It has been known to come up at the beginning of one of my teacher training workshops on AI. When I present the status quo of generative AI with its LLM-powered text-to-anything potential and quasi-omni infiltration of all worlds, a colleague will invariably react with concern and dismay - lamenting the fact that AI will kill creativity as we know it, along the lines of: "Students won't do anything any more. They won't have to come up with their own ideas when the AI thinks for them. The sky is falling!"
The reaction is understandable. Many have thought and written about this - the inevitable demise of creativity in the age of AI, when software can often “outperform” humans.
How AI Can Kill Creativity In Humans
Is Generative AI killing creativity?
AI can help − and hurt − student creativity
It's a logical next thought for educators, whose job it is to shape young minds and help young ones become independent thinkers. Having this thing at their fingertips which offers them shortcuts feels like a threat. Here's why I think the fear is ultimately unwarranted.
High time to reposition “creativity” in the educational world
The advent of generative AI gives us an opportunity to redefine and reclaim “creativity” for everyone. There has been a tendency in education to view it as a separate, almost magical entity, featured in certain subjects like art, design, music or language arts. Some people declare: “I’m not creative”, meaning “I can’t draw” or “I can’t play an instrument.” This is of course a very narrow definition and not at all accurate, as creativity is problem-solving at its core. It is thinking of a higher order and requires analytical reasoning, ideation and strategic planning. There are no non-creative school subjects. All domains, including Maths and Science, are disciplines where students need to thinking critically and creatively. If anything, I think we need to be helping students to understand the processes behind creativity and give them chances to develop their thinking skills in diverse scenarios.
Outsourcing some cognitive tasks while training others
We already have software that runs autonomously under human supervision (think of pilots flying aeroplanes), a fact that millions of people accept daily. Yet, while it is true that AI chatbots and “smart” engines have shown that they can reliably perform certain tasks, this does not mean that it is wise to rely on them 100% and delegate all activities to ChatGPT et al. The software plays a subordinate role and acts at our behest and on our behalf. Anyone who writes and uses technology to get things done needs to understand how to find a balance. The world is currently designing norms on this for educational contexts. The consensus with many institutions seems to be allowing the use of AI for theses and essays, provided proper citation has been given. If anything, it is evident that we need to double down on training students on how to write. I say double down because some tradition-minded instructors have fostered generic writing (à la “teaching to the task”) essentially training students to write as if they were robots. As with Math skills that are trained before the use of a calculator, students need to become literate writers, then learn how to augment their skills with digital aids, once they have mastered the technicalities and developed their own style. In my view, AI is not destined to kill off human thinking processes, since individuals will be using these tools in tandem with others to get things done.
Enabling and Empowering to create new things
Using AI chatbots to generate text and claiming sole authorship, just to avoid thinking for yourself, is an obvious ethical no-go. And no, it is not the AI that is creative. The AI generates, it does not create. The human prompts and the tool suggests a solution. What potential does AI have for student thinking and creative competency?
I’ve noticed that students getting excited when using AI to generate images according to their imagination. They are tickled at the thought that their images are innovations. And they are thrilled when they can produce images on their own. When given the chance, they gain a deeper understanding of image generation, of the effects of stylistic prompts. Because the output / feedback is almost instantaneous, they can move quickly, iterate and perfect their work. Students who may have had self-doubt, or who may have hesitated to put pen to paper, can experience aha-moments that are part of the learning process. And those who may have been bolder and more keen to try new things already, are empowered to go further and pursue their personalised paths. When guided in their learning process, students can experience firsthand what it feels like to find solutions, then work on them. This applies to all situations in the classroom. The application of generative AI can provide impetus and propel students. Once academic honesty and responsibility is part of the training, it has the potential to help students think creatively.
At the end …
… of a teacher workshop, the same educators who were sceptical about AI and creativity in the classroom, become enthusiastic about the possible use cases. I routinely have them create images using AI tools to get a feel for it themselves. They realize that they are the ones nudging the AI along to generate their unique products.
Ultimately, it's not about the AI replacing human creativity but enriching it, opening doors to new forms of expression and understanding that were previously unimaginable, for both students and teachers alike.
Thoughts?
P.S. I started collecting examples of creative postings on Instagram. People just blow me away with their ideas. Check some examples out here: https://www.instagram.com/aliciaginalee/saved/creative/18015378676924401/
P.S.S. I was curious about ChatGPT’s response to the question of the impact of generative AI on student creativity. The thread can be found here:
https://chat.openai.com/share/ee53ef2a-ba5f-4654-8297-154fa72ae38d
Title image:
Prompt: Generate an image of a child writing creatively with a quill in her hand which emanating rainbow coloured flourishes on a blank canvas - pixar style.
Tool: DALLE3.
I’ve had a few post drafts in my mind for a while, but I just haven’t taken a moment to put thoughts into the machine. It’s been end of term, so there are loads of administrative tasks to get done before we send students off to spend a week without school.
Great post, could not agree more with every ounce of this. I am also generally an optimist on AI staying in "augmented" or copilot mode for at least the next 5 years, and create so many situations where human + AI brings creative and more effective (thinking more in the workplace) results.