Generating a product image used to mean a studio, a model, a photographer and a day. Generating one now means a prompt and a few seconds, at a tiny fraction of the cost, and the quality is good enough that a large share of ecommerce imagery is already made or finished this way. Generative AI has not nudged the economics of product creative, it has broken them, and the same is happening to product video. For a brand with thousands of SKUs, that is a genuinely large shift, and most boards have not yet thought past the obvious conclusion.
The obvious conclusion is that imagery just got cheap, so cut the budget. That is true and it is the least interesting thing you can do with this. Treating a collapse in the cost of creative as a saving is like treating the arrival of cheap printing as a way to spend less on letters. The brands that win this shift will not be the ones that spent the least on imagery. They will be the ones that did things with near-free, near-infinite creative that were simply impossible before.
This went mainstream faster than most boards noticed
It is worth being clear-eyed about how far this has already gone. Generating model and lifestyle imagery at scale, producing dozens of variants of a product shot, turning a single still into video, all of it has moved from experiment to production line inside the largest retailers, and the tooling to do it is now in reach of brands of every size. The cost gap between traditional and AI-generated creative is not marginal, it is an order of magnitude, and where the cost of doing something falls that far, behaviour changes completely rather than incrementally.
For a leadership team the right framing is not "should we use this" but "what does it make newly possible". When creative is abundant and cheap, the constraint that shaped every previous decision, that good imagery is expensive and slow, simply lifts. Strategies that were uneconomic become obvious, and the brands that rethink from that new constraint move while everyone else is still negotiating a smaller photography invoice.
Cutting costs is the boring use
The interesting uses come from abundance. Imagery for the long tail of products that never justified a shoot, so the whole catalogue finally looks considered. A different image for every audience, channel and context, tested and personalised at a scale no studio could match. Rapid creative iteration, where you try twenty concepts in the time it used to take to brief one. Video for products that would never have had it. None of this is about spending less. It is about doing far more with creative than was ever feasible, and that is where the advantage sits.
This is the early-mover's window. The brands experimenting now are learning how to wield abundant creative, what works, what converts, how to keep it on-brand, while their competitors are still treating AI imagery as a procurement saving. That learning compounds, and by the time the laggards arrive the early movers will have a creative operation the others cannot quickly copy. It is the kind of capability we help brands build through joined-up creative and social work, where volume and brand have to hold together.
The authenticity tension is real, and it bites
Here is where the early-adopter energy has to stay grounded, because there is a genuine countercurrent. Customers are not naive about AI imagery, and trust matters more than novelty. When a brand uses AI to misrepresent how a product actually looks, fits or performs, it does not save money, it manufactures returns and erodes the trust the whole business runs on. There has already been visible backlash when brands pushed AI-generated models too far and had to walk it back toward real photography for the campaigns that define them.
Some customers, particularly for considered or fit-sensitive purchases, simply trust real imagery more, and a premium brand that quietly swaps authenticity for efficiency risks the very perception its price depends on. So the opportunity and the risk live side by side. Abundant AI creative is a gift for the unglamorous majority of imagery and a liability if it is used to fake the things customers most need to be true. Knowing the difference is the whole skill.
Where to use it, and where not to
The practical line is clearer than the hype suggests. Use generative AI to scale the work that was always going to be functional: the long tail, the variants, the contextual and channel-specific versions, the rapid testing. Protect the brand-defining moments, the hero campaign, the imagery that carries the brand's promise, where authenticity and craft are the point and where customers are paying attention. And never, in any case, use it to misrepresent the product itself, because that is not a creative decision, it is a trust decision, and you lose it.
This mirrors the discipline that has always separated good brands from careless ones, just with a far more powerful tool in play. The question is not whether to use generative creative, it is where it amplifies the brand and where it would quietly cheapen it. The same judgement applies to words: generative copy can produce product descriptions at catalogue scale, and the brands that let it flatten a distinctive voice into bland competence trade a real asset for a small efficiency. We have helped premium brands like Hackett London hold that line between scale and brand.
Governance is now part of the creative job
One genuinely new responsibility comes with all this. As AI-generated content fills the storefront, provenance and disclosure stop being abstract. Content authentication standards that embed where an image came from are becoming part of how trust is signalled, and regulation is arriving, with AI transparency rules tightening in the EU and elsewhere through 2026. A brand operating at scale now needs a position on disclosure, a quality bar that AI output has to clear before it goes live, and a record of what was generated and how.
This is not a reason to hold back, it is a reason to build the operation properly. The brands that treat governance as part of the creative process, rather than a compliance afterthought, will move faster and more safely than those improvising, because they will not have to slam the brakes when a rule or a backlash arrives. Early and responsible is not a contradiction here, it is the winning combination.
What to do now
Concretely: stop thinking of AI creative as a cost line and start mapping what abundant imagery and video would let you do that you cannot today. Draw the line clearly between the functional creative you scale and the brand-defining creative you protect. Set a quality bar and a disclosure position before you scale, not after. And start experimenting now, in a contained way, so your team learns to wield this while it is still an advantage rather than a baseline expectation.
The economics of creative have changed permanently, and the brands that treat that as an opportunity to do more rather than spend less will pull ahead. If you want help working out where generative creative fits in your operation, what to scale and what to protect, our marketing consultation is built to turn a fast-moving capability into a plan that strengthens the brand rather than diluting it.
