Rare Earths and Big Data: Who Is Mining Brazil?

Brazil is entering a dispute that is not only economic. It is a dispute over territory, data, attention, emotion, and human decision-making. On one side, we have rare earths, strategic minerals for wind turbines, electric cars, electronics, batteries, and defense systems. On the other, we have stochastic Big Data, capable of classifying people into groups and subgroups of consumption, emotion, desire, fear, and voting behavior.

The political question of this blog is simple: who is mining Brazil?

In rare-earth mining, physical territory is drilled, extracted, separated, refined, and transformed into industrial value. In data mining, the explored territory is different: behavior, attention, heartbeat, face, emotion, screen time, location, consumption, language, and decision-making.

Brazil has significant rare-earth reserves, but the national challenge is not only extraction. It is also processing, refining, technology, industrial policy, and production chains. Otherwise, Brazil risks repeating its historical role: exporting cheap raw material and importing expensive technology.

This is the first layer of the problem: Brazil may have wealth underground, but still remain poor in technological sovereignty. It exports minerals and imports dependence.

But there is a second, more invisible layer: Brazil is also being mined through data.

Shoshana Zuboff called this process surveillance capitalism: an economic logic that transforms human experience into raw material for extraction, prediction, and sale of future behavior. It is not only about “showing ads.” It is about producing predictability over what we will do now, soon, and later.

Here enters the idea of stochastic Big Data. Data mining does not need to know the intimate truth of each individual. It needs to calculate probabilities. It classifies people into groups and subgroups: who buys through fear, who buys through desire, who votes through anger, who reacts to humiliation, who stays longer in front of an image, who is vulnerable to urgency, who needs belonging, who easily enters Zone 3.

The individual thinks: “I accepted the terms.” But the problem is bigger. Individual consent does not solve collective mining. Even when one person authorizes data use, the model learns population patterns. The data of one person becomes inference about many. Mining stops being only personal and becomes territorial.

Brazil’s LGPD protects fundamental rights such as freedom, privacy, and the free development of personality. It defines personal data as information related to an identified or identifiable person, includes sensitive data such as health, genetic, and biometric data when linked to a natural person, and recognizes operations such as collection, classification, evaluation, processing, extraction, and profiling.

But today’s challenge goes beyond the classic text of privacy. The BrainLatam2026 question is: when behavioral, physiological, and affective data from a population are used to predict consumption, emotion, and voting behavior, is this still only personal data, or has it become the mining of cognitive territory?

Rosalind Picard, in the field of Affective Computing, showed that computers can recognize, interpret, express, and respond to emotions. This changes everything. If a platform can infer when a person is tired, irritated, lonely, excited, insecure, or seeking belonging, it can deliver exactly the emotion that keeps that person giving attention.

Advertising stops being only a message. It becomes a calculated affective environment.

In BrainLatam2026 language: advertising no longer speaks only to the cognitive mind. It begins to modulate the Tekoha — the individual’s internal territory — through stimuli that capture interoception, proprioception, desire, fear, and belonging.

And this does not require waiting for an invasive BCI or an EEG headset. A common camera can already begin to infer physiological signals.

The MIT CSAIL Video Magnification project shows that apparently static videos contain subtle changes invisible to the human eye, but extractable through algorithms, including human pulse, sound from vibrating objects, and the movement of hot air.

Another MIT work, Detecting Pulse from Head Motions in Video, showed that it is possible to extract heart rate and intervals between beats from micro-movements of the head caused by blood flow. In 18 participants, the method reported heart rates almost identical to ECG and also captured clinically relevant information about heart-rate variability.

This point is decisive: an ordinary camera can already function as an invisible BCI. It does not read thoughts, but it can read bodily signals that help infer autonomic state, arousal, fatigue, stress, or recovery. Alone, this does not say “the person wants to buy” or “the person will vote for someone.” But combined with clicks, face, gaze, history, location, time, and language, it becomes powerful behavioral ore.

In rare-earth mining, the crater appears in the territory. In data mining, the crater appears in attention, desire, fear, and decision-making.

That is why the comparison is politically strong: rare earths are the ore of the underground; stochastic Big Data is the ore of human decision-making.

In an election year, this becomes even more serious. The problem goes beyond false content. The deeper issue is the probabilistic modulation of attention and emotion before the person even realizes they are being guided.

Data mining can produce an algorithmic Zone 3. The individual is kept in fast reaction: fear, anger, urgency, comparison, envy, threat, artificial belonging. The body begins to attack, flee, freeze, or replicate patterns. It is “thinking fast” captured by the market and by propaganda.

In Zone 3, there is less criticality and less creativity to learn the new. The person does not decide from Fruition and Metacognition. They react to stimuli calibrated to reduce openness, increase defense, and shorten thought.

Here the debate stops being only technological. It becomes federative and political.

The Federal, State, and Municipal governments need to discuss legal instruments to recognize that data mining in a region generates economic value from the population of that territory. This must be done with legal care, compatible with the Constitution, LGPD, tax competencies, and fundamental rights. But the political principle is clear: if a company extracts value from the attention and behavior of a population, that population must receive protection, transparency, auditability, and social return.

This does not mean saying that any municipality can create any charge in any way. It means formulating a new public category: compensation for territorial data mining.

This compensation could fund digital literacy, mental health, protection of children and adolescents, algorithmic auditing, transparency in political advertising, public data laboratories, AI education, research in neurorights, and infrastructure for digital sovereignty.

The logic is similar to physical mining: if there is value extraction from a territory, there is local impact. In mineral mining, the impact appears in water, soil, health, cities, violence, inequality, and economic dependence. In data mining, the impact appears in attention, mental health, political decision-making, consumption, polarization, anxiety, screen addiction, and loss of autonomy.

BrainLatam2026 would call this territorial neurodigital sovereignty. Tekoha is not only forest, city, home, or neighborhood. Tekoha is also the person’s internal territory, now captured by cameras, sensors, platforms, ads, predictive models, and recommendation systems.

The central question for Brazil is:

will we allow companies to extract attention, emotion, heartbeat, behavior, and decision-making as if this were not national wealth?

The bridge with BCI and neurorights is direct. If today it is already possible to infer physiological signals through video, tomorrow EEG, fNIRS, eye-tracking, GSR, HRV, wearables, and BCIs may make this mining even deeper. The brain and body become sources of data for authentication, marketing, security, health, education, and politics.

For this reason, every physiological datum inferred from the body — even through an ordinary camera — should be treated as sensitive data of cognitive sovereignty. It does not matter whether it comes from EEG, fNIRS, smartwatch, camera, microphone, mouse, keyboard, or video. If the data helps infer attention, emotion, fatigue, arousal, vulnerability, or decision-making, it belongs to the field of neurorights.

The decolonial critique is simple: Brazil has already been mined for gold, wood, sugar, coffee, iron, oil, and now rare earths. We cannot allow the next colonial cycle to be the invisible mining of the Brazilian mind.

The bridge with DREX Cidadão appears as a proposal for metabolic inversion. If companies extract value from the attention and data of a population, part of this value should return to the citizen as public infrastructure, education, health, digital protection, and citizen income. The digital money of the future cannot be born only in banks, platforms, and behavioral prediction markets. It needs to nourish the citizen.

DREX Cidadão, in this context, is not only income. It is a policy of belonging. It is the State saying: the wealth produced by the social body must return to the social body.

Brazil needs to discuss rare earths with industrial sovereignty. But it also needs to discuss Big Data with cognitive sovereignty. One without the other will be incomplete. We may protect the underground and lose the mind. We may nationalize minerals and outsource decision-making.

Closing
Rare earths and Big Data are two faces of the same century. One mines the ground. The other mines behavior. One extracts from physical territory. The other extracts from digital, affective, and bodily Tekoha. In an election year, this discussion must leave laboratories and enter public policy. Because democracy is not only voting. Democracy is deciding with criticality, creativity, and belonging. If human decision becomes ore, Brazilian sovereignty must also defend the mind, the body, and the attention of its people.

References

Agência Brasil. (2026). Terras raras, minerais estratégicos e críticos: entenda as diferenças.

Zuboff, S. (2019). The Age of Surveillance Capitalism. PublicAffairs.

Picard, R. W. (1997). Affective Computing. MIT Press.

Balakrishnan, G., Durand, F., & Guttag, J. (2013). Detecting Pulse from Head Motions in Video. IEEE Conference on Computer Vision and Pattern Recognition.

MIT CSAIL. Video Magnification: Finding the Visible in the Invisible.

Brasil. Lei nº 13.709/2018 — Lei Geral de Proteção de Dados Pessoais.

Tribunal Superior Eleitoral. Resolutions and regulations on electoral advertising, AI, and disinformation.