Beyond Control and Constraint: Generative AI and the Personalization of Experience

Feb 4, 2026

Beyond Control and Constraint - Generative AI and the Personalization of Experience - Luc Nijman

The house of cards that human beings strive to build is now a bet that is entirely radical in its design and its evolution. As I discussed in my previous article (Iterative Layers of Continuous Improvement: Why Entrepreneurship Is Glorified Right Now), the levels of complexity keep stacking on top of each other. If I give you a shovel, and all shovel manufacturers disappear, you will agree that you would find within yourself the necessary abilities to make that shovel yourself. Right? It won’t be perfect, it will be of lower quality, less durable, made with substitute materials, but you would know how to reproduce it from memory. It is a simple technology that only needs to be seen and used to grasp the concept.


But if I ask you, tomorrow, if the entire modern world we have built collapses, to reproduce electricity, data centers, computers, autonomous robots, LLMs; you wouldn’t even know where to start, and even with the knowledge in hand, you wouldn’t succeed, especially not alone. The point I want to make here is that the world on which everything currently rests is incredibly fragile, like everything mankind has ever built. All the more so because it rests on complex and rare geo-economic-political conditions: relative peace, controlled conflicts, prosperous economy, average satisfaction and optimized health… So it is never to be taken for granted.


Now, I move to my point: entertainment and education in the generative computational era. What this enormous capacity for interconnected computation allows us, in a future that I genuinely envision as possible around 2040-2050, is a future where all digital content can be personalized according to the specific and subjective needs of each user. And this actually opens two worlds that are in permanent friction: control, often necessary and sometimes abusive, of populations, and the freedom of individuals and their particular needs. Indeed, every human being has a different vision of the world, even though we converge on points that allow us to connect in order to move forward collectively. This maintains a certain harmony.


That said, there are deviant individuals, people with intentions that exceed our capacity for acceptance, who step outside morality. Morality which, moreover, is also – I believe – utilitarian and a necessity for collective survival and abundance rather than a transcendental truth. I advance this notion in particular to justify the fact that control over certain precise aspects is a necessary reality, and not solely an arbitrary desire for oppression, as some would like to think. To protect what really matters, one must know how to cut and extract the element that disturbs the total equilibrium.


Now another question is to define what really matters, but I am not here to debate that.


The unlimited generation of content with entertainment or educational value is therefore the future of the internet, a place where friction decreases more and more, where user interaction and action become more and more optional. Imagine a series shaped according to your own ideals regarding the quality of interactions, the themes addressed, the expected visuals, and which reacts intelligently based on your state of mind, your physiological feedback, your level of fatigue, attention, need for stimulation, etc… Or educational content, where artificial intelligence brings you, at precise moments, knowledge and data that are important – but important in your eyes – according to your general or momentary interests.


It is a world that is both fascinating and disconcerting: it opens the possibility of total immersion, without friction, with almost no contact with reality, except for the photons traveling through optical fibers or the electricity running through cities. You could lose everything accidental, everything that makes the accident, the unexpected, the unpredictable: a virtual world where even accidents would be auto-generated so that you remain attentive. But also a world where it will potentially be possible, as we see for example with local LLMs, to build the world -virtual here, but with its real physical consequences- that we really desire, a world free from external inferences, at least in part.


Once again it is the same argument I will advance: it all depends on the usage. Knowing how to master such tools would bring comfort, quality of life, feedback on the environment of unparalleled quality, but for those who get lost in their impulses and delegate all reflection, it will prove potentially a trap, which will demand a great deal of dexterity and mental gymnastics to grasp the system that will have revealed itself to us without us even noticing.