Commercial Photography in the Age of AI 2026
A flood of plausible images did not lower the price of photography. It split it in two. Where the value went, what the machine can and cannot do, and how a real shoot pays now.
A field report from inside the market, not a think piece watching it. We shoot product, lifestyle, brand, campaign, editorial, headshots, and real estate for a living.
This is our read on commercial and brand photography and what generative AI did to it. We shoot this work for a living, so it is a field report from inside the market, not a think piece watching it.
We wrote it for the person who has to decide where the photo budget goes in 2026. You have a vendor pitching AI product shots for a few dollars a frame, a board asking why you would pay for a real shoot when the machine is free, and a feed full of images that look fine and feel like nothing.
We are going to give you the numbers, the legal exposure, and our own hard opinions about where real still wins and where it does not. Every stat is a setup. Our read is the payoff.
AI collapsed the price of a plausible image to near zero. That did not devalue photography. It moved all the value into the one thing AI cannot make, a real picture people trust.
AI did not make photography cheaper. It cut the field in two.
In June 2025 a prediction market ran a thirty-second spot during the NBA Finals. One person made it in two days for about two thousand dollars, generating a few hundred clips and cutting to fifteen. 1 Two years earlier that spot is a six-figure production with a crew, a location, and a month of post. The cost of a moving image fell by roughly ninety-five percent in the time it takes most brands to approve a mood board. 2
Sit with what that did to the price of a picture. Not the quality. The price. When a competent image costs almost nothing to make, the image stops being the scarce thing. The scarce thing becomes the one people believe.
The floor fell out from under the generic image, the kind that could be anything or nothing. At the same time the real, verifiable photograph got repriced upward as a trust asset, a picture that proves something happened. That gap is the most important development in commercial imaging since the smartphone, and most brands are on the wrong side of it.
When supply goes vertical, price goes to the floor.
The clearest casualty is stock photography, the business that sold you a generic handshake or salad bowl. Getty reported $939.3 million in 2024 revenue with its stock library down close to five percent, 6 the visible edge of what analysts call a silent collapse of photo licensing revenue. 7
The same agencies now make real money selling images to the machines instead of to you, and two giants that spent twenty years competing decided the better move was to merge and control what counts as verified, licensed, real. 8 Our read is that the merger is a bet on scarcity.
Stock photography was the cheap seat of commercial imaging. AI took the cheap seat. The agencies noticed, stopped fighting each other, and merged to sell the expensive seat, the one labeled real.
We watched this at the low end of our own inbox. The pure white-background catalog cutout, the flat product with no story, is a commodity now, and a good tool plus a careful editor gets most of the way there. We say that out loud, because pretending otherwise is how you lose trust. The moment the brief has a person in it, or a real place, or a claim a customer will judge you on, the math flips.
The squeeze is entirely in the middle. The working shooter supplying competent, generic frames is the position going away. The two ends, premium-real and commodity-cheap, both survive. The question is which tier you are buying, and whether you know it.
As images got infinite, people got suspicious.
Getty's research across more than 30,000 adults in 25 countries found 98 percent of consumers say authentic images are pivotal to trust, and about 90 percent want to know whether an image was made with AI. 4 Read that as a pricing signal. In a world drowning in plausible images, the believable real one is the scarce good, and scarce goods command a premium.
The same research found something sharper. Consumers forgive AI on objects and landscapes, and turn suspicious the moment it depicts people or real products. 4 The machine gets a pass on a sunset. It does not get a pass on your face or your bottle. Which is exactly backwards from how most brands deploy AI, on the hero shot the customer scrutinizes hardest.
Customers forgive AI on skies and forgive nothing on skin and products. 4 The smartest place to spend real money is the exact frame most brands hand to the machine, the hero shot of the product and the person. You are no longer paying for a photograph. You are paying for evidence, and evidence is worth more the more fakes surround it.
There is a name forming around this: the authenticity premium, the measurable value people place on content they believe a human made and a camera captured. 16 A real photograph is becoming a status signal, the visual equivalent of a handmade object in a world of injection molding.
We are not romantics about this. AI belongs in a modern photo workflow, and we use it. It earns its keep before the shutter and after it. In pre-production it is a fast, cheap ideation engine, forty lighting directions and set concepts in an afternoon, a sketchpad that talks back. In post it saves real time on the grind, cleanup, masking, and the dozens of crops and color variants a campaign needs.
Where it faceplants is anywhere truth is the point. Real products, because the customer will hold the actual thing. Real people, because the uncanny tells read as a lie even when the viewer cannot say why. Real places, because a hallucinated room is a lawsuit with a shutter sound. And trust itself, because once a customer catches you faking one image, they discount every image you have published.
Use AI on the parts of a shoot no one is asked to believe, the ideas, the cleanup, the variations. Keep it off the parts that carry the claim, the product, the face, the room. That line is the entire discipline.
AI is cheapest on the first frame and gets expensive as the stakes rise, the opposite of how brands budget. The true cost of an AI image is the render times the scrutiny it will face. On the frame that carries your claim, that multiplier is enormous, and no vendor prices it into the two-dollar quote.
The AI question does not have one answer. It has a different answer for every kind of commercial photograph.
Splits clean.
The flat cutout on white is close to commoditized. The hero shot is another animal, condensation on real glass, the weight of a watch on a wrist. Customers scrutinize product images hardest, and are least forgiving of synthetic products specifically. 4
Where AI does the most quiet damage.
Convincing at a glance, wrong on inspection. The coffee shop is every coffee shop, the couple is no one. It photographs the idea of a moment without the moment, and audiences feel the absence even when they cannot name it.
The safest from AI, for a reason.
A campaign is a promise a company makes about itself. The prestige is that it is real, made by named people on real sets with real talent. The moment you learn a luxury campaign was generated, the prestige evaporates.
They break in the most personal way.
A headshot is a claim about a specific real person. The AI-flattered version gets caught the moment that founder walks into the room, and now the first thing a client learns is that the photo was a lie.
A category of legal exposure.
A property image represents a physical asset a buyer will walk through. Generate a room that does not exist and you have not made a nicer listing, you have made a misrepresentation.
The hero product shot, the campaign, and the real estate listing share one trait. Someone will check them against reality, in the store, in the room, in the mirror. That is exactly where synthetic breaks, and exactly where a real shoot is worth every dollar.
The most common mistake we see is using AI on the highest-scrutiny frame to save money and shooting the low-scrutiny filler for real. It is backwards. Spend real budget where the customer looks hardest and let the machine handle the cutouts nobody studies.
The licensing and copyright ground under AI images is not settled, and the risk does not sit with the tool. It sits with you, the brand that published.
The landmark case, Getty versus Stability AI, was messier than either side wanted. In the UK, Getty abandoned its primary copyright claims before closing, and the court held that a model's weights are not a stored copy of the training images. 10 11 The biggest test case to date left the question open, which for a brand means the risk is open too.
If you publish an AI image that reproduces someone's copyrighted work, a real person's likeness, or a competitor's trade dress, the exposure can land on the publisher. This is why the one thing enterprises pay for is indemnification. Adobe built its enterprise pitch on it, 12 but even the safest offers exclude beta features and any prompt that names a person or a brand. 13
Indemnification means the vendor pays your lawyers if an AI image gets you sued. Most free and consumer tools do not offer it. The brand that hits publish carries the risk.
A real photograph comes with a paper trail, model and property releases and a licensing agreement that says exactly what you own and for how long. That documentation is not bureaucracy. It is the thing that lets you sleep, and clean rights are quietly becoming one of the strongest arguments for a real shoot.
Not a technology problem. A trust problem.
The brands that got publicly burned did not fail because the tech was bad. They failed because they announced it.
If you use AI, use it quietly and use it where it does not touch a claim. Do not build a campaign around the fact that you used it. Nobody has ever bought more of a product because the ad was cheaper to produce.
A category error about what a photo is for.
A commercial photograph is a persuasion instrument, not a picture of an object. AI is good at the object and bad at making a stranger trust it, and trust runs on the sense that a real person was present.
Ignoring the hidden cost.
The two-dollar image is two dollars until you count the revisions to fix the mangled logo, the frames you cannot legally use, and the equity that leaks when people decide you fake things.
The brands losing at AI photography are not losing on technology. They are losing on trust, rights, and hidden cost, three bills that come due after the cheap image ships.
Everyone can price a shoot. Fewer people can price what it returns.
Start with the asset disappearing everywhere else, trust. When 98 percent of consumers tie trust to authentic imagery, 4 a verifiably real image buys you the one thing your synthetic competitors cannot. Then durability. A real shoot produces an owned library with clean rights you can recut for years. Real images are an appreciating asset. Synthetic images are a recurring liability.
Do not benchmark a real shoot against the cost of an AI frame. That comparison is rigged and it is the wrong question. Benchmark it against outcomes you already track. Conversion on a real hero image versus a synthetic one. Return rate when the product photo matched the product versus when it flattered it. Measure image spend as trust per dollar, not cost per frame, and you tend to arrive at a real shoot on your own.
Real work compounds into a relationship. Synthetic output resets to zero every time you regenerate it. A real shoot hands you a rights-clean library you can cut for years; an AI campaign hands you outputs you have to remake and re-clear on every use.
One is an asset on the books. The other is a subscription to risk.
Before the shoot, AI is a sketchpad.
Lighting directions, set concepts, wardrobe moods, storyboard frames. It makes the pitch faster and sharper. None of that generated material ends up in the final work. It is scaffolding, and we take the scaffolding down.
On the day, everything that carries the claim is real.
Real product, real talent, real location, real light. This is the part your customer scrutinizes hardest, 4 and the non-negotiable part. The believable real frame is the deliverable.
After the shoot, AI does the grind.
Cleanup, masking, and the dozens of crops, ratios, and color variants a campaign needs. That is a cost saving we pass to you, and it does not touch the truth of the frame, because the frame was real to begin with.
Clean rights ship with the pictures.
Model and property releases and a licensing agreement that states exactly what you own and for how long. We travel worldwide to get the real thing, because the real thing is the point.
Ask any photo vendor exactly where AI touches the work. If they cannot draw the line between ideation, capture, and cleanup, they either do not have a process or do not want you to see it. The line is the discipline.
Be clearly premium and clearly real, or be honestly commodity. The mushy middle is where budgets die.
Labeling and disclosure harden. Platforms and regulators move toward mandatory AI labels, and demand for them is already at 90 percent. 4 Disclosure makes the synthetic-versus-real distinction visible in the feed, which raises the value of anything that gets to display the real label.
The price split widens into two clear markets. Commodity images keep sliding toward free. Premium images, real people, real products, real campaigns, hold and rise as scarcity and trust get priced in. 8 9 The mediocre real shoot that does nothing a machine cannot is the position that vanishes.
Provenance becomes a feature brands advertise. The way a food brand puts organic on the label, brands will signal real, human-made, verified capture. The legal ground keeps settling slowly, 10 which keeps clean rights an asset.
The market is splitting into a free commodity tier and a premium real tier, and the middle is vanishing. Pick the tier on purpose. Do not get caught in the gap.
Anyone can make a picture.
Few can prove a real one.
Use the machine where it belongs, in the ideas and the cleanup, and spend real money where trust lives, on the product, the face, the place, and the campaign a customer will judge you on. In a world where anyone can make a picture of anything, the brand that can prove a real one happened owns the scarce position.
Book a real shootRaised Media Co. · commercial photography & video production, NYC