When I first heard ChatGPT was released, I felt, for a split second, the way a Stegosaurus might have as they watched an asteroid plummet through the atmosphere and craterize a large chunk of the Yucatan Peninsula.
That didn’t last for too long.
As stories about generative AI writing blogs, devising limericks, designing images, and doing dozens of other creative tasks made the rounds, it became easier to see how AI-assisted creativity might be used. The technology itself is neither what the utopists nor doomsayers expected.
Rather than a substitute for artists and creative people, artificial intelligence is a tool that can accelerate people’s ability to achieve their capabilities – if done right.
What does that look like in practice? Curious myself, I did some digging into how writers and innovators imagine how artificial intelligence could drive the creative process in fiction as well as how real-world creators are experimenting with artificial intelligence.
How Do Writers Imagine AI Can Help with Creativity?
“One of the biggest roles of science fiction is to prepare people to accept the future without pain and to encourage a flexibility of mind.”
I love this Arthur C. Clarke quote. It envisions fiction as a sandbox where we can imagine scenarios, identify risks down the road, and avoid making worst-case mistakes in the real world. Much of what science fiction has imagined about AI has been less than flattering.
Writers have been quick to warn us about nefarious AI (HAL 9000, Skynet, etc.), but not as many have conceived AI tools willing to collaborate with people, especially not from a creative standpoint. However, there are a few options that come to mind.
There’s Samantha from Her. She helps Theodore save memorable emails and proof the letters he ghostwrites for a living. Samantha shares what she loves while redlining errors and providing alternative sentences for the more impressionistic letters. Eventually, she even handpicks his best work to submit to a publisher. She’s often a great collaborator (romantic relationship aside), but she and the other AIs eventually grow hyperintelligent and leave humanity. Adios, brainstorming buddy.
Another option is Prometheus in Max Tegmark’s “The Tale of the Omega Team.” Early in the story, the eponymous Omega Team uses their AI to create an animation empire, outsourcing everything from writing and acting to drawing and digital rendering to Prometheus. More than just sidelining creative people, the Omegas throw ethics to the wind. They create shell companies and foster disinformation campaigns to hide that their content is 100% AI-created. And that’s before the Omegas undermine and punish global governments for the sake of the “greater good.” Yikes.
When it comes to using AI as a creative tool, my favorite example is Janet from The Good Place. In her words, she’s “simply an anthropomorphized vessel of knowledge built to make your life easier.” But she’s so much more. Janet can create an entire simulated world, collaborate on endless froyo puns, and propose reality-shifting solutions that help save humanity. Better yet, she brings out the best in everyone she meets.
How Are Real-World Artists Using AI-Assisted Creativity?
So, we have a wide range of possibilities (hopeful, cynical, and in between), but what’s actually happening in the real world? Right now, it’s all a bit more experimental.
Since we’re still in the Wild West phase of AI-powered creativity, artists and companies are trying to figure out where human creativity and machine output intersect to create engaging art and content. There definitely are some growing pangs left to go, but the output we’re seeing emphasizes the importance of human input during the planning stages.
Let’s look at two industries I know and love to see how AI is reshaping them.
Film
Let’s start with movies. Advocates of AI-assisted creativity in filmmaking think of it as an equalizer, giving micro-budget filmmakers and hobbyists the means to achieve their creative vision. Generative AI allowed “The Frost” to be set in Antarctica’s inhospitable cold and “The Mannequin” to cast deepfake Marilyn Monroe. Artificial intelligence clearly drives the creation of images, animations, cinematography, music, and script – much to the detriment of cohesion or watchability.
“/Imagine” plays with the technology in a more satisfying way. A narrator brainstorms an AI prompt aloud, revealing things about herself, her perceptions, and her insecurities over a montage of AI-generated images. There’s a rift between what’s said and what shows up on the screen, making a statement about the nuance lost when “intelligence is an artificial thing.”
It’s not just indie filmmakers or disruptors using the tool. In Hollywood, we’re seeing some examples of artificial intelligence during secondary creative stages. Late Night with the Devil included three retro-ish commercial-break title cards created by AI. The marketing team promoting Alex Garland’s Civil War used AI-generated images of war-torn cities during the promotional campaign leading up to its release.
Both instances have garnered intense backlash, especially because artificial intelligence was such a hot-button issue during the 2023 Writers Guild of America strike. In fact, the 148-day strike resulted in guardrails that prevented 100% artificial intelligence generated scripts and AI source material. Additionally, studios must obtain permission from actors and pay them for the time their digital doppelgängers are on screen. So, we might see fewer explicit uses of this tech in the near future.
Publishing
It’s no surprise that the world of writing has been hesitant to use artificial intelligence. As we’ve talked about previously, authors and publishers have filed lawsuits against OpenAI to challenge the organization’s claims of fair use to produce AI-generated content. However, even the world of writing is not without its own examples of AI-assisted creativity.
What happens when you prompt ChatGPT to write a book of poems in 24 hours? You get Aum Golly, a collection of 55 poems produced by artificial intelligence but guided by human co-author Jukka Aalho. The poems themselves are fairly basic, often repetitive, and feel like modern musings rather than anything poetic. However, art is in the eye of the beholder, and online reviews show some interest in the art form.
Or there’s the AI-written 1 the Road. Inspired by Jack Kerouac’s westward road trip, Ross Goodwin connected video, audio, and GPS sensors to create a stream-of-conscious travelogue. Reviews indicate that much of the book lacks plot or story arc with occasional moments of inspiration. It’s an interesting experiment, but not really destined for mass appeal.
Outside of the pathfinders exploring AI’s potential, the industry at large is adapting this technology for various steps throughout the publishing lifecycle. In “State of Play: Exploring Generative AI’s Transformative Effects on the Media and Entertainment Industry,” researchers found publishers were using AI for content generation, editing, summarization, translation, audiobook recording, localization, and sales chatbots. In some instances, this means optimizing what their creative team can do, but in others, it means removing people entirely.
Lessons for Using AI in Your Creative Projects
Though each creative profession is going to have its own challenges and opportunities with artificial intelligence, there are some cornerstone lessons to keep in mind.
Generative AI may have advanced faster than most imagined, but it still has some considerable limitations. In general, artificial intelligence is good at remixing art, but it currently lacks the intention that makes the work of human artists resonate. Moreover, some experts are suggesting that artificial intelligence will run out of training data in the near future, which could cause the technology’s capabilities to plateau.
As a result, creative people and companies are most likely going to find these use cases the most appealing if they want to stay fresh rather than put out stale content that doesn’t differentiate them from the noise:
- Brainstorming – Generative AI is great at exploring concepts. Some of the creative people I know use it to storyboard ideas, shape outlines, or even give them a sense of what’s the run-of-the-mill approach is (so they can avoid it). They even use it to fill gaps in their knowledge or experience as they put things together.
- Audience Research – How well do you know your target audience? Sometimes, it helps to clear out your assumptions and reassess your customers, users, or buyers from a fresh perspective. ChatGPT and other generative AI tools can help to identify customer pain points and differentiators, crafting a range of nuanced customer segments in less time. With sales and lead generation growing more difficult, this is an all-around win.
- Research and Summarization – Creative people are specialists in their craft but are expected to know a variety of niche subjects to speak authoritatively to an audience or present concepts in an authentic way. ChatGPT can help them quickly learn the fundamentals of new subject matter in a fraction of the time.
How to Hire AI-Savvy Creative Professionals
In most cases, innovation doesn’t rewind (barring a dark age), so we expect creative people will need to embrace artificial intelligence.
That doesn’t mean creative professionals need to be trained as AI engineers. However, they do need to dabble in some technical skillsets to maintain flexibility and leverage this tool to the fullest. As a result, your organization should start considering whether to include these requirements in your job descriptions:
- Prompt engineering – Artificial intelligence isn’t going to read anyone’s mind. Current versions of generative AI need clear and specific instructions that provide context, examples, constraints, and boundaries. Professionals who know how to iterate, enhancing the output in a fraction of the time, are worth taking the time to find.
- Data manipulation – The results of any Larger Language Model (LLM) will depend on the quality, diversity, and usability of your data for its intended purpose. Creative professionals who have some familiarity with data manipulation will be able to remove errors, integrate various data sources, and boost training data to improve the overall performance of their AI tools.
- Coding – Though coding isn’t required to use artificial intelligence, it can help if your organization wants someone who can integrate existing LLMs into your own applications or create specialized models.
As these skillsets reflect, creative professionals will likely need to pursue a STEAM education (STEM + the arts) to accelerate their creative process and keep their careers and your organization competitive. In the short term, that translates to a market-wide skills gap, but one that’s not insurmountable.
For starters, it helps to work with a creative tech staffing company that knows how to source, screen, and interview AI-savvy professionals. Since this caliber of creative professionals are not abundant, you’ll need a large talent network to boost your odds of finding candidates with AI know-how. Upskilling is another option. This allows you to guide seasoned creative professionals through your own programs or those of an upskilling partner to accelerate your time to value.
What You Can’t Ignore
If there’s one thing for hiring managers to understand, it’s that AI is here to stay. How it will be used depends on breakthroughs, regulatory restrictions, and people experimenting in real time.
That said, when creative people are using smart practices, it should be embraced.
If you need help finding people willing to explore new worlds of innovation while also showing them your organization is the right home for their skills, it’s time you reach out to the Dexian Tech Creatives Staffing team.