Opinion: Photography, AI and the Question of What Makes Art

Photography has never been particularly good at standing still. From film to digital, darkrooms to Photoshop and manual focus to cameras that can now identify and track the eye of a bird in flight, photography has always evolved alongside technology. Each advancement has brought some version of the same debate. Is this making photography better? Is it making it easier? And, inevitably, is it cheating?

Artificial intelligence is simply the newest technology to force photographers to ask those questions. Except this time, the questions feel different.

AI is no longer limited to helping us take better photographs. It can create the image itself. That distinction has opened an interesting and sometimes uncomfortable conversation about what photography is, what role the photographer plays in creating an image and where AI-generated work belongs within the world of fine art.

I don’t believe the answer is as simple as AI is good or AI is bad. AI has tremendous potential within photography. It also has the potential to remove some of the very things that make photography meaningful.

Perhaps the more important question is not whether photographers should embrace AI, but where we draw the line between using technology to support our vision and asking technology to create that vision for us.

 

AI as a Tool for Photographers

 

There is a lot to like about AI when it operates behind the photograph rather than in place of it. Modern photographers already use AI constantly. Subject recognition, autofocus, noise reduction and masking. Image organization and editing tools all rely on some form of machine learning or artificial intelligence. Used well, these tools remove technical barriers between the photographer and the finished image.

I think the overall debate is, can it be a good thing?

A wildlife photographer can concentrate on composition and behavior while the camera tracks an animal's eye. A portrait photographer can spend less time creating tedious masks and more time refining an image. AI-assisted culling can reduce hours spent sorting nearly identical frames. Noise reduction can make photographs possible in conditions that would have produced unusable images not very long ago. None of those things diminish the photographer's creative involvement.

The photographer still chose to be there. The photographer still recognized the subject, selected the lens, composed the frame, understood the light and decided when to press the shutter. AI simply helped execute the photographer's intention. In that sense, AI is not particularly different from many of the technological advances photographers once resisted and eventually accepted. The complication begins when AI moves from assisting with a photograph to manufacturing what appears to be one.

 

When the Photograph Never Happened

 

Generative AI can create extraordinary images. A few words can produce dramatic landscapes, elaborate portraits, impossible wildlife encounters and scenes with lighting photographers might spend years learning to create. That capability is impressive. It is also fundamentally different from photography.

Photography has traditionally had a physical connection to the world. Light reflected from something that existed, passed through a lens and was recorded on film or a sensor. Whatever happened afterward in the darkroom or on a computer, there was an originating moment.

Something was there.

Someone saw it.

That relationship between the image and the moment is one of photography's defining characteristics. With generative AI, there does not have to be a moment. The mountain may never have existed. The light never fell across the valley. The person was never standing in front of a camera. The animals never walked through the snow. The resulting image may still be beautiful, compelling and artistically meaningful, but its relationship with reality is fundamentally different.

Does that necessarily make it less valuable?

 

And Then There Is Fine Art

 

This is where the conversation becomes considerably more complicated.

Fine art has never required literal representation of reality. A painter does not have to prove that a landscape existed. A sculptor does not have to recreate something found in nature. Digital artists have spent decades constructing worlds that never existed outside their imagination. Fine art is often about concept, interpretation, emotion and intention rather than documentation.

So why should AI-generated art be automatically excluded?

If an artist uses AI deliberately as part of a larger creative process, there is a reasonable argument that the resulting work can belong within contemporary fine art. The tool itself does not determine whether something has artistic merit, but AI introduces another question that fine art has always cared deeply about: authorship.

Who created the work?

With traditional photography, the answer is relatively easy. The photographer made a series of creative decisions about where to stand, what to include, what to exclude and when to press the shutter. With generative AI, authorship becomes less clear.

The person provides direction and the AI interprets it based on a model trained to find existing visual material. The artist evaluates the result, changes the prompt, generates again and continues refining until the image reflects the desired idea. However, is that process equivalent to painting the canvas, sculpting the clay or taking the photograph?

I don't think we have completely answered that question yet.

 

The Value of Difficulty

 

There is another part of this conversation that deserves more attention, the experience required to make the image. Technology has always made photography easier. That isn't inherently bad. Nobody receives extra artistic credit because their camera was harder to operate.

A landscape photographer learns how weather can affect light. A wildlife photographer learns animal behavior and patience. A portrait photographer learns how to make another human being comfortable enough to reveal something authentic in front of the camera.

Photographers learn through failure. We miss focus. We chose the wrong shutter speed. We arrived too late. We stood in the wrong place. We wait for light that never comes. We take thousands of mediocre photographs while slowly learning why a handful of them work. That process shapes the photographer.

AI can dramatically shorten the distance between an idea and an impressive image. That accessibility is one of its greatest strengths. It may also be one of its greatest weaknesses. When we remove too much of the process, we have to ask whether we are simply making creativity more accessible or whether we are removing some of the experiences that develop creative judgment in the first place.

 

Perfection Has Never Been the Point

 

AI is remarkably good at giving us what we ask for. Perfect light, perfect skin, perfect mountains, perfect wildlife and yet, photography rarely behaves that way. Real light changes. People can blink. Animals move in unpredictable ways and can altogether leave. Weather refuses to cooperate and locations can become rapidly crowded. The photograph you imagined may not be the photograph you ultimately make. Sometimes the limitations produce something better.

There is an unpredictability to traditional photography that forces the photographer to respond rather than simply direct. We work with the world instead of having complete control over it. That tension between intention and reality is part of what makes photography interesting. As AI-generated imagery becomes increasingly flawless, imperfection may become more valuable.

A photograph's significance may increasingly come not from how extraordinary it looks, but from the knowledge that the moment occurred.

 

Maybe Authenticity Becomes the Luxury

 

For years, digital photography has moved toward technical perfection. Cameras have become increasingly more capable. Editing software can correct problems that once ruined photographs and AI now promises to take that perfection even further. But when anyone can generate something spectacular in seconds, spectacular may eventually stop being enough.

Authenticity could become the differentiator. Did this person exist? Did this landscape look like this? Did this animal stand there? For authenticity the photographer should say I was there, I saw it, I photographed it, nothing changed about this image. There is something powerful about those statements, particularly within fine-art photography.

A beautiful AI-generated image can show us an imagined world. A photograph can show us how another person saw the real one. Neither necessarily needs to replace the other.

 

So Where Does AI Belong?

 

I think AI belongs in photography. I also think boundaries matter.

Using AI to reduce noise, improve focus, organize images or simplify editing allows technology to support the photographer's vision. Generative tools can also have legitimate roles in conceptual art, digital art and mixed-media work.

Where we should be more careful is in pretending those things are interchangeable. An AI-generated image does not need to be called a photograph to have value.  I don’t think giving the image the title of “AI Art” diminishes the value. A digital composite does not become less creative because we acknowledge that it is a composite. Fine art is broad enough to accommodate all these mediums without erasing the distinctions between them.

And perhaps that is where this conversation ultimately lands. AI does not have to destroy photography to transform it. It may force photographers to reconsider what makes photography valuable in the first place. The answer probably isn't megapixels, perfect focus or flawless editing. AI will eventually become indistinguishable from what has been taken with a digital camera versus what was created with a prompt.

The value of the image may come from a photographer who went somewhere, they experienced something, they noticed something worth remembering and for one fraction of a second, they decided to press the shutter. Technology can help us make that photograph better, but the moment itself still belongs to the photographer.