What was the starting point for Everything Must Go?
As a project, it is a continuation of two previous works I created with Forced Entertainment around lip-syncing. On stage, the performers lip-sync texts that I wrote and that are vocally generated by artificial intelligence. We are interested in this fusion between the human and the non-human because it reveals something about what it means to be human today and to live surrounded by technology that constantly limits and redefines the space in which we evolve. There is a form of anger and sadness in that.
You chose to stage this reflection late at night, in a bar that may be located in the centre of a city, unless it lies beneath the surface of the Earth… Why this setting?
I was drawn to the idea of a fictional place with a distinctive atmosphere. In a bar, people can meet, speak with strangers, reflect on their own lives… It is a place where countless lives intersect. I like the idea that it is more fantastical than real, a refuge after catastrophe, after collapse. We do not know whether it is late at night or early in the morning, or even whether the bar is closed. At certain moments in the performance, we are truly inside a bar, while at others it becomes simply the stage itself. The black-and-white floor resembles a chessboard, conducive to optical illusions. The fiction is in perpetual motion, endlessly transforming itself: much like today’s digital realm, or AI-generated images and videos that morph into one thing or another, like a dissolving landscape. In the performance, everything changes constantly: the actors change costumes or voices, the stories they tell seem to be heading somewhere before suddenly veering off course. I am interested in this fluidity, this instability that echoes the world we live in today.
In this universe, the stories being passed around are often fragmentary.
The fragmentary form lies at the heart of my work. It has fascinated me throughout the roughly forty years of my creative practice. I am drawn to the idea of fragments that spectators must complete themselves: you do not know where they begin or end, yet you must negotiate with that uncertainty. It creates a very active experience for the audience of a performance or for the reader of a novel. For me, it is also a way of understanding the nature of our existence, as we constantly live surrounded by images and fragments of our lives. Your phone randomly shows you a picture from six months ago or ten years ago, and you have to reconstruct the memory attached to it…
Another characteristic of your work is humour—a certain kind of absurd humour. What do you mean when you say that Everything Must Go is a joke in which nothing is funny?
With Forced Entertainment, we like to establish a dynamic relationship with audiences. Our performances dance with the audience’s attention and reactions. I think one of the most striking things a performance can do is play on the boundary between what is funny and what is terrible, between what is comic and what is tragic or painful. Shifting from one to the other is a powerful gesture: it forces the audience to watch the stage with great attentiveness. The performers’ energy is another essential aspect of my work. Because I work with AI-generated voices, there is a risk of creating something cold and authoritarian. But what matters most is this collective of six actors working together on stage, their extraordinary attentiveness to one another, their ability to exist together, the life they inject into the performance. Without them, this technical setup would simply be a sequence of recordings. Yet throughout the creative process, the actors ignite it and lead it in completely unexpected directions. That is the key to the performance. Very often, the best projects are those in which we know how to listen to what is happening on stage and allow ourselves to be surprised by it.
Would you go so far as to say that this ability to allow oneself to be surprised is one of the differences between AI and artistic creation? Because in the first case, AI responds to a request that we formulate by giving us what we expect, whereas in artistic research, you look for one thing and end up finding something else?
In the case that interests me, I think AI developers are trying to create a useful tool capable of producing text: a serious, pleasant voice that can read us a book or explain things to us like a friend would. They try to anticipate our uses and give us a certain degree of freedom to control the reading speed, slower or faster. What I am trying to do is render those predictions ineffective, to make the result unpredictable, to disrupt the programming. During the creative process, I sometimes wrote texts without punctuation, texts that were incorrect or ungrammatical. When I had the AI read them aloud, the result was strange, surprising, full of glitches: what I could hear was the machine struggling with my text. I think technologies continue to progress, and failures are constantly fed back into the system in pursuit of a certain idea of perfection. But perfection does not interest us. What interests us are the mistakes and the bugs.
Interview conducted by Simon Hatab in March 2026