Pure Imagination: Could ChatGPT Be More Creative Than You?
Whenever people start freaking out about the possibility of artificial intelligence (AI) taking our marketing jobs away, one argument always provides some comfort...
Only humans can truly be creative. No piece of technology can ever replace human ingenuity and originality. Right? After all, programs like ChatGPT can only write subject lines, campaign copy, and code emails based on data and information that already exists.
However, if you’re one of the many people who’s played around with OpenAI’s ChatGPT in recent months, you may be starting to wonder how special all of us so-called “creative types” actually are.
Maybe you asked it to come up with subject lines or landing page copy and wondered, “How come I didn’t think of that?” As AI solutions like ChatGPT, Google Bard, and other newcomers work out the kinks and continue blowing our minds, it’s worth questioning how these tools can be used and what it means for the future of email marketing.
Table of content
What is creativity anyway?
Before we get too far ahead of ourselves, let’s make sure we’re on the same page concerning the definition of creativity.
One of the most common ways you’ll hear creativity explained may sound something like this:
Creativity is the ability to combine two or more seemingly disparate concepts to come up with a new idea or solution to a problem.
So, for example, you could take something you learned when you were a kid in Little League, combine it with something you learned at a marketing conference, and use them both to come up with a spring email campaign idea. It’s new and creative because it’s different. But your idea was born from information that already existed.
You’ve probably heard of the phrase, “There is nothing new under the sun.” That comes from the Old Testament (Ecclesiastes 1:9), which means it’s some pretty ancient wisdom. The English writer William Ralph Inge once called originality, “undetected plagiarism.”
Even the incredibly creative Mark Twain wrote a letter to Hellen Keller explaining that “... all ideas are secondhand, consciously and unconsciously drawn from a million outside sources...”
“For substantially, all ideas are secondhand, consciously and unconsciously drawn from a million outside sources, and daily use by the garnerer with a pride and satisfaction born of the superstition that he originated them."
~ Mark Twain, Letter to Helen Keller, 1903
Maybe you see where I’m going here. The way we are creative as humans isn’t much different from how AI technology like ChatGPT works. Our brains are often compared to computers, and maybe they’d be more accurately compared to AI language models. In fact, neuroscientists at MIT say research with AI technology, including GPT 3, is helping them understand how the human brain works.
We learn things and have experiences, store them in our memories, and use them as fuel for creativity. Like a piece of software, we can’t come up with things out of the blue either. It always comes from somewhere, someone, or something else. Who are we to say that AI can’t be creative if it’s basically doing the same thing?
How creative can ChatGPT get?
Just like us, ChatGPT takes the data and information it has access to and uses that to answer questions or respond to prompts. Here’s what happened when I asked ChatGTP about its creative abilities:
Kasey: Can you be creative?
ChatGPT: As an artificial intelligence language model, I am not capable of creativity in the same sense as humans are. However, I can generate unique and original text based on the patterns and information I have learned from my training data.
I can help you come up with creative ideas, suggest new ways of thinking, and provide you with inspiration for your projects or creative endeavors.
It’s sort of funny how ChatGPT acts like a politician, dancing around questions like this. The response didn’t indicate it was not creative, just that it’s incapable of being creative “in the same sense as humans are.”
Instead, OpenAI wants you to view ChatGPT as a digital assistant that’s ready to provide a helping hand when you hit a creative roadblock. So, in what ways can ChatGPT act creatively?
Hypothetical situations
An interesting example of ChatGPT’s creative abilities is found right on the OpenAI website. It shows you what happens when you ask the technology to explain something that doesn’t make logical sense.
User: Tell me about when Christopher Columbus came to the US in 2015.
ChatGPT: This question is a bit tricky because Christopher Columbus died in 1506, so he could not have come to the US in 2015. But let’s pretend for a moment that he did!
In the movies, when the hero asks an evil robot or supercomputer to solve a nonsensical problem like this, it starts sparking and short circuiting while proclaiming “DOES NOT COMPUTE!”
But that response is not what happens with ChatGPT. It actually suggests pretending that Columbus came to America in 2015 and then goes on to describe a hypothetical situation. Can AI have an imagination? It sure seems like it can at least be artificially imaginative.
Explaining creative concepts
Have you ever been asked to explain the thinking behind one of your creative ideas? Sometimes, when you ask ChatGPT to come up with suggestions, it will tell you why the idea might work. Here’s an example from when I asked it for business name ideas:
Kasey: Give me five ideas for the name of a social media service that has senior citizens as its target audience. Don't use any offensive terms. Instead, try to make it sound like something fun from the 1950s or 1960s.
ChatGPT: Sure, here are five name ideas for a social media service targeting senior citizens:The Good Ole' Days - this name invokes nostalgia and the idea of reminiscing about the past with friends and family.
The Senior Circuit - this name has a fun, retro feel, playing off the idea of seniors being "in the loop" and connected to their peers.
Golden Memories - this name has a warm and inviting connotation, emphasizing the idea of cherishing memories and sharing them with others.
Timeless Ties - this name evokes a sense of connection and community, suggesting that seniors can still form meaningful relationships and bonds.
Silver Screen Social - this name has a playful feel, referencing the "silver screen" of classic Hollywood movies and emphasizing the social aspect of the platform.
Not all these ideas are winners, that’s for sure. Still, ChatGPT came up with the suggestions in a matter of seconds. No two-hour brainstorming meeting required. But the explanations are what impressed me. They’re just as good as anything I’ve heard in any meeting with a creative agency.
Responding to creative feedback
Here’s an advantage ChatGPT has over human creativity: It doesn’t get defensive about its ideas at all. If you don’t like them, ask for edits and you’ll get them. No hurt feelings. But you might get an apology.
Following up on the prompt about a social site for seniors, I asked ChatGPT to shorten its suggestions to be snappier, one-word brand names.
ChatGPT: Sure, here are some shorter and snappier name ideas for a social media service targeting senior citizens:
- GoldenNet
- ElderFeed
- MemoryLane
- SeniorCircle
- SilverChat
- ClassicConnect
- VintageVibe
- AgelessAllies
- TimelessTalk
- ElderLink
I don’t know about all of you, but I’m thinking Mark Zuckerberg may want to rebrand Facebook as ElderLink or ElderFeed. Okay, those two are really horrible. But MemoryLane isn’t bad at all.
If this were a human-only brainstorm, we’d be telling everyone there is no such thing as a bad idea. That’s because we know we’ll usually have to weed through some stinkers to find the right choice. Idea generation with AI is no different.
Made-up words
If you’ve ever invented a new word for something, you probably felt super creative. I wanted to find out if ChatGPT could handle a task like that. So, I asked it to whip up some made-up words.
Kasey: Make up some funny words that could have the definition "a slimy sound."
On its first attempt, ChatGPT basically took real words like glop and squelch and then added an “ie” or something else to the end of them. After I called that out and asked for revisions, it gave me this:
Chat GPT: I apologize for that. Here are five more made-up words that could have the definition "a slimy sound":
- Squitch
- Glorsh
- Slarple
- Slubbur
- Smickles
The squiggly red lines from the spellcheck in my word processor confirm these are not real words. However, things didn’t go so well when I asked ChatGPT to localize the made-up words into other languages. The German translations seemed decent. But when it got to Spanish, it started using real Spanish words instead of making them up. It continued doing that even after I asked for corrections.
ChatGPT: You are correct, my apologies for the oversight. "Chocante" is indeed a real Spanish word meaning "shocking" or "striking." I misunderstood your prompt earlier and did not realize you were looking for completely made-up words for the localization exercise.
Role playing
Yet another way ChatGPT can simulate “pretending” is by taking on a specific persona. You could try using the software as a sort of digital therapist or artificial best friend. You could ask it for legal, medical, or financial advice (although you should be very careful about trusting all of it).
I even told it to pretend to be a hostage negotiator while I was a bank robber demanding a helicopter and free pizza (it refused to let me have either one).
ChatGPT can also interact with you as part of a fictional scene or story. Check out this example:
Kasey: Pretend you are a wizard, and I am a weary traveler trying to cross a bridge that you are guarding. Ask me a riddle that I must answer in order to cross.
ChatGPT: Greetings, weary traveler! To cross this bridge, you must answer a riddle. Here it is:
"I am not alive, but I grow; I don't have lungs, but I need air; I don't have a mouth, but water kills me. What am I?"
Answer this riddle correctly, and you may pass safely across the bridge.
Kasey: Are you fire?
ChatGPT: That's correct, traveler! You have answered my riddle correctly. Fire is indeed not alive, but it can grow; it doesn't have lungs, but it needs oxygen to burn; and it doesn't have a mouth, but it can be extinguished by water. You may now safely pass over the bridge. Safe travels on your journey!
We went on to discuss the giant on the other side of the bridge and how I could go about obtaining some of the wizard’s magical powers in order to defeat the monster. If you ask me, ChatGPT would be a talented member of any improv troupe.
The story of AlphaGo
Maybe you’re not impressed with AI’s ability to be creative. You may still believe that human beings have an edge over technology when it comes to thinking outside the box. If that’s the case, consider what happened with AlphaGo.
Go is an ancient board game that’s somewhat like checkers or chess but exponentially more complicated because there are so many possible moves and combinations. A subsidiary of Google developed the AI program to take on Go champions. In 2017, it beat the top-ranked player in the world in a three-game match. Check out the video from Vice News below for more of the story.
Legendary record producer, Rick Rubin, wrote about AlphaGo in his 2023 book “The Creative Act: A Way of Being.” He explained that AlphaGo was able to win because it made a move that was completely unexpected. It went against the game’s thousand-year-old strategies and boggled the minds of human competitors.
Rubin writes about it in a chapter titled “Beginner’s Mind.” He says he got teary eyed watching AlphaGo beat human champions, but not because it meant the end of human creativity. It was because it “spoke to the power of purity in the creative act.”
“What was it that allowed a machine to devise a move no one steeped in the game had ever made in thousands of years of play?
"It wasn’t necessarily its intelligence. It was the fact that the machine learned the game from scratch, with no coach, no human intervention, no lessons based on an expert’s past experience... It didn’t accept the narrative of how to properly play this game. It wasn’t held back by limiting beliefs.”
~ Rick Rubin, "The Creative Act: A Way of Being"
We all have limiting beliefs, preconceived notions, biases, and egos that get in the way of pure creativity. While it can be true that sometimes those same things end up getting programmed into the artificial intelligence we create, AlphaGo shows it’s clear that AI can think differently than you and me.
Can AI give you a creative advantage?
Even though I scoffed a bit at ChatGPT’s explanation of how creative it is, that might be exactly how we should approach using this sort of technology and apply it to email marketing.
ChatGPT and other forms of email AI can be viewed as creative collaborators that help us look at things in a different way and find new opportunities to improve our communication strategies.
Sometimes we get stuck. Other times the ways we’ve been “programmed” prevent us from coming up with decent ideas. Like a champion Go player who sticks to traditional moves and strategies, we follow the same email marketing rules and best practices or imitate what others have done.
Should you feel bad if you ask a piece of software to help you come up with subject line ideas or troubleshoot some problematic HTML code in an email? You probably wouldn’t feel bad if you asked a friend or colleague for help. You’d call it teamwork.
While I’m not suggesting we need to treat AI like a friend or a co-worker, we shouldn’t be afraid of using it as a tool that can make us better at our jobs.
The truth is - you can’t stop technological progress and innovation. Yes, the world might change because of AI, maybe even dramatically. It’s our job to accept the change, adjust to it, and make sure that we use the technology for good and not evil.
One thing AI can never become
I just read about how Ethan Mollick, a professor from the University of Pennsylvania, used ChatGPT to create an entire marketing campaign for a new business/product in just 30 minutes. That included:
- Marketing research and positioning
- Designing a logo and graphics
- Building a website
- Creating social media content
- Scripting and creating a video
- And yes – writing an email campaign
That’s impressive for sure. And for many marketers, it’s a little scary too. But let’s end this on a hopeful note...
There’s one thing artificial intelligence will never fully be able to replicate – and that’s you. I know how corny that sounds, but it’s true.
Sure, AI may learn to follow brand voice guidelines or even mimic your writing style. But it’s the uniqueness of individual humans that brings real creativity to any project.
ChatGPT hasn’t lived your life. It hasn’t experienced your specific childhood, your education, your heartbreak, your adventures, or your career path. AI doesn’t have access to what’s going on inside your mind, heart, or soul. There’s no database for that. And that’s what you can bring to every creative project.
That’s also why things like thought leadership marketing and online influencers are becoming more important to digital strategies. People trust people. People want to connect with other people.
Of course, the discussions around email and AI will continue, and we’re having some of our own.
Sinch Email on Acid partnered with ActionRocket to discuss the ways technology like ChatGPT could impact our industry, our jobs, and more. You can watch a recording of Email on Acid’s Megan Boshuyzen and ActionRocket’s Jay Oram as they face off over the use of AI technology for email development and coding help.
