A consensus is emerging
Among all the hype at GTC, one point about AI content in product marketing was easy to miss. A consensus is forming on how to make AI production-ready: anchor it to a more fundamental digital truth.
When Adobe and NVIDIA state that a single-governed 3D product identity is critical for generative AI, and LVMH, L’Oréal, Unilever, and Nestlé are all moving in the same direction, the point carries weight.
The real divide is reusable vs single-use
The shift is not about traditional vs AI content. It is about reusable foundations vs single-use ones.
On the surface, AI looks like the opposite of traditional production: fast, flexible, infinite. But without the right foundations, AI shares the same weakness as traditional production: poor repeatability. The ability to recreate a product image, or key elements from it, accurately and consistently over time, without starting from scratch.
Most content is one-off. That is not the issue. The issue is whether it was created from something reusable, or whether its fundamentals have to be reinvented each time.
One-off outputs are fine. Single-use foundations are not
AI without foundations can produce great-looking results, but they are approximations: of the product, of the lighting, of the brand world. For low-value items in simple set-ups, that is often enough.
For higher-value products, pixel-perfect requirements, or complex lighting, approximation does not hold. The issue is not that single-use content exists. It is businesses thinking they are building a repeatable content capability when they are still just producing single-use outputs faster.
“
"We're moving from a world of content production to one of content infrastructure."Adam Cleaver
Chief Creative & Strategy Officer, Collective
It starts with a governed master
Reusable content starts with a governed master. Single-use content starts with an output.
A digital master holds more than a picture. It holds the logic of how the product should appear: geometry, materials, variants, pack wording in every language, surface finish, the rules around lighting and framing. It is the building block that makes AI repeatable.
Around that master, brands can build a library of reusable components: environments, scene templates, props, lighting set-ups, camera passes. The pieces that otherwise get reconstructed or approximated, over and over.
This is what makes meaningful automation possible. Once the product truth, the rules, and the reusable components are defined, more of the future production work can be automated. The library compounds, the same way brand consistency does.
The choice
If you want to introduce AI into production, your content operation is doing one of two things. Building reusable capability. Or producing disposable content faster.
Read Adam’s full piece on LinkedIn.
Related articles
16.04.2026 / News Seven Key Marketing Take-outs from NVIDIA’s GTC Conference
05.11.2025 / News Collective Adopts OpenPBR: A Major Step Forward for Digital Twins and 3D Content
