Nov22
To succeed with AI adoption, you need to know which AI is good for what. Expecting magic will lead to disappointment, like expecting creativity from chatbots.
It depends on who you ask.
According to Wikipedia and the Cambridge Dictionary, creativity is the ability to “form novel and valuable ideas or works using one's imagination” and “the ability to produce or use original and unusual ideas”. According to AI (Google AI Overviews and ChatGPT), creativity also includes combining existing ideas and making unexpected connections between existing concepts.
Generative AI (GenAI), which some claim is creative, can of course create only something it has seen before. So, it’s not surprising that AI chatbots claim that creativity includes combining existing ideas. If you think about it, that’s very human. It’s almost cute. People always redefine topics or change goals to fit their narrative. Take Sam Altman for example. How many times has he already redefined Artificial General Intelligence (AGI)?
There are two studies from 2023 and 2024, which concluded that AI chatbots generally produced more creative responses than humans. Even if you ignore the small sample sizes, it’s crucial to understand how these studies measured creativity.
https://www.nature.com/articles/s41598-023-40858-3
https://www.nature.com/articles/s41598-024-53303-w
Both studies used standard tests to measure divergent thinking tasks. These include generating uncommon and creative uses for everyday objects, and entering ten words that are as different from each other as possible.
No wonder that AI chatbots scored better than the recruited 256 English-native humans in the first and 151 in the second study. AI chatbots are Large Language Models (LLMs) trained on the vast amount of text generated by humans. LLMs probably know all the 170,000 English words that are in current use according to the Oxford English Dictionary. And most LLMs know some of the word meanings. So, it shouldn’t be difficult for a machine to come up with words that are as different from each other as possible.
But is that creativity? Would GenAI be able to come up with inventions like those created by these people?
I doubt it. Based on what GenAI can do – create output leveraging existing human knowledge – GenAI cannot create novel and original ideas. It cannot be the next Leonardo da Vinci, Thomas Edison, Albert Einstein, Jimi Hendrix, or Steve Jobs. LLMs cannot create new technologies, new physics theories, new product categories, or new music styles.
However, GenAI is probably better than the average human when it comes to mash-ups. After all, LLMs have access to a larger data set to start with than the average human. They can mash up what humans have done before in unexpected combinations. And, they can create these combinations much faster than humans.
For example, LLMs should be able to create an even better version of the Metallica & Mariah Carey mash-up “For Whom The Bell Tolls” and “All I Want For Christmas Is You”. Which AI chatbot accepts this challenge?
https://www.youtube.com/watch?v=Y9w3GSkJBzU
By the way, the research methods mentioned above basically measured only this mash-up ability of humans vs. GenAI. They didn’t measure creating novel, previously unknown ideas.
There are many business needs requiring new ideas for products, processes, or business models. It’s sufficient in many of these cases to mash up prior content. This is where GenAI adds value by creating versions and mash-ups of ideas that humans have not thought about before. Many companies can benefit from new creations in product design, new marketing visuals, or business model reinventions.
Some people call this GenAI creativity “everyday creativity”. I called it “everyday use”. I would put a warning label on GenAI that reads: „Creative outputs intended for everyday use only. Breakthroughs in science, arts, or business not expected.“
Business leaders whose companies depend on incremental creativity improvements should consider GenAI for mashing up prior ideas. But don’t expect novel breakthroughs. With these reasonable expectations it will be easier to achieve AI adoption among your employees.
By Jack Lampka
Keywords: Change Management, Generative AI
The Grace Gap: What Boards Should Learn from Nvidia's Leadership Crisis
Can GenAI be creative?
The Supply Chain Doesn't Need More Alerts. It Needs More Time to Act.
Friday’s Leadership Reflection Quote – Saeculum Leadership: Leadership Establishes the Conditions for Knowledge
The Corix Partners Friday Reading List - November 20, 2026