BENGALURU — The question facing marketing teams has quietly changed. A year ago it was whether AI photography was good enough to use. Now it is where to stop. In Adobe’s 2026 AI and Digital Trends Report, produced with Oxford Economics from responses by 3,000 executives and practitioners worldwide, 76% of organisations said generative AI had delivered moderate to significant gains in the volume and speed of their content production. The capability is settled. The judgement is not.
Vinay Kumar Nevatia, a Bengaluru-based AI photographer working with startups and direct-to-consumer brands, spends more of his time on that second question than the first.
“Nobody asks me any more whether the image will look real,” he says. “They ask which images they are allowed to make.”
Why AI Photography Became a Default So Quickly
Adoption ran ahead of policy because the pressure was practical. HubSpot’s AI Trends for Marketers research, drawn from more than 1,000 marketing professionals, found that among those already using generative AI for content, 48% are creating images with it.
The reasons are unglamorous. Paid social media consumes creativity faster than any small team can shoot it. Marketplace listings want six angles per product. A seasonal campaign arrives whether or not a studio has a free day. AI product photography solved a volume problem that conventional production was never priced to solve for a brand doing modest revenue.
What changed underneath, Vinay Kumar Nevatia argues, is the shape of the cost. A traditional shoot front-loads nearly everything, and each new variation means starting again. Generative work inverts that: the first usable image is the slowest, and every version after it is cheap. For a brand deciding which of six concepts deserves a budget, that inversion changes what is worth attempting at all.
Vinay Kumar Nevatia on the Difference Between Illustration and Evidence
His working distinction is a simple one, and he applies it before opening any tool.
Some images illustrate — they set a mood, suggest a context, show how a thing might sit in someone’s life. Others act as evidence — the buyer studies them to decide what they are actually getting.
“Generate the illustration. Photograph the evidence,” he says. “Almost every problem I have seen came from someone blurring those two.”
Where generative work holds up
Brand campaign visuals are the strongest case. A single product can be placed across a dozen settings, seasons and lighting moods while holding one identity, which gives a team without an art director a supply of fresh assets that still look related.
Concept testing is the second. A marketer can produce several treatments of an idea, run them, and commit real money only to the one that earns attention.
AI headshots are the third. Distributed teams rarely have matching portraits, and a founder page assembled from five different rooms undercuts a brand before a visitor reads a word.
Where a camera is still required
Anything a buyer relies on to make the purchase decision stays photographic: the actual item, its finish, its true colour, its scale in a real hand. So do results claims of any kind, and so does anything depicting a real customer.
“Fabric is the obvious one,” Vinay Kumar Nevatia says. “A generated weave looks lovely and tells the customer nothing about what arrives in the parcel. That is a return waiting to happen.”
The Disclosure Question Brands Are Now Facing
The regulatory position is less exotic than the technology suggests. Advertising standards bodies in India and abroad assess whether a claim misleads, not how the visual was produced. A synthetic image that implies something a product cannot do is treated as the claim it makes.
Major platforms have separately tightened their handling of synthetic and manipulated media, and disclosure expectations continue to firm up across ad networks and marketplaces. For a brand, the practical consequence is that generative AI in marketing has become a documentation problem as much as a creative one: knowing which assets were generated, and being able to say so.
His own rule is procedural rather than philosophical. Every generated asset is logged as generated. Nothing that a customer could reasonably read as proof goes out without a real photograph behind it.
Craft Is What Decides Whether Anyone Notices
The gap between AI photography that performs and the kind that reads as cheap is, in his account, almost entirely post-generation.
“The first output is a draft. Treating it as the finished piece is where the stock-image look comes from.”
He reworks generated images for the details a trained eye registers before the conscious mind does: light that falls the way light actually falls, texture that survives a close crop, edges and hands that hold up under attention, and colour pulled back into the brand’s palette rather than the model’s defaults. The benchmark is an image a customer scrolls past without pausing to question it. Brands considering the approach can see how he combines generative tools with a photographer’s eye through Vinay Kumar Nevatia.
The wider industry appears to be reaching the same conclusion. Access to the tools is now near-universal and therefore worth little. Art direction, brand discipline and the willingness to discard nine outputs to keep one are what separate the results.
What This Means for India’s D2C Sector
The effect is sharpest among Bengaluru’s smaller consumer brands, where ambition has rarely been the constraint. Creative volume was the ceiling on how fast a four-person team could move, and AI photography has lifted it.
Vinay Kumar Nevatia does not read that as a displacement story. In his experience, the photographers around him are spending less time on repetitive catalogue and background work and more on the shoots that genuinely need a lens — the hero product frame, the founder portrait, the campaign that has to carry a year.
“It removed the part of the job nobody wanted to pay for,” he says. “It did not remove the part that requires someone to decide what good looks like.”
For brands starting out, his advice is dull on purpose: settle the visual identity first, then generate. Teams that lead with the tool end up with a large pile of images that share no family resemblance. Teams that lead with a defined look — palette, framing, lighting, mood — end up with a library.
The Outlook
The pattern emerging is redistribution rather than replacement. Production is getting cheaper; judgement is getting more valuable. The brands doing well with AI photography are not the ones generating the most images, but the ones that decided in advance which images they were prepared to stand behind.
That is the position Vinay Kumar Nevatia argues for, and it is a restrained one. Generative tools have handed small brands a creative budget they never had. The ones still benefiting in three years will be those who kept a clear line between what they illustrated and what they proved.
About Vinay Kumar Nevatia
(Vinay Kumar Nevatia is a Bengaluru-based AI photographer and visual storyteller who produces studio-quality product images, brand campaign visuals and professional AI headshots for startups, D2C brands and independent professionals without a physical shoot.)