Can AI Create Real Art? What Artists, Scientists, and Tech Experts Say

Artificial Intelligence (AI) is everywhere, from chatbots to driverless cars – and now it even paints and composes music. But can AI really create “art” in the way we humans do? This question has sparked a lively debate among artists, scientists and tech gurus. On one hand, AI can spit out stunning images, poems or songs in seconds; on the other hand, many argue that true art requires human emotion, originality and intent. In this article I’ll explore what we mean by “art” and “AI”, and then dive into the diverse opinions of artists and experts. Spoiler: I’ve wrestled with this myself – part of me marvels at AI’s creativity, and part of me fears it misses something essential. Together we’ll look at the evidence, the science, and the soul of the matter.

What Is Art, Anyway?

Before judging AI’s artwork, let’s ask: What even is art? Philosophers and cultures have wrestled with this for centuries. Some think art is mimesis (imitation of reality), as Plato taught – a painting was valued by how accurately it copied its subject. In the Romantic era, others said art is expression of emotion – a painting or song should stir our feelings. Immanuel Kant argued that art should be judged by its form and creativity, not just beauty. And some modern theories (like Dickie’s “institutional theory”) even say context matters: an object becomes art when the “art world” (galleries, critics, museums) agrees to call it so.

In plain language, art is usually thought of as a human creative act. It’s how we express ideas, stories or emotions – from cave paintings to rock music. Art can make us feel joy, sadness, wonder or even challenge our assumptions. As composer Pablo Picasso said, “Art washes away from the soul the dust of everyday life.” (He was Picasso, so we’ll trust he knew a bit about art.) What’s clear is that art is deeply tied to experience: it comes out of living, feeling and imagining. It’s partly about the finished piece, but even more about the person who made it.

Personally, I’ve always felt that art is a conversation between creator and audience. When I sketch or write, I tap into memories, dreams or frustrations – something uniquely me. Even comedy or cartoons carry the imprint of a human joke. This idea – that art is an expression of human experience – is at the heart of the debate on AI art. Can a machine, with no feelings or memories, capture that spark? As we’ll see, some experts say yes, while many artists strongly disagree.

What Is AI – and How Does AI “Make” Art?

AI stands for Artificial Intelligence, which basically means software that mimics tasks we thought only humans could do (like understanding language, recognizing faces, or learning patterns). In practice, modern AI often means machine learning: algorithms that learn from huge amounts of data. For instance, we train an AI by feeding it millions of images or songs so it can spot patterns and make new creations.

“AI art” usually refers to content created or enhanced by AI tools. For example, a popular approach is text-to-image generation. You, the user, give a program a text prompt (say, “a surreal forest with glowing trees”), and behind the scenes the AI uses neural networks to generate a matching image. These systems (like DALL·E, Midjourney or Stable Diffusion) have been trained on vast collections of existing art and photos.  They learn which colors go together, how light and shadow work, what a tree looks like, etc. Technically, many use Generative Adversarial Networks (GANs) or diffusion models to craft images . In plain terms: one neural network tries to make images, another tries to judge them as “realistic,” and through many rounds they improve.

Other AI art tools do things like style transfer (repainting your photo in the style of Van Gogh or Picasso) or even write music and poems. Essentially, the AI digests the “essence” of art it was fed, then recombines patterns to create something new. It can do this very quickly: a job that might take a human artist hours or days can often be done in seconds by AI. In fact, as the TechTarget definition notes, AI art tools have made it “possible to change or enhance existing images or generate new images based on submitted text prompts,” even for people without any artistic skill.

Because AI art is still new, it raises lots of questions. Technologically, AI art challenges the age-old idea that humans are the only creators of art. Ethically and legally, it raises issues about copyright and credit (since the AI was trained on other artists’ work). But it also empowers new creativity: now “almost anyone can create digital art in a fraction of the time”. You don’t need to know how to draw – you just write or say what you want, and the AI paints it. It’s like having a very obedient, very fast collaborator.

To sum up the tech side: AI art is when machines use learned patterns to generate images, music, text or other media that people might call “art.” Behind the scenes it’s a lot of math and data crunching, but the output can look amazingly creative. The big question is whether that output is truly art in the human sense – which brings us to what real people (and experts) think about it.

How Artists View AI Art

Unsurprisingly, many working artists have strong opinions – most of them skeptical or negative. A common theme is that creative process matters to art, and simply hitting a button lacks soul. Children’s book illustrator Rob Biddulph put it bluntly: AI art “is the exact opposite of what I believe art to be… true art is about the creative process much more than the final piece. And simply pressing a button to generate an image is not a creative process”. In other words, for Biddulph real art comes from an artist’s feelings and experiences. He feels an image created by AI (no matter how pretty) is just someone else’s inner world summarized by an algorithm, not born from the artist’s heart.

Copyright and authenticity are big concerns. Many illustrators worry AI “stole” their style. As artist Anoosha Syed warns: “AI doesn’t look at art and create its own. It samples everyone’s then mashes it into something else.” To her, AI art isn’t genuinely original. Another illustrator, Harry Woodgate, accused AI tools of relying on “the pirated intellectual property of countless working artists”. (There is legal debate on this, but artists fear losing credit for their creativity.) Woodgate and others even started the #NoToAIArt campaign, arguing that using scraped artwork is unethical and devalues the human craft.

Many artists also point out that current AI images have a recognizable “look” – often uncanny or generic. Syed notes that AI pictures have an inauthentic feel and, over time, people might “turn away from [them] because of its inauthenticity and ‘cheapness’”. Indeed, if you stare at half a dozen AI faces or landscapes, you start to notice quirks (like oddly shaped hands or repeating patterns). Some illustrators believe we will grow tired of the novelty and return to appreciating hand-made work. In fact, AI’s very unpredictability can be a double-edged sword. Independent animator Ruth Lingford observes that AI is “acting like a sort of collective unconscious”: it randomly merges bits of many images. She finds some of these unexpected blends “very interesting,” but she also suspects fully AI-made films would fall flat. Without human guidance, it “closely approximates some aspects of the creative process” but lacks intentional meaning.

Music teachers have similar doubts. Jazz musician Yosvany Terry explains that on stage, art is a dialogue and an emotional performance – something an algorithm can’t replicate. He says “that sense of interplay, or the ability to react in the moment, is something that artificial intelligence can’t reproduce”. In his view, AI can technically compose music, but it “lacks surprise, emotion, and even silence”. It might create a perfectly polished tune, but without the soul and spontaneity of humans. Terry compares it to technology like radio: initially feared, but eventually just another tool. Even so, he thinks musicians should “open our eyes, ears, and arms” to work with AI’s strengths, while remembering that true artistry comes from human creativity.

Some artists are more optimistic, treating AI as a new tool. Mixed-media artist Matt Saunders sees AI as a challenge to break old habits: “every new technology upends conventions… We should be grateful to be challenged and knocked out of our habits and assumptions!”. He notes that many artists are already experimenting with AI (after all, artists have long used computers and cameras to make art). To Saunders, art is ultimately what we ascribe to it – it’s about the conversation, not the code that made it. If an artist uses AI in a thoughtful way, the result can be meaningful. Architect Moshe Safdie echoes a balanced view: he says AI can produce beautiful images or functional designs (like optimizing light in a building), but he draws the line at deep creativity. For Safdie, art has a “spiritual, emotional” element (seeing Guernica might make you weep for human suffering, he explains). He frankly states: “In terms of art created by AI, I don’t think we can call it art… AI can imitate something that’s already been created and regurgitate it... but that is not an original work.”

In short, most practicing artists say AI might make cool stuff, but it’s either not genuine art or it’s at best a tool for the human artist. The common thread is that art involves human experience and intent – something machines simply lack.

What Scientists and Researchers Say

Turning to science, studies of how people react to AI art reveal interesting patterns. Cognitive researchers have found that even when images look similar, viewers tend to prefer art believed to be made by humans. In one study, participants were shown AI-generated paintings and told half were by humans and half by AI (randomly). Across various criteria (liking, beauty, meaning), the “human” label made people rate the same image more positively. In other words, the context – who we think made the art – heavily influences our judgments. Study authors summarize: people tend to be “negatively biased against AI-created artworks relative to... human-created artwork”. Another experiment using eye-tracking found a subtle bias: viewers unconsciously spent more time looking at paintings they believed were human-made, even though the art itself was often indistinguishable. The researchers conclude that “human and AI art may be perceived as having similar aesthetic values, [but] an implicit negative bias toward AI art exists”. In plain terms: our brains still give a slight edge to the human touch, even if the image looks equally good.

Why? Scientists suggest it’s partly because art is seen as a communication of human experience. We value the idea that an artist was behind a piece – that it has a story, emotion, or hard work inside. One paper noted that knowing a work is human-made leads viewers to attribute more meaning and effort to it. It’s like paying more for a handcrafted item than a factory-made one – even if they look alike, the “soul” of human labor adds value in our minds.

On the AI side, some experts in AI research have weighed in too. AI “godfather” Yoshua Bengio (a Turing Award winner and pioneer in deep learning) argues that there’s nothing inherently sacred about human-only creativity – it’s “entirely plausible that we could one day create [AI] systems that surpass human abilities in various fields, including the arts”. In other words, from a technical standpoint, machines might eventually be better than us at making beautiful things. But Bengio points out a social question: even if AI can out-paint Picasso or out-compose Beethoven, will we value that artwork the same way? He notes that society may attach different value to an AI’s work versus a human’s creative process.

Similarly, AI pioneer Yann LeCun (Meta’s Chief AI Scientist) acknowledges AI can now generate art and music on par with humans in a purely technical sense. However, he believes there’s an extra layer to art that AI doesn’t have: authenticity and truthfulness. In a Wired interview, LeCun remarked, “If I played you a recording that topped Charlie Parker at his best, and then told you an AI generated it, would you feel cheated?”. He answered that yes, we probably would feel cheated, because music (and art) is not just about pattern; it’s about culture and context. “Truthfulness is an essential part of the artistic experience,” he said. LeCun compares a Master’s painting to a cheap AI copy: even if the copy looks alike, we’d still prefer the original hand-painted piece, just as we’d pay more for a handmade bowl than a factory clone. His point is that art carries human stories and imperfections that machines can’t replicate.

Another tech leader, Sam Altman (CEO of OpenAI, makers of DALL·E), has a more upbeat view. After an AI tool created images in the style of Studio Ghibli anime, he defended the technology by saying it lowers the barriers to creativity for everyone. Altman argues that making art easier and more accessible is a “net win” for society: “Giving everyone more tools, making things easier, lowering the barriers to entry… increases the number of people that can contribute to society,” and we all benefit. In his view, even if AI art isn’t “classic art,” it democratizes expression – millions more can now visualize their ideas.

We also hear voices from thinkers who straddle art and science. Composer Carol Reiley (co-founder of an AI-music startup) notes that AI is entering what was once thought of as a uniquely human creative space. But she also emphasizes the need for artists to guide the process. Grammy-winning violinist Hilary Hahn (Reiley’s collaborator) enthusiastically says “there’s room for AI music to coexist with human composers… to gracefully merge tech with humanity”. They see AI not as a replacement, but as a collaborator or amplifier of human creativity.

In summary, researchers and AI experts recognize that current AI can produce high-quality art and music – in fact, sometimes indistinguishable from human-made content – but they also note something missing. Studies suggest people inherently favor human-made art and attribute more meaning to it. AI won’t put on a concert or attend a gallery opening; it doesn’t have intentions or consciousness. As LeCun put it, no matter how smart machines get, “AI is not the same experience”.

The Ongoing Debate

So where does this leave us? The verdict isn’t unanimous. On one side of the debate, many traditional artists insist that real art requires human agency and emotion. They see AI as a sophisticated copier at best, and a thief at worst (if it used copyrighted images). To them, pressing a button and waiting for a picture feels lazy and inauthentic. On the other side, some researchers and tech leaders see AI as a powerful new tool that expands creative horizons. They argue the final piece – how it moves you – can be appreciated regardless of the process, and that humans guiding AI still imbue it with meaning.

Personally, I’ve been both amazed and uneasy. I remember playing with Midjourney one evening: I typed “a cat astronaut exploring Mars” and in seconds had a glossy, uncanny image that looked straight off a graphic novel. It was exhilarating – the computer did in a moment what I could never draw. But I also felt a twinge of guilty wonder: did I really create that art? It raised questions about skill and effort. Then again, I consider myself more of a writer than a painter, and maybe that nighttime AI image was just a new form of creative expression for me.

The truth might be a bit of both worlds. As artist Matt Saunders noted, “art means what we ascribe to it… it can be a provocation, but it is essentially always part of a conversation.”
If an AI-generated image sparks joy or insight, perhaps it is art for the beholder. But as architect Moshe Safdie reminds us, art also has a soul beyond pixels. We still treasure seeing a master’s brush strokes or hearing a live jazz improvisation. In the end, the question “Is AI art real art?” may come down to how we define “real.” If real art means something new and meaningful, maybe AI can sometimes get there. If it means a living human creating with intent, then AI will always feel one step removed.

One thing is clear: the debate has energized both camps. Many artists are experimenting with AI, and many tech developers are listening to artists’ concerns. The result may not be a winner-takes-all “AI vs human” scenario. Instead, we might see more collaboration. Just as computers didn’t end drawing and writing, AI might become another tool – like a camera or a synthesizer – that artists use creatively.

At the very least, AI art forces us to think about what we value in creativity. As we navigate this “brave new art world,” we should remember that technology has always changed art. The printing press, the camera, even the synthesizer were once controversial, and yet art flourished. Whether AI is a passing trend or a permanent partner, it has already pushed humans to reconsider and reaffirm what makes art meaningful to us.

In the words of jazz musician Yosvany Terry, new tech upends the status quo, but “Any new technology is first seen as a threat… we must remember that all these innovations are man-made, and as humans we can create and innovate.” Perhaps that’s a hopeful note: even in an age of AI, the human capacity to imagine, to feel, and to give meaning remains at the heart of art.