Spitballing with ChatGPT: An Experiment in Innovation

It’s 7pm, and this evening’s innovation class will be different. 

Students take their seats around tables formed in a U-shape in the small, neutrally-painted room. A set of students also appear on a wall screen via Zoom. 

Tuning in from a laptop occupying a space on one of the tables, and introduced as a surprise to the students, is a new member of the team: ChatGPT, bearing the nickname, Chad. Chad G. P. Tee, to be formal. 

I’m teaching Innovation and Creativity in Organizations this term to a group of creative, curious students. I’ve been immersed in innovation for years but never imagined I would be bouncing ideas shoulder to shoulder with an AI bot. 

Judging from the opening interaction, Chad is eager to begin.  

“Will you join us in a group brainstorming?” We ask. 

“Sure,” Chad responds. “I’d be happy to join in a group brainstorming session! What’s the topic or problem you’d like to brainstorm about?” 

Thus begins our class session. 

The topic for this evening’s brainstorming, selected by students: How might we obtain better information and knowledge about what is in our food? 

Tonight’s goal is twofold: First, we want to see how ChatGPT performs in the specific activity of brainstorming. 

Second lies a deeper goal. 

We are wondering what role generative AI tools like ChatGPT might play in organizational innovation. We know there is innovation in AI. Will there be AI in innovation? Will AI have a seat at the table in the process of innovation? 

Students begin sharing ideas aloud in the room, in popcorn fashion, which we enter onto sticky notes in an online program, IdeaFlip. This allows us to capture both in- person and zoom-attendee ideas in the same place. 

We turn to Chad for ideas, and it begins to spew out a rapid stream of possibilities: 

We hit “stop generating,” and clarify: “Our team has ground rules, only one idea at a time.” 

“Apologies for that,” Chad replies. “Here’s the first idea: Develop a food labeling system that provides clear and concise information about the ingredients, nutritional content, and potential allergens in the food. This could include a color-coding system to make it easy to quickly identify certain ingredients or nutritional content.” 

After more student ideas are shared with the room, we ask for a second idea from Chad. 

“Develop a food database that contains information on the nutritional content, ingredients, and origin of various foods,” says Chad. “This database could be accessible to the public and updated regularly to ensure accuracy.” 

We continue capturing ideas from students and from Chad, suggestions that span more comprehensive information about food sources, benefits, and allergens to a database of food and health information and best practices, to a SmartPhone app, to some sort of portable device. 

One idea seems to garner special interest: a wearable device that can tailor food information to our individualized health status and portfolio, taking into account our momentary as well as long-term wellness needs. 

As part of brainstorming, we are trying to generate ideas that are “out there,” and so we up the ante: “Ok, how about one more idea, and make it wild.” 

“All right, here’s a wild idea,” says Chad. “Develop a food scanning device that can analyze the molecular composition of any food item and provide a detailed breakdown of its nutritional content, potential allergens, and even potential contaminants.” 

Students smile. 

A few final ideas from students and Chad, and the session comes to a close. 

We thank Chad for joining us. 

“You’re welcome!” our gracious participant replies. “I’m glad I could contribute some ideas to your brainstorming session. Let me know if you need any more help or ideas.” 

Now it is time for classroom reflection. 

Students report feeling the tool was both “human” and “dehumanizing.” Students describe Chad as “helpful,” “groundbreaking” and “like a smart Alexa,” but also “scary.” 

“Let’s pause for a second and give this something to think about,” says one student. “A piece of artificial intelligence [just] joined the group. That was terrifying, shocking, and exhilarating all in one go. To put it more succinctly, the future is already here and will encourage further expansion of innovation.” 

Another student echoes the human, un-human mix: “It was a little scary to imagine you can treat ChatGPT like speaking with a human.” 

Several students talk about the enlivening nature of having ChatGPT in the process. 

“It was great to see how [the] ideation stage came alive during our live class when we conducted several brainstorming exercises,” said a student.  

Later, a student shares this input: “A real eye-opener for most of us was our guest speaker Chad G. P. Tee. I say eye-opener because when it was prompted, we were all mind-blown.” 

Tonight’s experiment was just a start; there are more parts of the process to try out. Next time, we will ask Chad to build on the ideas of others in real time. We can also involve AI in the sorting exercise—grouping similar ideas so the team can further develop them. It may be valuable to ask Chad to help judge the ideas for commercial viability, a post-brainstorming step in innovation. 

Still, this evening leaves us thinking there is sure to be a potent role for generative AI models in the innovation process. It can clearly add to the set of ideas in the room (another “voice” or “brain”), and it can get a little “out there” when needed. As students noted, it will be able to un-stick a group that is experiencing idea-block or seed an ideation process with starter ideas. 

“My take on this can best be summed up by Marie Curie’s famous line, ‘Nothing in life is to be feared, it is only to be understood. Now is the time to understand more, so that we may fear less,’” said one student.  “If we understand how to use and harness this power, it will likely become an even more valuable tool for mankind.” 

Indeed, we are likely to invite Chad again soon.  

About Larry Ebert 

Larry Ebert is a management consultant and senior instructor in Management and Organizational Leadership and Human Skills Development at Golden Gate University, teaching Leadership, Communication, and Innovation. He consults clients on innovation strategy and change management, including the application of emerging AI tools into business processes. Clients have ranged from Netscape/AOL and Hewlett Packard to Stanford University, Wells Fargo Bank, and internet startups. He served as co-CEO of InDefense, Inc. Larry is also a musician and composer.