Creating your stand with AI: possibilities, limits and real risks for exhibitors

Today, a few lines in ChatGPT are all it takes to get a first stand proposal. A premium atmosphere, a big lit-up arch, a lounge area, a few screens, sophisticated materials. In a few seconds, your next trade show already seems to be taking shape. And honestly, it's pretty impressive. For a brand looking to explore a world, test a mood, or simply put images to an idea that's still a bit vague, artificial intelligence has become a fantastic tool. But there's a problem. The AUKARÉ team explains.

What AI really does well

Giving shape to a vague idea

When a project is still just at the intuition stage, AI is useful. It explores worlds, color combinations, lighting moods, architectural references. For an exhibitor who knows what feeling they want to create without knowing how to translate it visually, it's a wonderful help.

Want something premium, warm and contemporary? AI can test ten directions in an hour. Torn between an architectural world and something more organic? It generates both in a few minutes.

At this stage, it plays its role perfectly: turning an idea into a visual. And for a design agency, receiving a moodboard built from several references can be very useful.

"Sometimes it's more useful for us to get a moodboard than an already-finished 3D render straight out of AI."
Rémi, designer at AUKARÉ

Enriching 3D renders without slowing down the process

In the day-to-day practice of a stand design agency, AI has one concrete, effective use: once the structure has been designed by a designer, it generates the decorative elements that bring the render to life. People, plants, trinkets, accessories. Time-consuming elements to produce manually, done in a few minutes with AI.

"It's time-consuming and doesn't add much to the design itself, beyond putting it in a more 'real-life' setting."
Rémi, designer at AUKARÉ

This is probably the fairest way to integrate AI into a design process: on tasks that consume time without adding creative value to the concept.

What AI doesn't always get right for stand design

Distorted stand dimensions

This is the most common mistake, and often the least detected. AI adjusts proportions so the render looks aesthetically balanced on screen. It doesn't always try to respect the actual show floor plan.

"When you ask it for a 6 by 4, it won't always stick exactly to the right dimensions."
Rémi, designer at AUKARÉ

This isn't a bug. It's how the model normally works: it optimizes the image, not the dimensions. A 6 x 4 stays a 6 x 4. You can't quietly enlarge the room to fit in the bar, the lounge, the storage area and the LED screen. What you're really paying for your stand per square meter starts with real dimensions, not visually appealing proportions.

Show regulations not factored in

Every organizer imposes its own rules. Maximum height per hall, safety distances, constraints on hanging signage, fire regulations, floor load limits. This data isn't part of the AI's training dataset. It generates what looks beautiful in the abstract, not what's authorized in Hall 5 of a specific venue for a specific edition. Discovering on-the-ground constraints during the floor plan validation stage means starting the whole design over again.

Visitor flow not anticipated

In a 3D render, everything looks smooth. The visitor enters here, discovers a zone, heads toward the demonstration area. On a real show floor, it's rarely that clean. Visitors arrive from several directions. They stop, cross paths, change course, and sometimes come in from the side nobody had planned for.

"You end up with oddities, like a counter placed against a partition that nobody can get behind."
Rémi, designer at AUKARÉ

In the image, the counter looks lovely. In reality, someone still has to serve visitors from it. Visitor flow isn't a matter of geometry, it's a matter of behavior. And that behavior is learned by walking the halls over dozens of show editions, not by training a model on photos of stands.

"You expect people to come in from there, but actually, no. In reality, people often come in from the other side."
Rémi, designer at AUKARÉ

Budget unknown

AI has no trouble imagining spectacular architecture, high-end materials, suspended elements and fully custom furniture. The render can be magnificent. The budget, however, doesn't get generated along with the image. Two visually similar stands can have radically different production costs. And some materials that appear available in a render have manufacturing lead times that make taking part in that particular show impossible. AI doesn't know any of that.

Logistical constraints not factored in

In an image, a material stays a material. In a stand, it becomes a technical choice. A wall has a thickness, a structure has a weight, a suspended element has to be fixed, a piece of furniture has to be manufactured, transported, assembled, used over several days, then sometimes dismantled and reused. This shift from image to object requires expertise and an understanding of the entire manufacturing, build-up and tear-down process that AI cannot simulate.

Brand distinctiveness that ends up blending into the crowd

This is the most discreet problem, and one of the costliest. By using the same tools, the same models and the same references, you gradually end up with the same visual codes. Rounded volumes, LEDs, giant screens, light wood, greenery. Taken separately, none of it is a problem. But when a hundred brands use the same codes at the same show, differentiation disappears.

"You end up with the same stands all the time, always the same kind of thing."
Rémi, designer at AUKARÉ

A stand must above all be recognizable. It has to look like your brand, speak to your audience, create an experience visitors will associate with you. An AI trained on millions of stand images tends to produce the statistically average version of what "premium stand" means in its data. That's not your stand.

How AI can fit into your brief to your stand agency

The AI render isn't the project, it's a lead

An AI moodboard gives a direction. An AI-generated 3D render can create the illusion of a solution. These are two very different things. When an agency receives a moodboard, it can take that direction and translate it with its know-how. When it receives a stand already locked into an image, it sometimes has to start by deconstructing what doesn't work before it can even start creating.

Don't try to arrive with the stand already drawn

Try to arrive with a very clear intent instead. A moodboard, references, colors, materials, worlds. And above all, examples of what you don't want. This gives the designer a direction without taking away their playground.

"It gives direction, you kind of know the overall mood you want, but it still leaves us free rein to actually create the stand."
Rémi, designer at AUKARÉ

And if you spent your Sunday evening generating dozens of versions of your future stand with ChatGPT: keep them. Complete them with what gives the project its substance: exact dimensions, the show's floor plan, regulatory constraints, available budget, your objectives, your target audience, the essential features. A well-built brief upstream makes the difference between a useful creative lead and a back-and-forth that drags on for weeks.

AI has changed how brands approach stand design. That's good news. It makes creative exploration accessible, turns an abstract idea into visual material, and speeds up certain production phases. What it still can't do: factor in the show's technical regulations, how your visitors actually behave in that specific hall, build-up constraints, the materials available within your budget, and real manufacturing lead times. That data can't be prompted into existence.

A stand isn't an image. It's a real space with real constraints, moved through by real visitors, built with real materials, used over several days by real teams. AI opens up the creative conversation, but field expertise has the final word. Let's talk about it?

FAQ

Can AI design a trade show stand for me?

It can generate useful visual directions and moodboards. It cannot design a stand in the operational sense: real dimensions, show regulations, visitor circulation, budget and manufacturing constraints can't be captured in a prompt. These elements determine the stand's performance and require field expertise that AI doesn't have.

What are the most common mistakes in an AI-generated stand render?

The five most common: proportions distorted relative to the actual dimensions of the space, ignored circulation (impassable counters, blocked zones), structures that don't meet height regulations, materials that don't exist within the available lead times, and budget left out of any constraint. These errors are structural and independent of prompt quality.

How should you use an AI render to brief a stand agency?

By presenting it as a creative direction, not a specification. Complete it with the information that gives the project its substance: exact floor area, the show's floor plan, regulatory constraints, budget, objectives and target audience. The render provides direction. The brief lets you build.

Is an AI moodboard or an AI-generated 3D better for briefing your agency?

The moodboard is often more useful. It gives a creative direction without locking in an architecture that might be unworkable. An AI-generated 3D render can create precise visual expectations that the designer may have to deconstruct before they can design the real stand.

Will AI replace stand designers?

No. It automates repetitive visual tasks and speeds up creative exploration phases. It doesn't replace knowledge of trade shows, regulations, suppliers, real visitor flows, or the ability to design a space that performs commercially. These skills are learned through experience and can't be generated by a prompt.