AI is most effective when its constructed nature is clear. Abstraction, transformation, speculative worlds, impossible environments, symbolic and narrative world-building, this is where generative work amplifies imagination rather than competing with reality. It becomes fragile in the opposite territory: hyper-real simulation that mimics documentary or lived experience. When realism is ambiguous, trust erodes. When the work is openly expressive, it resonates.

When the brief lives somewhere a camera cannot follow, or where the value of the work comes from creative possibility rather than physical reality. That includes impossible or speculative environments (a city built from light, a product floating through a surreal landscape, a brand world that does not exist anywhere on earth), large-scale visual concepts that would be logistically or financially out of reach (multiple environments in a single film, ambitious art direction that no location or set could practically deliver), and high-volume creative work where consistency and iteration matter more than a single hero asset. Full AI is also a strong choice for early-stage creative exploration, letting concepts be visualised and tested at near-final quality before bigger commitments are made.

Human performance is the clearest case. While AI can carry emotionally resonant work when it is well directed, there is still a category of moments, the unscripted glance, the live chemistry between performers, the involuntary detail that nobody planned, that only a real shoot can deliver. Dynamic product-in-motion work sits in similar territory: liquid pouring, fabric moving with a body, food being prepared in real time, the kind of continuity and physical behaviour that is still best served by a camera. When the brief lives in those spaces, we shoot.

Yes. Photorealism and brand consistency sit at the centre of our process. Lighting, perspective, texture, reflection and composition are controlled with the same rigour you would expect from a traditional shoot, and every output is held to professional marketing standards. Visuals integrate cleanly with your existing brand identity and deliver at the resolution and quality required for campaign-level use.

Yes. Lifestyle work depends on emotional truth, and emotional truth comes from direction, not from the tool that renders it. When AI is steered with real cultural perspective, considered casting, and a clear point of view on what feels true, it can carry a lifestyle campaign and evoke the same response as a traditional shoot. The camera never made the film. Neither does the AI tool.

Only if we would let the tool lead. Generative models, left alone, produce by probability: a statistical average of what already exists. That is where sameness comes from. The antidote is point of view: a consistent visual logic, a clear sense of what belongs and what does not, and the willingness to refuse anything that feels obvious. AI is the given. Judgment is the variable.

Both can be part of the work, and we treat them as two separate decisions.

‍

AI-generated people are a strong fit when the role is to represent a type rather than a specific individual: a family in a kitchen, a crowd at a concert, commuters in a city, characters in a stylised or speculative world. The brief calls for human presence, not human recognition.

‍

Real talent is the right choice when the campaign relies on a specific person being recognised, when the performance carries something only that individual can bring, or when the role itself is built around association with them. We work with proper licensing for their likeness, and that licensing extends to any AI-assisted use of their performance, things like lip-sync localisation into other languages, additional coverage generated from existing footage, or scene extensions where they cannot physically be present. Consent, scope, and compensation are agreed before any of that work begins.

Yes, and this is one of the strongest arguments for modern production. Once your visual system is defined, a single core production becomes the foundation for a wide range of versions: regional backdrops, localised dialogue with lip sync, market-specific edits, and channel-specific formats across social, TV, e-commerce, print and OOH. You get global consistency and local relevance from the same creative backbone, rather than rebuilding from zero for every market.

In most cases, it makes production meaningfully cheaper. A set that would once have required a full art department, build, and strike can sometimes be generated entirely with AI, or built physically only for the shots that genuinely need it. The same logic applies to environments, crowd extension, product variants, and many other elements that traditionally drove cost. Just as importantly, the savings change what becomes possible. Ideas that would have been dismissed at the first budget review (multiple environments, large-scale visual worlds, dozens of campaign variants) move inside reach of a brief that previously could only afford one execution. Cheaper is part of the answer. The bigger part is what the budget can now do.

Significantly, when the brief allows for it. Modern production compresses traditional production stages without compromising creative output, which means moving from brief to finalised assets in a fraction of the time a fully traditional process would take. For brands operating at the speed of culture, that responsiveness has shifted from luxury to a competitive requirement. That said, faster is not the goal in itself. We use the time saved upstream to invest more attention into the parts of the work that benefit most from it: direction, iteration, and craft.

On training data: We do not prompt with direct references to other creatives' work. References inspire, the way they always have, but we never feed another artist's images to AI as something to copy. Every project carries a creative signature: our own art direction, or the vision of the artist we are working with. We actively seek out AI artists who have a visual perspective of their own, not the ones who are skilled at imitating an existing style. That choice does real work to sidestep one of the core ethical problems with generative tools.

‍

On the people who appear in the work: When we generate human faces, we run them through similarity checks (tools like FaceCheck.ID, PimEyes, and platform-level safeguards) before anything reaches a client, to flag any drift toward a real person's likeness. That said, we cannot guarantee a generated face is 100% unique, which is why our actual preference is to work with real models under proper consent agreements that explicitly cover AI-generated extensions of their likeness: additional coverage, lip-sync localisation, scene extensions. The generated layer sits on top of a consented performance, not in place of one.

‍

On bias: Cultural bias in AI output is not a model problem we hand off to better software; it is a creative decision made at the art direction stage, every time. It is addressed in how we brief, how we prompt, which references we choose, and (just as importantly) what we actively reject. Experienced creatives carry that responsibility into the work. The model does not.

‍

On the people working in the creative industry, whose roles are changing: We will not pretend the disruption is not real. What we can say is that the AI artists we work with came almost entirely from traditional production: retouching, photography, motion graphics, CGI. The roles are changing shape, not disappearing. The craft transfers; the workflow adapts. That is the more honest picture than either "AI takes the jobs" or "nothing has changed.

‍

AI without ethics is a shortcut. Modern Production is not.

The licensing framework is the same as in any production: usage is agreed per project across media, markets, duration, and scope. What changes with modern production is what sits inside that framework.

‍

Fully AI-generated talent carries no portrait rights and no buy-outs, which keeps usage costs significantly lower. Real talent, or AI-assisted use of a real person's likeness (lip-sync localisation, scene extensions, additional coverage), involves licence agreements that travel with the work, and that's reflected in the cost. Hybrid productions sit somewhere between the two, depending on which elements are real and which are generated.

‍

Usage costs are always included in the total quoted amount, and the licence terms are always laid out explicitly in our proposal. The pricing follows what has actually been used, not a flat assumption.

‍

Yes. Our workflows are designed for commercial safety across markets, with clear ownership of inputs and outputs and alignment with the regulatory frameworks that apply to each region, from the EU AI Act to emerging standards elsewhere. Where provenance and transparency are required, we work with standards like C2PA and SynthID. You receive clear documentation of how each asset was produced, so your legal and compliance teams have what they need to sign off with confidence.

All production has a carbon footprint, and AI usually has a smaller one than traditional alternatives: fewer flights, smaller crews, less physical build. In the right project, it directly replaces the highest-carbon parts of a traditional shoot, like long-haul travel to multiple locations or large physical sets that are built once and dismantled. That said, "usually" is not "always," and the honest answer depends on the project. The brands we work with increasingly want carbon data alongside their final assets. We treat that as part of the deliverable, not a separate request.

‍