results showed that out of the 40 selected ideas, 35 came from GPT, while only 5 came from human students.

In conclusion, the professors at the Wharton School stated that it is no longer necessary to consider competition between humans and machines, but rather to find a way for both to collaborate.

By having humans play the role of the navigator and artificial intelligence serve as the co-pilot, this strategy has already emerged in the field of software development. They predict that such human-machine collaboration will provide better products and services to the market and offer superior solutions for any future societal needs.