Adobe has dropped “AI Photo Editor” from the name of its Lightroom mobile app after photographers pushed back on the label. I think the photographers were right, and the reason has little to do with Adobe.
It comes down to words. In imagery, the word you use tells the client what they are getting.
What changed
The Lightroom mobile app had been listed as “Lightroom: AI Photo Editor” for most of this year. Photographers objected on Reddit and Threads, pointing out that Lightroom has long been used to edit photographs taken on ordinary cameras, and that the label made their work sound like something it was not. Dataconomy reported on 4 November that Adobe has now taken “AI” out of the App Store name.
According to the same report, an Adobe employee had said last month that the AI wording was there to help people find the generative features in search. That is a fair commercial reason. It is also a good example of the problem: one word, used to describe very different things.
Why the word matters
A photograph is a record. Someone pointed a camera at something real, and the image shows what was there. Retouching finishes that record. It does not replace it.
A generated image is something else. It may be beautiful, but nothing was in front of a lens. For a client commissioning product or brand imagery, that is the difference between showing a customer what they will receive and showing them what a model imagines.
Shoppers seem to feel the difference. A UK survey published by Photoroom in May found that 31% of consumers say visible AI-generated marketing makes them trust a brand less (IT Brief UK). Photoroom sells AI image tools and the report gives no methodology, so treat the figure as indicative. But it points the same way as my own experience: when people are about to spend money, they look for reasons to trust what they see.
Three words that mean different things
Retouched. A real photograph, finished by a person. What was shot is what you see. For example, removing a stray cable or correcting colour to match the real product.
Composited. Several real photographs combined into one image. Every element was photographed, but the final image never existed as a single moment. For example, placing a product shot onto a photographed background.
Generated. Created by a model from a prompt or a reference. Nothing in it was necessarily photographed. For example, an image built from a text description.
All three are legitimate. They are not the same, and the client should know which one they are paying for.
Where I draw the line
I use AI-assisted tools in retouching every week. Noise reduction, removing a stray cable, sharpening a soft detail. They save time, and I would not give them up.
My line is simple. The tool can do the repetitive part, but the judgement about what to change stays with a person who understands what the client needs. That includes knowing what must not change: the weave of a fabric, the way light falls on a metal edge, the exact shade of a product. These are the things a customer will check against what arrives.
This is what I mean when I say: by hand where it matters, modern tools where they help, and a human eye where it counts.
What to ask when you commission imagery
- Is this a record of something real, or was any of it generated?
- Which tools were used, and for what? Noise reduction is not the same as replacing a background.
- Will the finished image match the product the customer actually receives?
Any good retouching studio should answer these without hesitating. If the answer is vague, that tells you something too.
Plain language
I would like to see the industry use clearer words. “Retouched”, “composited” and “generated” mean different things, and clients deserve to know which one they are paying for.
Where do you draw the line between editing and generating?
If you are planning product or brand imagery and would like to talk it through, get in touch on imaging@naish.co.uk or 07976 752417.
