Can AIs really be creative?
For centuries, creativity has been one of the defining pillars of humanity. From cave paintings to artificial intelligence writing poems, creating has been one of our most profound forms of expression. However, in recent years, machines have begun to venture into realms we once considered exclusively ours. What happens when an AI composes music, designs an illustration, or writes a story? Can we call it creativity? Or are they just imitating patterns? In this article, we explore this fascinating question from different angles.
Creativity: an evolving concept
Before answering whether an AI can be creative, it’s worth asking: what do we mean by creativity?
Creativity isn’t a single thing. It can be the ability to generate something novel, solve a problem in an unconventional way, or combine known elements to produce something original. Some definitions emphasize the intention behind the creation, others focus on the outcome. In many cases, it’s associated with inspiration, emotion, even the irrational.
This range of meanings complicates the answer. If creativity is simply generating something new and valuable, does an AI that writes a symphony that moves millions fall into that category? If intention or awareness of the creative process are essential, then perhaps not.
How do they create artificial intelligences?
Artificial intelligences that generate creative content work through algorithms that learn from vast amounts of data. For example, an AI trained to write poetry has analyzed thousands of poems and learned patterns of style, rhyme, structure, and vocabulary. It can then combine these elements in new ways to produce a text that resembles what it read, but is not an exact copy.
The same is true of AIs that create images, such as MidJourney or DALL·E. From natural language descriptions, they generate artificial illustrations or photographs. In music, programs like AIVA or Amper Music compose complete melodies in a matter of seconds.
The process may seem magical, but there is a clear mechanism behind it: learning from previous data, identifying common structures, and generating new combinations that respect those rules.
Imitation or authentic creation?
One of the biggest debates revolves around whether what an AI produces is a mere imitation or truly an original creation. Can an AI invent something?
The reality is that, although the process is based on existing patterns, it’s not very different from what humans do. A painter learns from other painters, a writer is inspired by books they’ve read, a musician draws influence from hundreds of melodies. No one creates from nothing. Even our most original ideas are rooted in what we’ve seen, heard, or experienced.
So, if an AI can combine known elements in a novel way and generate a result that didn’t exist before, wouldn’t that be a form of creativity?
Creativity without conscience
One of the key points is that AIs are unaware of what they are creating. They don’t know if something is beautiful, nor do they feel pride in their work. They have no emotions or life experiences that filter into their art. Their creations are not born from an internal impulse, but from external instruction.
This leads many to think that, although the result seems creative, the process is not. In other words, there may be originality in the product, but there is no creative intent on the part of the person creating it.
But again, we might ask: is the process or the result more important? If a song created by AI moves someone, is it less valid because it didn’t stem from a real emotion?
Surprising cases of artificial creativity
There are real-life examples that challenge traditional ideas about creativity:
In 2018, a painting created by an AI trained by the art collective Obvious sold for over $400,000 at a Christie’s auction. The portrait was not copied from any previous work. It was a combination of styles learned by the algorithm, which generated a fictional face that never existed.
In music, AIs like AIVA have composed symphonic pieces that have been performed in concerts. Although experts can notice certain differences with human composers, many listeners cannot distinguish whether it was written by a machine or a person.
In film, there are short films with scripts written by AI, even automatically edited trailers. Some of these works, although imperfect, demonstrate an uncanny ability to replicate narrative structures.
Collaborative creativity: humans + machines
Beyond the “pure” creativity of AI, an increasingly popular approach is augmented creativity. Rather than replacing humans, AIs act as assistants or co-creators.
A graphic designer can use AI to generate hundreds of variations of a logo in seconds. A writer can ask an AI to suggest titles for a novel. A musician can experiment with algorithm-suggested melodies to step out of their comfort zone.
This collaboration allows for faster exploration of ideas, overcoming creative blocks, and finding new forms of expression. In that sense, AI enhances human creativity rather than competing with it.
Current limits and ethical challenges
Despite its power, AI creativity has significant limits. Artificial creations may lack symbolic depth, irony, subtle humor, or genuine emotion. They struggle to understand the cultural or social context of a work. An AI can make a joke, but it hardly understands why it’s funny.
Furthermore, there are ethical debates surrounding copyright. If an AI was trained on millions of copyrighted works, is its creation truly original? Should the original artist receive any kind of recognition or compensation?
And on the other hand, if an AI-generated work goes viral or becomes a commercial success, who is the real author? The user who wrote the prompt? The AI developer? The company that created it?
The future of artificial creativity
The future points to an increasingly integrated coexistence between humans and machines in the creative field. We will likely see more tools designed for artists, designers, musicians, and writers that integrate AI naturally into their processes.
It is also possible that new creative genres, new forms of hybrid art, and even new professions will emerge: curators of AI-generated art, prompt editors, algorithmic composers.
What is clear is that AI is no longer just a technical tool, but also a source of aesthetic and cultural exploration. Creativity is ceasing to be a human monopoly, at least in its execution.
Conclusion: Redefining creativity
So, can AIs really be creative? The answer depends on how we define creativity. If we understand creativity as the ability to produce something new, interesting, and valuable, then AI can do that. If creativity involves emotions, lived experiences, or conscious intentions, then no.
But perhaps the question shouldn’t be whether AI is creative, but rather how it can enrich our own creativity. Artificial tools challenge us to think of new ways to create, collaborate, and imagine.
Accepting that creativity is not an exclusively human capacity doesn’t mean losing something, but rather expanding it. Perhaps the true creativity of the future will be neither human nor artificial, but an encounter between the two worlds.