Creativity Under Siege: Between Algorithms and Academic Bureaucracy
A reflection on how academic bureaucracy constrains creativity—and how AI can help us spend less energy on form and more on the architecture of ideas.
Notebook note: This reflection was prompted by a video originally published by @la_prosodia. The embedded video provides context; what follows is my own reading of the ideas it raises.
The Labyrinth of Form
In this excerpt, the speaker confronts us with an unsettling paradox through the work of Remedios Zafra. We tend to fear that machines will turn us into automatons, yet the video suggests that academia had perfected its own machinery of repetition long before.
It describes a system that rewards a mechanistic, procedural tone: a bureaucratic labyrinth in which researchers are forced to repeat empty forms and write texts aimed less at truth than at satisfying an evaluation metric. This is packaged culture—the endless repetition of the same sterile gesture, with the container taking precedence over its contents.
The Scribe and the Architect
This is where I depart from the usual fear of technology. Generative AI is often said to make us less creative by solving the problem of writing for us. I suspect the opposite: the real danger is remaining trapped on the ground floor of the building, obsessing over the placement of every brick and losing sight of the dome.
Academic bureaucracy wants to turn us into scribes: copyists who spend their lives policing form, protocol, and insignificant details. It prevents us from moving to a higher level of abstraction.
Technology—and these probabilistic models in particular—can offer a staircase out of that ground floor. By delegating part of the scribe’s mechanical work, we can move toward the layer of the architect.
Iteration as a Compass
Far from being a deterministic oracle, an algorithm works with probability. Like a hall of mirrors, it returns variations and branching paths in response to the same prompt.
That is where the key lies: iteration.
Creativity stops looking like generatio ex nihilo—creation from nothing—and becomes an act of direction and selection. The machine proposes variants from the context available to it; we iterate from the standpoint of our own vision. We reject, refine, and ask again, shaping the result step by step.
We do not enter into dialogue with a machine so that it can think for us. We use it to discard the obvious more quickly and reach the precise idea sooner. The genuinely human act in this digital garden is not laying every brick; it is having the judgment to know when the work is finally complete.