I’ll start in a slightly unusual way. It’s quite possible that if we reread this article in a few months or a few years, we’ll have a good laugh at it. Artificial intelligence is currently developing so fast that the word “exponential” is gradually ceasing to sound like an exaggeration. What was a technology demo a year ago is an everyday work tool today, and what we now consider the limits of AI may be completely irrelevant a year from now. That’s why I don’t see this text as a definitive answer to the question of what AI will do to the creative industry. It’s more a record of where we stand in September 2026 and how we use AI at MAISON D’IDÉE today.
Honestly, we don’t know what its real long-term impact on creativity will be. We don’t know how the perception of authenticity will change, or how audiences a few years from now will respond to communication created by humans versus communication created by machines. Perhaps this distinction will eventually stop making sense altogether. Nor do we know to what extent AI will change the very essence of communication design, advertising and brand building. But one thing we already know: AI is an extraordinarily powerful tool. And I think the word tool is what matters.
As Creative Director of MAISON D’IDÉE, I look at AI primarily from the perspective of visual communication and creative work. And despite all the possibilities it offers today, our creative process still very often starts in the most ordinary way – with paper, a pencil, a sketch and an idea. Not because we’re nostalgic or reject technology. Quite the opposite, we use AI intensively. But paper forces you to do something I consider essential in the creative process: to use your own head first. To come up with an idea, to have a sense of what we want to say, why we want to say it, and how it relates to the brand we’re creating for.
Only then does AI come in. We use it to quickly visualise ideas, create variations, complete photographs and illustrations, try out different directions, or simply to get an idea out of our heads and into visual form very quickly. The result is then often further refined by hand by a graphic artist or designer. So AI helps us realise a vision, but it shouldn’t be the one defining that vision. At least today, I see a huge difference there.
Our clients themselves give us an interesting test. Recently, we’ve presented a large number of proposals in which we used AI to varying degrees. And we keep observing one thing. When we put concepts created primarily through AI side by side with concepts that began as a human idea or sketch, with AI used only afterwards to develop them, clients repeatedly choose the second group. Of course, this isn’t a scientific experiment, and I’m not claiming it will always be this way. But for me it’s a very interesting practical experience. The human ability to create a vision, work with context, draw on long-term experience and make a qualified judgement remains absolutely critical in creative work. AI can show you very quickly what something could look like. But a human still has to know why it should look that way.
On the other hand, it would be a mistake to see AI as the enemy of creativity. Quite the contrary. From the client’s point of view, I believe it significantly raises the quality of the service. It allows us to explore considerably more options in the same amount of time, test more directions and get much further in the design phase. So it’s not just that AI lets us do some things faster. What matters more is what we do with the time we gain. Instead of handing the client the same work in less time, we can deliver more quality work in the same time.
We see this very clearly, for example, when working with stock photo libraries. Until recently, as graphic artists and designers, we spent a great deal of time searching for a photo, illustration, icon or other visual element that at least came close to what we had in mind. We often then had to adapt the creative itself, to some extent, to whatever was available. Today this relationship is reversing. We can first have an idea and then create visual material that matches it. Already in the design phase, we can show the client a visualisation that may be 80 or 90 percent of the way to where we want the result to go. The client doesn’t have to “fill in the blanks” as much about what we’re actually presenting, and at the same time we see much sooner whether the concept really works.
Logically, a question then arises: if AI can create an image, write a text or prepare dozens of variations in a few seconds, why doesn’t the client simply create the ad themselves? Of course they can. Just as they have long been able to create their own logo, website or advertising banner. But the question was never only how to create something. The more important question is what to create and why.
If you ask AI to create an ad for a bank, a carmaker, a real estate development or a fashion brand, you’ll probably get something usable. Maybe even something very nice. But who decided that this is the right direction? How does this proposal relate to what the brand communicated five years ago and where it wants to be in the next five? How does it differ from the competition? What is sustainable in the long term, and what is just a passing visual trend? What should a person feel after encountering the communication, and what should they remember? And above all – why is this solution right for this particular brand? At this point, knowing how to use the tool well is no longer enough. This is where strategy, experience, taste and human judgement come in.
This is also where I see the greatest current risk of AI in creative work. If you start the whole process with the question “What could we do?” and ask it of artificial intelligence first, you can very easily end up with something that is nice, professional, polished and technically correct, yet strangely familiar. Generative AI works with an enormous number of existing patterns, looks for connections between them and produces probable solutions. Put simply, to a certain extent it averages what already exists.
And the average can be very nice. It can be technically perfect, professional and, at first glance, even creative. But it can still remain average. And a client who comes to a creative agency usually doesn’t come to get an average solution. They come precisely because they need to find something of their own.
That’s why I think an even more interesting question will be what happens when practically everyone uses the same or very similar AI tools. A strange paradox may emerge. Never in history have we had such powerful creative technologies at our disposal, and yet we may end up in a world where a huge amount of communication looks alike. Even today, you sometimes see an image, an ad or a post and, without anyone telling you, you get the feeling that “this is AI”. Technically, there may be nothing wrong with it. The problem arises when the communication of dozens of different brands evokes the same feeling.
So I believe that the more powerful AI becomes, the more, paradoxically, the value of an original human idea may grow. Technical execution will keep getting easier and more accessible. Creating a beautiful image, animation, video or text will become less and less exceptional. But having a good idea won’t automatically get easier. And perhaps the very ability to come up with an idea that an algorithm wouldn’t offer on its own as the most probable one will be among the most valuable creative skills.
The question of authenticity remains similarly open. Will people accept AI photographs, AI models, AI videos, synthetic voiceovers, or entire campaigns without a single real photographed person? Probably yes, at least in some areas. At the same time, though, we’ll have to think much more carefully about when to use AI and when not to. Not every technological possibility is automatically the right communication decision. If the essence of a brand is craftsmanship, a human story or authenticity, a perfectly generated image may paradoxically be weaker than an imperfect real photograph.
That’s why I don’t think a good agency of the future will be defined by using AI absolutely everywhere. Rather, by knowing where to use it, how to use it, and where not to use it. AI should expand the possibilities of a creative team, not dictate its taste.
For a brand’s communication to be truly good, it isn’t enough for it to be nice. There has to be an opinion, a vision, and an understanding of the client, their product, market, competition and audience behind it. There has to be a reason why the solution is exactly this way and not another. This is where years of experience of the creative director, strategist, designer, copywriter and the whole team come in. Things that worked and things that didn’t. Hundreds of previous projects, discussions with clients, customer reactions, intuition, taste and cultural context. And sometimes simply the judgement that tells you that even though something looks great at first glance, it isn’t right for this particular brand.
That’s why at MAISON D’IDÉE we don’t want to fight AI. Quite the opposite. We want to use every tool that allows us to do better work for our clients. If AI lets us create ten alternatives instead of three more quickly, we’ll use it. If it lets us create a more accurate visualisation, we’ll use it. If it saves us hours of mechanical work and gives us more time to think, it’s a clear benefit for us. But we don’t want to mistake efficiency of production for quality of thinking.
Because our job isn’t to produce as much graphic design as possible. Our job is to find the right idea and then use all available tools – including AI – to execute it as well as possible.
Maybe I’ll read this article in two years and think about how wrong I was. At the current pace of AI development, that’s entirely possible. But in September 2026, this is how I see it: AI is a fantastic tool in the hands of a creative person. Not a replacement for the reason they create.
And the easier it becomes to create a nice image, video, text or an entire advertising campaign, the less exceptional the mere ability to create something will be. What will matter all the more is knowing what to create, who to create it for, why to create it exactly this way, and where it takes the brand in the long run.
And that is where I see the role of a creative agency, even in the age of artificial intelligence.
