Ethics, Technology, and Human Space

In "Seremos Dados," Marcus Bruzzo proposes a fundamental reversal of the debate about artificial intelligence. The question that dominates popular culture and much of academia ("will AI be more intelligent than us?", "is AI conscious?", "will AI replace us?") is, for Bruzzo, the wrong question. It assumes the problem is one of comparison: human versus machine. But the real question is not about competition. It is about space.

What happens to human space as we progressively delegate our cognitive, creative, and decision-making functions to automated systems? The question is not whether AI writes better than a human, but what happens to the human capacity to write when billions of people stop exercising it. It is not about whether AI makes better decisions, but what happens to the human capacity to decide when decisions are outsourced. Bruzzo’s thesis is that the risk is not the machine’s superiority, but human atrophy.

The first dimension of risk identified by Bruzzo is the erosion of cognitive agency. We outsource memory to smartphones, spatial navigation to GPS, writing to generative AI, and calculation to calculators. Each delegation seems reasonable on its own: why memorize a phone number if the device stores it? Why learn to find your way around the city if GPS shows you the route? But the cumulative effect is a progressive atrophy of the capacities that make us cognitive agents.

The neuroscience behind this concern is neuroplasticity, studied in Track 3 of this course. The brain strengthens the circuits it uses and weakens the ones it doesn't. If we stop exercising spatial memory (because GPS takes care of it), the hippocampus loses measurable volume. If we stop practicing argumentative writing (because AI generates texts), the circuits involved in organizing thought weaken. This is not a metaphor: it is synaptic reorganization.

The second dimension of risk is the erosion of cultural diversity. When billions of people use the same systems to generate text, images, music, and code, cultural production tends toward homogenization. Each large language model (LLM) was trained on a particular corpus, with particular biases, and optimized for particular statistical patterns. When that same model generates content for billions, it produces variations within a limited aesthetic space.

Bruzzo calls this the opposite of the "imaginary surplus." Human culture has historically produced more forms, styles, narratives, and modes of expression than any individual could consume. This abundance was the substrate of innovation: the unexpected, the strange, what made no immediate sense. When generation passes through an algorithmic funnel optimized for engagement and recognizable patterns, the tendency is to compress this diversity. Not eliminate it all at once, but gradually reduce it, like erosion.

The third dimension is the most directly political: the transfer of decision-making power to opaque systems. Decisions that affect real lives (granting credit, diagnosing health conditions, judicial sentencing, job selection, access to social benefits) are delegated to algorithms whose inner workings are not understandable even to their creators, much less to the people affected.

This opacity drastically reduces the possibility of challenging a decision. When a human denies a loan, the applicant can ask why and, in principle, argue against the decision. When an algorithm denies one, the answer is "the model said no." The technical explanation (a combination of weights across millions of parameters) cannot be translated into reasons that can be challenged. The result is an erosion of accountability: no one decides, so no one answers for it. The system makes the decision, but the system is not a moral agent.

If Bruzzo’s diagnosis is bleak, his proposal for resistance is surprising: what remains human is the useless. Not useless as failure, but useless as a philosophical category. Art that optimizes no metric. Love that maximizes no utility. Play with no productive purpose. Conversations that exist simply for their own sake. A walk with no destination. Contemplation with no output. Silence that does not need to be filled.

This proposal engages with a long philosophical tradition. Aristotle considered contemplation (theoria) the highest activity precisely because it was the least useful in a practical sense. Kant saw aesthetic experience as "purposiveness without purpose." Hannah Arendt distinguished labor (biological necessity), work (the fabrication of objects), and action (free activity among equals), with action being the properly human dimension precisely because it has no instrumental utility. Bruzzo updates this tradition for the algorithmic age: resisting the technological colonization of the imagination means radically valuing what cannot be reduced to data.

This module wraps up the course "Mente, Poder e Comportamento". We traced an arc that began with the mechanics of individual decision-making (Systems 1 and 2, heuristics, biases), moved through social influence (conformity, obedience, manipulation), explored the neural infrastructure (neuroplasticity, dopamine, amygdala), moved up to the economic systems that exploit these vulnerabilities (behavioral economics, nudge, mental accounting), questioned the philosophical foundations (free will, consciousness, semiotics), and reached the current frontier: what happens when machines inherit, amplify, and automate all of this.

Bruzzo’s answer, and this course’s final argument, is not a technical one. It is an ethical and existential answer. Being human in the age of AI does not mean competing with machines in efficiency, speed, or scale. It was never about that. It means insisting on the dimensions of experience that cannot be reduced to data, optimized by algorithms, or delegated to systems. Vulnerability. Ambiguity. Productive error. Useless beauty. Presence without purpose. Encountering another person who cannot be predicted, modeled, or replicated.