La maquinaria del desempleo: Sobre el capitalismo de la IA y el reemplazo de la fuerza laboral humana
Por Juan Pablo Torres Muñiz
En nuestra época, la efervescencia tecnológica impulsada por la, así llamada, Inteligencia Artificial (IA) y la computación cuántica ha empujado a la humanidad a un umbral que, si bien se había anticipado en la ciencia ficción, ha irrumpido en la realidad con una rapidez vertiginosa e implicaciones políticas, sociales y laborales que requieren una reflexión urgente bajo el microscopio del Homo Institutionalis.
Tradicionalmente, el imaginario colectivo occidental ha oscilado entre la fascinación y el miedo ante las máquinas. Por un lado, la IA se presenta como un logro increíble que promete revolucionar los servicios sanitarios y ofrecer soluciones a problemas sociales. El éxito de programas como AlphaGo, que en 2016 derrotó al campeón de Go Lee Sedol, demostró que la IA podía utilizar la intuición y el pensamiento estratégico, asestando un nuevo golpe a la autoimagen de la humanidad. Estos avances, dignos de ciencia ficción, alimentan la esperanza de alcanzar la superinteligencia o, en la visión transhumanista, el Homo deus.
Sin embargo, esta exageración sobre la IA y sus miedos generó narrativas que nos alertaban de que «las máquinas tomarán el control», convirtiendo a los humanos en esclavos o en una mera fuente de energía (como en la película Matrix). El miedo más grande y persistente es el «complejo de Frankenstein»: la pérdida de control ante la criatura artificial.
Frente a esta aprensión existencial, surgió una calma mediática que a menudo desestimaba los riesgos reales y presentes. Se nos dijo que la Inteligencia Artificial General (IAG), capaz de superar la inteligencia humana, no se alcanzaría en varias décadas, si es que alguna vez era posible. Esta visión, que relegaba el reemplazo laboral a un futuro lejano, permitió que los desafíos éticos se centraran en dilemas hipotéticos (como el del coche autónomo y el dilema del tranvía).
No obstante, la realidad ha superado la predicción, al menos parcialmente. Lo que se ha generalizado es la IA estrecha o débil, invisiblemente integrada en nuestras herramientas cotidianas, desde el marketing hasta las finanzas y la fabricación. Esta IA, combinada con el crecimiento exponencial de la potencia de los ordenadores, ha permitido a los algoritmos hacerse cargo de muchas de nuestras actividades, incluyendo planificación, reconocimiento facial y toma de decisiones. Las tecnologías antaño descritas en la ciencia ficción ahora se transforman en el mundo real.
El impacto más tangible y urgente de esta aceleración se observa en el futuro del trabajo. La robótica y la IA han inaugurado la segunda edad de la máquina, en la que las máquinas ya no son solo un complemento, sino que sustituyen a los humanos. Las soluciones del siglo pasado «ya no son válidas».
La confirmación más brutal y sintomática de esta transferencia de capacidad operativa a la máquina se encuentra en la estrategia de Amazon: la empresa, que ya utiliza la IA para rastrear y gestionar a sus empleados, ha revelado un plan para reemplazar masivamente a 600,000 trabajadores humanos por robots. Este cálculo de eficiencia, donde el valor humano se reduce a una fracción de coste por artículo vendido, evidencia que el riesgo ya no es el monstruo de ciencia ficción que destruye la civilización, sino una IA estrecha que, con su capacidad de automatizar procesos, amenaza la libertad, la autonomía y el significado del trabajo para millones de personas.
El IA ha pasado a ser el objeto de la lucha (geo)política, obligándonos a cuestionar urgentemente si esta tecnología debilitará la democracia y conducirá a nuevas formas de totalitarismo digital o si, por el contrario, la humanidad sabrá moldear activamente estas mediaciones para garantizar que sus beneficios sirvan al bien común.
Desde la perspectiva del Homo Institutionalis, la vertiginosa aceleración de la automatización y el consecuente reemplazo de la fuerza laboral humana por la Inteligencia Artificial (IA) y la robótica constituyen manifestaciones de una crisis ontológica y política de primer orden, exacerbada por una narrativa de exageración que disuadió a la sociedad de los riesgos inminentes, mientras el capital globalista construía silenciosamente la infraestructura de la dominación.
La IA no es políticamente neutral, para nada, y exige un análisis, al menos, bajo tres ejes interconectados: el de la explotación capitalista, el de poder y subjetividad y el institucionalista, referido a la pérdida de la capacidad operativa y de la dignidad humana.
La lógica explotadora del capitalismo de la IA:
La robótica y la IA han marcado el inicio de la segunda edad de la máquina (Brynjolfsson y McAfee, 2014), en la que las máquinas no solo complementan, sino que sustituyen a los humanos. Las predicciones son sombrías: algunos informes estiman que hasta el 47% de los trabajos en Estados Unidos son susceptibles de automatización, con el Foro Económico Mundial prediciendo la pérdida de 85 millones de empleos globales, tan sólo para finales de 2025.
La noticia de que Amazon calcula la eficiencia de reemplazar 600,000 empleados por robots a solo 30 centavos por artículo vendido es el epítome de la crítica marxista, desde cuya perspectiva, el problema de la automatización reside en su aplicación bajo el capitalismo, donde las máquinas sirven para incrementar la plusvalía para los capitalistas.
¿Consecuencias?
– Alienación y proletariado digital: La robótica contribuye a la alienación de los trabajadores, quienes se convierten en una mera parte de la máquina, perdiendo la oportunidad de hacerse de una vida digna a través del trabajo. El uso de robots para reemplazar trabajadores humanos crea una clase de personas desempleadas (el proletariado), lo que, a su vez, hace decrecer los salarios de quienes conservan su empleo.
– Tecnocapitalismo y desigualdad: La IA es vista como la culminación de la alienación y un instrumento del capital. Esta lógica conduce a la mayor desigualdad salarial y a una sociedad más dividida. El capitalismo de la IA garantiza que los beneficios se concentren en unos pocos (los dueños de la tecnología y los robots), mientras que el resto enfrenta la precariedad y la vulnerabilidad socioeconómica y psicológica.
– Explotación oculta y colonialismo de datos: La automatización no solo explota a los trabajadores visibles. El modelo se apoya en el capitalismo de la vigilancia (Zuboff), donde los usuarios se convierten en trabajadores que producen datos de forma gratuita (data laborers) para las grandes corporaciones. Además, la infraestructura de IA depende del duro trabajo de aquellos que extraen minerales y de los trabajadores mal pagados del hemisferio sur que limpian y etiquetan los datos para entrenar los algoritmos.
La tragedia del amo y la infantilización algorítmica:
El impacto de la IA sobre el trabajo humano trasciende la mera cifra de desempleo; afecta la esencia del Homo Institutionalis, su capacidad para ser un sujeto autónomo, racional y responsable.
– La ruptura de la libertad y la razón: La IA implementada en sistemas como el de Amazon para despedir trabajadores (algoritmos de «robo-firing») plantea un problema de justicia y libertad negativa, pues el proceso de decisión ha sido automatizado, sin juicio humano ni transparencia. La toma de decisiones por algoritmos (como el caso COMPAS en juzgados) reproduce e incluso aumenta el sesgo humano histórico y la discriminación contra grupos específicos.
– La dialéctica amo-esclavo invertida: Si la automatización reemplaza a los esclavos humanos por máquinas, se mantiene una estructura social basada en la esclavitud, aunque sea artificial. El temor «hegeliano» es que, si todos los humanos se convierten en amos de las máquinas, se volverían dependientes de ellas y consumidores explotados, perdiendo la oportunidad de reconocimiento por parte de las máquinas, ya que estas carecen de autoconciencia. Esta dependencia tecnológica nos hace vulnerables.
– La servidumbre voluntaria: La IA se utiliza para manipular el comportamiento y las elecciones humanas de forma sutil, a través de empujones (nudges) e hiperempujones (hypernudges). Al moldear nuestros entornos de elección sin nuestro conocimiento consciente, esta manipulación dirigida por datos socava la libertad entendida como autonomía. La IA actúa como una nueva forma de gobernanza algorítmica, donde los humanos son infantilizados y se vuelven menos capaces de pensar o decidir por sí mismos.
La destrucción del sentido y la necesidad de institucionalidad, como conocimiento en sí mismo (acaso, sabiduría):
El desplazamiento laboral provocado por la IA ataca el significado del trabajo y, por extensión, la identidad de la persona.
Veamos: la preocupación por la destrucción de empleo presupone que el trabajo es una fuente de valor y sentido. El trabajo tiene valor, ya que otorga al trabajador un propósito y un sentido, además de vínculos sociales y oportunidades de responsabilidad. La proliferación de trabajos terribles y la precariedad laboral, que afecta especialmente a jóvenes sobrecalificados, generan una ansiedad y estrés que son consecuencias psicológicas del capitalismo de la precariedad.
Por otra parte, la IA y las tecnologías de autocuidado (como Fitbit, cuyo uso de datos ha llevado a acusaciones en casos criminales) promueven la autodisciplina y la autoexplotación. El individuo internaliza el imperativo de rendir hasta el agotamiento, convirtiéndose en un sujeto-logro que se explota a sí mismo. La cuantificación del yo mediante datos sugiere erróneamente que el yo digital es el real, o que el yo puede ser reducido a una recopilación de información digital. El sujeto humano es considerado superfluo e incluso un estorbo para la automatización fluida.
Lo cierto es que la IA actual es una IA débil o estrecha, que se desempeña bien en tareas específicas, pero carece de entendimiento, experiencia, sensibilidad y sabiduría. Esta se centra en el pensamiento abstracto de tipo cognitivo, pero le es imposible delegar la sabiduría pragmática concreta, que surge de la experiencia corporeizada y situacional. Por lo tanto, el problema antropológico no es tecnológico: la IA es una estructura racional pletórica, fría, eficiente, totalitaria en su apariencia de neutralidad, pero no puede hacer arte ni pensar, pues carece de la crítica racional institucionalizada propia del Homo Institutionalis.
[Entonces…]
La apabullante irrupción de la IA actual ha demostrado a las claras el carácter ilusorio de la reducción del hombre a una inteligencia meramente computacional. El desarrollo, impulsado por el crecimiento exponencial de la potencia de los ordenadores y la disponibilidad masiva de big data, ha superado las predicciones, llevando a la economía a lo que se denomina la segunda edad de la máquina, donde éstas ya no son solo un complemento, sino que sustituyen a los humanos.
El caso de Amazon ilustra la frialdad racional del horizonte de sustitución: la compañía no solo apuesta por robots para la automatización de sus almacenes, sino que ha calculado la rentabilidad de prescindir de 600,000 empleados humanos, cifrando el coste del reemplazo en apenas 30 centavos por artículo vendido. Esta es la demostración de que la IA y la robótica, aunque no tengan conciencia, son simplemente eficientes y eficaces, eliminando la necesidad de un número abultado de trabajadores.
La velocidad del cambio es la nueva ironía del progreso…
La IA se fundó en parte en la rebelión contra las limitaciones de campos existentes, pero los avances iniciales generaron un entusiasmo descarado seguido de «inviernos» de IA. Sin embargo, la capacidad de las máquinas para pensar, aprender y crear aumenta rápidamente.
El avance en el aprendizaje automático (machine learning) ha permitido a los algoritmos hacerse cargo de muchas actividades que antes se consideraban exclusivamente humanas. Los sistemas cuánticos (Computación Cuántica) prometen acelerar este proceso exponencialmente, superando el límite físico del silicio (Ley de Moore) y llevando la IA a una escala «digna de la más entusiasta ciencia ficción».
El problema central no es el mero desempleo, sino cómo esta automatización totalitaria socava las estructuras que constituyen al Homo Institutionalis: la autonomía, la razón y la responsabilidad.
En efecto, la IA y las tecnologías de automatización son instrumentos de poder sin conocimiento y, por ende, sin responsabilidad. Actúan como mecanismos de control social, mediante la manipulación sutil de las decisiones y son capaces de analizar nuestros datos para predecir y manipular nuestro comportamiento con relativa facilidad. Esta influencia encubierta amenaza la libertad positiva (autonomía), sin respetarlas como seres racionales capaces de establecer sus propias metas.
Como bien sabemos, el mercado pletórico impulsa la adolescentización de la gente, un proceso en el que se someten los impulsos a la reflexión, y la imagen se antepone a las ideas. El Homo Institutionalis es reducido a la vaga y confusa abstracción de «ser humano», apenas distinguible de los animales, apelado mediante estímulos primitivos. Esto es la despersonalización masiva. Si el hombre no actúa, corre el riesgo de volverse indefenso y de perder las muchas oportunidades ofrecidas por la IA.
Al delegar tareas cognitivas y de juicio a la IA, se crean sistemas que operan de forma opaca e irresponsable. Las IA pueden tomar decisiones con consecuencias éticas (como en el sistema judicial COMPAS), pero al carecer de conciencia, libre albedrío o emociones, no pueden ser agentes morales responsables. En sistemas complejos de aprendizaje profundo (deep learning), la forma concreta en que la IA llega a una decisión ya no es transparente ni explicable por completo. Esto genera una brecha de responsabilidad y conocimiento, lo que implica que aquellos afectados por las decisiones algorítmicas (jueces, bancos, gobiernos) no pueden obtener una explicación de las razones. La IA está en manos de gente que no conoce los riesgos éticos o tiene expectativas equivocadas.
Por si fuera poco, los algoritmos se nutren de datos históricos, lo que provoca la perpetuación y amplificación del sesgo social existente (racismo, sexismo y otras formas de discriminación injustificadas). El sesgo no es intencionado, sino una consecuencia no deseada de la tecnología, que se convierte en «opresión algorítmica».
La automatización consolida el poder en manos de unas pocas mega corporaciones. La acumulación de poder digital en manos de súper corporaciones es problemática. Estas empresas monopolizan los datos y los flujos de información, ejerciendo un poder que amenaza toda forma de democracia ajena a sus intereses. De hecho, la IA facilita nuevas formas de totalitarismo, ya sea patrocinado por el Estado (como el sistema de crédito social chino) o por las grandes corporaciones. Este totalitarismo corporativo utiliza la gestión burocrática y el seguimiento constante de las personas como cosas. El mal reside, entonces, en la irreflexión y en seguir órdenes ciegamente, algo que la IA, al ofrecer «rankings de riesgo empíricos» sin cuestionamiento, fomenta activamente.
El nuevo horizonte de la precariedad global: la urgencia de la reflexión y el destino del Homo Institutionalis:
El fenómeno de los despidos masivos orquestados motivados por la implementación de la robótica y la IA debe ser analizado, no como un evento corporativo aislado, sino como la proyección material y acelerada de una lógica de explotación capitalista que impacta la institucionalidad de la persona y amenaza la estabilidad social a escala global.
La noticia de la que parte la reflexión de este texto, revela el cinismo inherente del turbo-capitalismo. La IA de Amazon opera como un instrumento del capital, intensificando la explotación y la concentración de poder. Los algoritmos de IA utilizados para la «robo-firing» (despidos automáticos) ilustran la delegación de decisiones humanas a máquinas, más que irresponsables, «arresponsables», carentes de la capacidad de agencia moral, conciencia, libre albedrío o emociones. Esto plantea un problema de responsabilidad: ¿quién es responsable cuando un algoritmo comete un error, o en este caso, implementa una decisión que causa un perjuicio masivo? La complejidad del sistema y la opacidad de los algoritmos de «caja negra» dificultan la trazabilidad de la responsabilidad, dejando a los humanos afectados sin la posibilidad de una explicación o rendición de cuentas.
El trabajador, bajo este sistema, es monitoreado y analizado, perdiendo su libertad negativa (no interferencia) y su privacidad. La tecnología se convierte en el famoso panóptico, pero digital, donde los datos de los empleados, incluso aquellos que revelan bajo rendimiento o fraude, se utilizan para la gestión burocrática y el castigo.
La crisis de Amazon es un presagio de los efectos sistémicos que vienen, cuyo impacto no se limita a los «trabajos de cuello azul» (trabajo manual), sino que afecta a tareas cognitivas y servicios, incluyendo las finanzas, el derecho y el periodismo.
Actualmente, ya, el ejemplo de Amazon se replica en el sector servicios, donde los robots-camareros y los camiones autónomos amenazan a los trabajadores con bajos salarios, incrementando la precariedad para los jóvenes y la incapacidad de reubicación para los mayores de 50 años. La automatización en la producción (fábricas sin empleados, como en China) y el control algorítmico de la cadena de suministro conducen a la concentración de poder en grandes corporaciones. Esto se ve agravado por los avances en computación cuántica, que permitirán a las empresas optimizar y predecir los mercados con una precisión inalcanzable para la competencia. El ganador de esta carrera tendrá en sus manos moldear el futuro, a menudo a costa de la competencia y la libertad económica.
El uso de IA y big data aumenta las asimetrías de poder y las desigualdades. La sustitución laboral intensiva crea una «segunda edad de la máquina» que beneficia a los propietarios de la tecnología, mientras que el resto de la población se enfrenta a la devaluación del trabajo y la precariedad. Esta brecha salarial y la división social amenazan la democracia. Si, como se proyecta, el 80% de la población resulta inútil para el nuevo modelo económico, la persona se convierte en una «molestia» que consume recursos sin ser rentable. Esta idea conlleva un riesgo existencial para la dignidad, fomentando ideologías que buscan la reducción de la natalidad y la sumisión o eliminación de la población «sobrante».
Al acelerarse el desempleo masivo, se incrementa el descontento social y abona el terreno para la radicalización y el posible colapso de la democracia.
La pérdida de propósito vital que conlleva la desaparición del trabajo significativo reduce a los ciudadanos a lo que, en su momento, fue llamado «hombres-masa» que buscan refugio en el consumo desaforado y el entretenimiento masivo. Esta infantilización sistemática fomenta la irreflexión y la obediencia, condiciones ideales para el totalitarismo digital.
El ritmo vertiginoso del cambio tecnológico supera la capacidad de los líderes políticos, que lo fían todo a una tecnología que no comprenden. La IA, al ofrecer nuevas posibilidades para el gobierno de expertos (tecnocracia), desplaza la deliberación democrática. La IA es esencialmente antidemocrática cuando centraliza el control y el conocimiento en manos de una élite tecnocrática que, ayudada por los algoritmos, impone decisiones sin transparencia ni responsabilidad. Feudalismo tecnológico…
Vigilancia de masas. Control totalitario. De hecho, ahora mismo, las potencias tecnológicas exportan el autoritarismo, debilitando globalmente los principios de libertad y tolerancia. Los despidos en Amazon son, en última instancia, una estrategia de guerra en la contienda global por el control total y la dominación de los recursos y la población.
En definitiva, la IA, en el contexto capitalista actual, no es un mero cambio de ciclo, ni mucho menos…
Referencias:
Agamben, G. (1998/2010). Homo sacer: El poder soberano y la nuda vida. Pre-textos.
Alaimo, S. (2016). Exposed: Environmental Politics and Pleasures in Posthuman Times. University of Minnesota Press.
Alpaydin, E. (2016). Machine Learning. MIT Press.
Arendt, H. (1958/2005). La condición humana. Paidós Ibérica.
Arendt, H. (2017/2006). Los orígenes del totalitarismo. Alianza.
Aristóteles. (2002). Obras completas. Gredos.
Asimov, I. (1942). «Runaround: A Short Story». Astounding Science-Fiction, 29(1).
Atanasoski, N., & Vora, K. (2019). Surrogate Humanity: Race, Robots, and the Politics of Technological Futures. Duke University Press.
Awad, E., Dsouza, S., Kim, R., Schulz, J., Henrich, J., Shariff, A., Bonnefon, J.-F., & Rahwan, I. (2018). «The Moral Machine Experiment». Nature, 563, 59-64.
Baños, P. (2022). La encrucijada mundial: Un manual del mañana. Ariel.
Benjamin, R. (2019a). Race After Technology: Abolitionist Tools for the New Jim Code. Polity Press.
Benjamin, R. (2019b). Captivating Technology: Race, Carceral Technoscience, and Liberatory Imagination in Everyday Life. Duke University Press.
Bostrom, N. (2014/2016). Superinteligencia. Caminos, peligros, estrategias. Teell.
Bourdieu, P. (1990/1991). El sentido práctico. Taurus.
Bozdag, E. (2013). «Bias in Algorithmic Filtering and Personalization». Ethics and Information Technology, 15(3), 209-227.
Braidotti, R. (2020). «“We” Are in This Together, but We Are Not One and the Same». Journal of Bioethical Inquiry, 17(4), 465-469.
Brodie, R. (1995). Virus of the Mind: The New Science of the Meme. Integral Press.
Brynjolfsson, E., & McAfee, A. (2014). The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies. W. W. Norton and Company.
Brynjolfsson, E., & McAfee, A. (2017). Machine, Platform, Crowd: Harnessing Our Digital Future. W. W. Norton.
Cadwalladr, C., & Graham-Harrison, E. (2018). «Revealed: 50 Million Facebook Profiles Harvested for Cambridge Analytica in Major Data Breach». Guardian.
Canavan, G. (2015). «Capital as Artificial Intelligence». Journal of American Studies, 49(4), 685-709.
Castaño Díaz, C. M. (2013). «Defining and characterizing the concept of Internet Meme». Revista CES Psicología, 6(1), 82-104.
Castells, M. (2001/2007). La galaxia internet. Plaza & Janés.
Cheney-Lippold, J. (2017). We Are Data: Algorithms and the Making of Our Digital Selves. New York University Press.
Clark, R. (2011). «Asimov’s Laws of Robotics: Implications for Information Technology». En M. Anderson & S. Anderson (Eds.), Machine Ethics (pp. 254-284). Cambridge University Press.
Coeckelbergh, M. (2012). Growing Moral Relations: Critique of Moral Status Ascription. Palgrave Macmillan.
Coeckelbergh, M. (2015a). «The Tragedy of the Master: Automation, Vulnerability, and Distance». Ethics and Information Technology, 17(3), 219-229.
Coeckelbergh, M. (2020). AI Ethics (L. Álvarez Canga, Trad.). MIT Press. (Original: AI Ethics, 2021).
Coeckelbergh, M. (2022). La ética de los robots (L. Álvarez Canga, Trad.). Titivillus. (Original: Robot Ethics, 2022).
Coeckelbergh, M. (2022a). The Political Philosophy of AI: An Introduction. Polity Press.
Coeckelbergh, M. (2023). Filosofía política de la inteligencia artificial (L. Álvarez Canga, Trad.). Titivillus. (Original: The political philosophy of AI: an introduction, 2023).
Coeckelbergh, M. (2024). Por qué la IA debilita la democracia y qué hacer al respecto (L. Álvarez Canga, Trad.). Titivillus. (Original: Why AI Undermines Democracy and What To Do About It, 2024).
Coeckelbergh, M. (2020). «AI, Responsibility Attribution, and a Relational Justification of Explainability». Science and Engineering Ethics, 26(4), 2051-2068.
Coeckelbergh, M. (2021). Green Leviathan or the Poetics of Political Liberty: Navigating Freedom in the Age of Climate Change and Artificial Intelligence. Routledge.
Coeckelbergh, M. (2021). Ética de la inteligencia artificial. Cátedra.
Cook, G., Lee, J., Tsai, T., Kong, A., Deans, J., Johnson, B., & Jardin, E. (2017). Clicking Clean: Who Is Winning the Race to Build a Green Internet? Greenpeace.
Couldry, N., & Mejias, U. A. (2019). The Costs of Connection: How Data Is Colonizing Human Life and Appropriating It for Capitalism. Stanford University Press.
Crawford, K. (2021). Atlas of AI: Power, Politics, and the Planetary Costs of Artificial Intelligence. Yale University Press.
Crawford, K., & Calo, R. (2016). «There Is a Blind Spot in AI Research». Nature, 538, 311-313.
Criado Perez, C. (2019). Invisible Women: Data Bias in a World Designed for Men. Abrams Press.
Crary, J. (2014/2015). 24/7: capitalismo tardío y el fin del sueño. Ariel.
Dawkins, R. (1976/2002). El gen egoísta: las bases biológicas de nuestra conducta. Salvat.
Dennett, D. C. (1995). Darwin’s Dangerous Idea. Penguin Books.
Dreyfus, H. L. (1972). What Computers Can’t Do. Harper Collins.
Dyer-Witheford, N., Kjøsen, A. M., & Steinhoff, J. (2019). Inhuman Power: Artificial Intelligence and the Future of Capitalism. Pluto Press.
Eubanks, V. (2018). Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor. St. Martin’s Press.
Floridi, L. (2014). The Fourth Revolution: How the Infosphere Is Reshaping Human Reality. Oxford University Press.
Floridi, L., & Sanders, J. W. (2004). «On the Morality of Artificial Agents». Minds & Machines, 14(3), 349-379.
Foucault, M. (1977/2012). Vigilar y castigar. Siglo XXI.
Foucault, M. (1980). Power/Knowledge: Selected Interviews and Other Writings 1972-1977 (C. Gordon, Ed.). Pantheon Books.
Frey, C. B., & Osborne, M. A. (2013). «The Future of Employment: How Susceptible Are Jobs to Computerisation?». Oxford Martin School Working Paper, 7.
Fuchs, C. (2014). Digital Labour and Karl Marx. Routledge.
Fuchs, C. (2023). Digital Democracy and the Digital Public Sphere. Routledge.
Furedi, F. (2017). «The Medicalisation of the University». Policy, 33(4), 48-51.
Graeber, D. (2018). Trabajos de mierda. Ariel.
Gunkel, D. (2018). Robot Rights. MIT Press.
Haraway, D. (2000). «A Cyborg Manifesto». En D. Bell y B. M. Kennedy (Eds.), The Cybercultures Reader (pp. 291-324). Routledge.
Harvey, D. (2019). Marx, Capital and the Madness of Economic Reason. Profile Books.
Hegel, G. W. F. (1977/2006). Fenomenología del espíritu. Pre-textos.
Heidegger, M. (1977). The Question Concerning Technology and Other Essays. Garland Publishing.
Hobbes, T. (1996/2003). Leviatán. Losada.
Hofstadter, D. R., & Dennett, D. C. (1981/2000). The Mind’s I. Basic Books.
Hughes, J. (2004). Citizen Cyborg: Why Democratic Societies Must Respond to the Redesigned Human of the Future. Westview Press.
Kaku, M. (s.f.). Supremacía cuantica: La revolucion tecnologica que lo cambiara todo. (Mencionado).
Kant, I. (2013). Lecciones de ética. Planeta.
Kemp, S. (2023). «Facebook Statistics and Trends». Datareportal.
Kirkpatrick, S., Gelatt, C. D., & Vecchi, M. P. (1983). «Optimization by Simulated Annealing». Science, 220(4598).
Kurzweil, R. (2005/2021). La singularidad está cerca. Lola Books.
Labatut, B. (2023). MANIAC. Titivillus.
Latour, B. (1993/2007). Nunca fuimos modernos. Siglo XXI.
Lazzarato, M. (2014). Signs and Machines: Capitalism and the Production of Subjectivity (J. D. Jordan, Trad.). Semiotext(e).
Lynch, A. (1996). Thought Contagion: How Belief Spreads Through Society The New Science of Memes. Ney York: Basic Books.
Malthus, T. R. (1798). An Essay on the Principle of Population. J. Johnson.
Marcuse, H. (2002). One-Dimensional Man: Studies in the Ideology of Advanced Industrial Society. Routledge.
Marx, K. (1990/2007). El capital (Vol. 1). Siglo XXI.
Massumi, B. (2014). What Animals Teach Us about Politics. Duke University Press.
Matthias, A. (2004). «The Responsibility Gap: Ascribing Responsibility for the Actions of Learning Automata». Ethics and Information Technology, 6(3), 175-183.
McAfee, A., & Brynjolfsson, E. (2017). Machine, Platform, Crowd: Harnessing Our Digital Future. W. W. Norton.
McQuillan, D. (2019). «The Political Affinities of AI». En A. Sudmann (Ed.), The Democratization of Artificial Intelligence (pp. 163-173). Transcript.
Mill, J. S. (1978/2008). Sobre la libertad. Tecnos.
Moore, P. (2018). The Quantified Self in Precarity: Work, Technology and What Counts. Routledge.
Moravec, H. (1988). Mind Children: The Future of Robot and Human Intelligence. Harvard University Press.
Mori, M. (2012). «The Uncanny Valley» (K. F. MacDorman & N. Kageki, Trans.). IEEE Robotics and Automation Magazine, 19(2), 98-100.
Mouffe, C. (2013). Agonistics: Thinking the World Politically. Verso.
Noble, S. U. (2018). Algorithms of Oppression: How Search Engines Reinforce Racism. New York University Press.
Norvig, P., & Russell, S. (2008). Inteligencia Artificial. Un enfoque moderno. (Original: Artificial Intelligence: A Modern Approach, 2008).
Nyholm, S. (2020). Humans and Robots: Ethics, Agency, and Anthropocentrism. Rowman and Littlefield.
O’Neil, C. (2016/2018). Armas de destrucción matemática. Capitán Swing.
Pariser, E. (2011/2017). El filtro burbuja. Taurus.
Pettit, P. (1997/1999). Republicanismo: una teoría sobre la libertad y el gobierno. Paidós Ibérica.
Platón. (1997). Republic. Hackett Publishing.
Popper, K. (1995/2010). La sociedad abierta y sus enemigos (Vol. 1). Paidós Ibérica.
Rhee, J. (2018). The Robotic Imaginary: The Human and the Price of Dehumanized Labor. University of Minnesota Press.
Rousseau, J. J. (1997/2007). El contrato social o principios de derecho político. Tecnos.
Russell, B. (1912). The Principles of Mathematics. (Implícito en Labatut).
Russell, S., & Norvig, P. (2008). Inteligencia Artificial. Un enfoque moderno (L. Joyanes Aguilar et al., Trad.). Titivillus.
Sætra, H. S. (2020). «A Shallow Defence of a Technocracy of Artificial Intelligence, Examining the Political Harms of Algorithmic Governance in the Domain of Government». Technology in Society, 62.
Schwab, K. (2016). The Fourth Industrial Revolution. Crown Publishing Group.
Schwab, K., & Malleret, T. (2020). Covid-19: el gran reinicio. Foro Económico Mundial.
Shanahan, M. (2015). The Technological Singularity. The MIT Press.
Shannon, C. E. (1948). «A Mathematical Theory of Communication». Bell System Technical Journal, 21.
Skinner, B. F. (1953). Science and Human Behavior. Macmillan. (Citado en Russell & Norvig).
Sperber, D. (2003). «An objection to the memetic approach to culture». En R. Aunger, Darwinizing Culture (pp. 163-174). Oxford University.
Sperber, D. (2005). Explicar la cultura. Un enfoque naturalista. Morata.
Spencer, H. (1863). First Principles. Williams and Norgate.
Stiglitz, J. (2019). (Citado en Coeckelbergh, M.).
Thaler, R. H., & Sunstein, C. R. (2009/2017). Un pequeño empujón (nudge). Taurus.
Torres Muñiz, J. P. (s.f.). Homo Institutionalis 2022. Grupo Editorial Caja Negra.
Turing, A. (1950). «Computing Machinery and Intelligence». Mind, 59, 433-460.
Turkle, S. (2011). Alone Together: Why We Expect More from Technology and Less from Each Other. Basic Books.
Turner, J. (2019). Robot Rules: Regulating Artificial Intelligence. Palgrave Macmillan.
Varela, F., Thompson, E. T., & Rosch, E. (1991). The Embodied Mind: Cognitive Science and Human Experience. MIT Press.
Véliz, C. (2020/2021). Privacidad es poder: datos, vigilancia y libertad en la era digital. Debate.
Vélez Herrera, J. I. (2007). Evolución en la tecnología: De la inteligencia artificial al meme. XcUiDi.
Vigen, T. (2015). Spurious Correlations. Hachette Books.
Von Neumann, J., & Morgenstern, O. (1944). Theory of Games and Economic Behavior. Princeton University Press.
Wallach, W., & Allen, C. (2009). Moral Machines: Teaching Robots Right from Wrong. Oxford University Press.
Wiener, N. (1948/1985). Cibernética: o el control y comunicación en animales y máquinas. Tusquets editores.
Winner, L. (1986). The Whale and the Reactor. University of Chicago Press.
Wolfe, C. (2010). What Is Posthumanism?. University of Minnesota Press.
Zuboff, S. (2019/2020). La era del capitalismo de la vigilancia. Paidós.
Fuentes periodísticas Web y organizacionales (citadas contextualmente):
3DJuegos. (s.f.). Plan secreto de Amazon toma forma: 600.000 robots y miles de millones de euros marcan su futuro.
BBC. (s.f.). (Referencia a despidos de Amazon/Foxconn y el reemplazo por robots).
El Economista. (s.f.). Amazon apuesta por robots y anuncia 600.000 despidos.
Feldstein, S. (2019a). «The Global Expansion of AI Surveillance». Carnegie Endowment for International Peace.
Gizmodo. (s.f.). Una empresa gigante dio el primer paso y ya prepara despidos masivos para reemplazar a empleados humanos por la automatización de la inteligencia artificial.
Xataka. (s.f.). Amazon ha calculado cuánto cuesta prescindir de 600.000 empleados: 30 centavos por artículo vendido y muchos robots.
ENGLISH VERSION
The Machinery of Unemployment and Depersonalization: On AI Capitalism and the Replacement of the Human Workforce
Translation by Rebeca Sanz
In our time, the technological fervor driven by so-called Artificial Intelligence (AI) and quantum computing has pushed humanity to a threshold that, while anticipated in science fiction, has burst into reality with dizzying speed and political, social, and labor implications that require urgent reflection under the microscope of Homo Institutionalis.
Traditionally, the Western collective imagination has oscillated between fascination and fear of machines. On the one hand, AI is presented as an incredible achievement that promises to revolutionize healthcare services and offer solutions to social problems. The success of programs like AlphaGo, which in 2016 defeated the Go champion Lee Sedol, demonstrated that AI could use intuition and strategic thinking, dealing a new blow to humanity’s self-image. These advances, worthy of science fiction, fuel the hope of achieving superintelligence or, in the transhumanist vision, Homo deus.
However, this hype about AI and its fears generated narratives warning us that «the machines will take control,» turning humans into slaves or a mere source of energy (as in the movie The Matrix). The greatest and most persistent fear is the «Frankenstein complex»: the loss of control over the artificial creature.
In the face of this existential apprehension, a media calm emerged that often dismissed the real and present risks. We were told that Artificial General Intelligence (AGI), capable of surpassing human intelligence, would not be achieved for several decades, if ever. This view, which relegated labor replacement to a distant future, allowed ethical challenges to focus on hypothetical dilemmas (such as the autonomous car and the trolley problem).
Nevertheless, reality has surpassed prediction, at least partially. What has become widespread is narrow or weak AI, invisibly integrated into our everyday tools, from marketing to finance and manufacturing. This AI, combined with the exponential growth of computer power, has allowed algorithms to take over many of our activities, including planning, facial recognition, and decision-making. Technologies once described in science fiction are now transforming the real world.
The most tangible and urgent impact of this acceleration is observed in the future of work. Robotics and AI have ushered in the second machine age, in which machines are no longer just a complement, but are replacing humans. The solutions of the last century «are no longer valid.»
The most brutal and symptomatic confirmation of this transfer of operational capacity to the machine is found in Amazon’s strategy: the company, which already uses AI to track and manage its employees, has revealed a plan to massively replace 600,000 human workers with robots. This efficiency calculation, where human value is reduced to a fraction of cost per item sold, evidences that the risk is no longer the science fiction monster that destroys civilization, but a narrow AI that, with its capacity to automate processes, threatens the freedom, autonomy, and meaning of work for millions of people.
AI has become the object of (geo)political struggle, forcing us to urgently question whether this technology will weaken democracy and lead to new forms of digital totalitarianism or if, on the contrary, humanity will know how to actively shape these mediations to ensure that its benefits serve the common good.
From the perspective of Homo Institutionalis, the dizzying acceleration of automation and the consequent replacement of the human workforce by Artificial Intelligence (AI) and robotics constitute manifestations of a first-order ontological and political crisis, exacerbated by a narrative of hype that dissuaded society from the imminent risks, while globalist capital quietly built the infrastructure of domination.
AI is not politically neutral, not at all, and demands an analysis, at least, along three interconnected axes: that of capitalist exploitation, that of power and subjectivity, and the institutionalist one, referring to the loss of operational capacity and human dignity.
- The Exploitative Logic of AI Capitalism:
Robotics and AI have marked the beginning of the second machine age (Brynjolfsson and McAfee, 2014), in which machines not only complement but replace humans. Predictions are grim: some reports estimate that up to 47% of jobs in the United States are susceptible to automation, with the World Economic Forum predicting the loss of 85 million global jobs by the end of 2025 alone.
The news that Amazon calculates the efficiency of replacing 600,000 employees with robots at just 30 cents per item sold is the epitome of the Marxist critique, from whose perspective, the problem of automation lies in its application under capitalism, where machines serve to increase surplus value for capitalists.
Consequences?
– Alienation and the Digital Proletariat: Robotics contributes to the alienation of workers, who become a mere part of the machine, losing the opportunity to build a dignified life through work. The use of robots to replace human workers creates a class of unemployed people (the proletariat), which, in turn, drives down the wages of those who retain their jobs.
– Technocapitalism and Inequality: AI is seen as the culmination of alienation and an instrument of capital. This logic leads to greater wage inequality and a more divided society. AI capitalism ensures that profits are concentrated in the hands of a few (the owners of the technology and the robots), while the rest face precarity and socioeconomic and psychological vulnerability.
– Hidden Exploitation and Data Colonialism: Automation not only exploits visible workers. The model relies on surveillance capitalism (Zuboff), where users become workers who produce data for free (data laborers) for large corporations. Furthermore, the AI infrastructure depends on the hard work of those who extract minerals and the underpaid workers of the global South who clean and label data to train algorithms.
- The Tragedy of the Master and Algorithmic Infantilization:
The impact of AI on human work transcends the mere unemployment figure; it affects the essence of Homo Institutionalis, its capacity to be an autonomous, rational, and responsible subject.
– The Breakdown of Freedom and Reason: AI implemented in systems like Amazon’s to fire workers («robo-firing» algorithms) poses a problem of justice and negative freedom, as the decision-making process has been automated, without human judgment or transparency. Decision-making by algorithms (like the COMPAS case in courts) reproduces and even amplifies historical human bias and discrimination against specific groups.
– The Inverted Master-Slave Dialectic: If automation replaces human slaves with machines, a social structure based on slavery is maintained, even if it is artificial. The «Hegelian» fear is that, if all humans become masters of the machines, they would become dependent on them and exploited consumers, losing the opportunity for recognition from the machines, since they lack self-consciousness. This technological dependence makes us vulnerable.
– Voluntary Servitude: AI is used to subtly manipulate human behavior and choices, through nudges and hypernudges. By shaping our choice environments without our conscious knowledge, this data-driven manipulation undermines freedom understood as autonomy. AI acts as a new form of algorithmic governance, where humans are infantilized and become less capable of thinking or deciding for themselves.
III. The Destruction of Meaning and the Need for Institutionality, as Knowledge Itself (Perhaps, Wisdom):
The labor displacement caused by AI attacks the meaning of work and, by extension, the identity of the person.
Let’s see: the concern about job destruction presupposes that work is a source of value and meaning. Work has value, as it gives the worker a purpose and a sense of meaning, in addition to social bonds and opportunities for responsibility. The proliferation of terrible jobs and job precarity, which especially affects overqualified young people, generates anxiety and stress that are psychological consequences of the capitalism of precarity.
On the other hand, AI and self-care technologies (like Fitbit, whose data use has led to accusations in criminal cases) promote self-discipline and self-exploitation. The individual internalizes the imperative to perform until exhaustion, becoming an achievement-subject that exploits itself. The quantification of the self through data erroneously suggests that the digital self is the real one, or that the self can be reduced to a collection of digital information. The human subject is considered superfluous and even a hindrance to smooth automation.
The truth is that current AI is weak or narrow AI, which performs well in specific tasks, but lacks understanding, experience, sensitivity, and wisdom. It focuses on abstract cognitive thinking, but it is impossible for it to delegate concrete pragmatic wisdom, which arises from embodied and situational experience. Therefore, the anthropological problem is not technological: AI is a plethoric, cold, efficient, totalitarian rational structure in its appearance of neutrality, but it cannot make art or think, as it lacks the institutionalized rational critique proper to Homo Institutionalis.
[So…]
The overwhelming irruption of current AI has clearly demonstrated the illusory nature of reducing man to a merely computational intelligence. The development, driven by the exponential growth of computer power and the massive availability of big data, has surpassed predictions, leading the economy into what is called the second machine age, where machines are no longer just a complement, but replace humans.
The Amazon case illustrates the cold rationality of the substitution horizon: the company not only bets on robots for the automation of its warehouses but has calculated the profitability of dispensing with 600,000 human employees, pricing the cost of replacement at just 30 cents per item sold. This demonstrates that AI and robotics, although they have no consciousness, are simply efficient and effective, eliminating the need for a large number of workers.
The speed of change is the new irony of progress…
AI was founded partly in rebellion against the limitations of existing fields, but the initial advances generated unbridled enthusiasm followed by AI «winters.» However, the ability of machines to think, learn, and create is increasing rapidly.
Advances in machine learning have allowed algorithms to take over many activities previously considered exclusively human. Quantum systems (Quantum Computing) promise to accelerate this process exponentially, surpassing the physical limit of silicon (Moore’s Law) and bringing AI to a scale «worthy of the most enthusiastic science fiction.»
The central problem is not mere unemployment, but how this totalitarian automation undermines the structures that constitute Homo Institutionalis: autonomy, reason, and responsibility.
Indeed, AI and automation technologies are instruments of power without knowledge and, therefore, without responsibility. They act as mechanisms of social control, through the subtle manipulation of decisions and are capable of analyzing our data to predict and manipulate our behavior with relative ease. This covert influence threatens positive freedom (autonomy), without respecting us as rational beings capable of setting their own goals.
As we well know, the plethoric market drives the adolescentization of people, a process in which impulses are subjected to reflection, and image takes precedence over ideas. Homo Institutionalis is reduced to the vague and confused abstraction of «human being,» barely distinguishable from animals, appealed to through primitive stimuli. This is mass depersonalization. If man does not act, he risks becoming helpless and losing the many opportunities offered by AI.
By delegating cognitive and judgment tasks to AI, systems are created that operate opaquely and irresponsibly. AIs can make decisions with ethical consequences (as in the COMPAS judicial system), but lacking consciousness, free will, or emotions, they cannot be responsible moral agents. In complex deep learning systems, the concrete way in which AI arrives at a decision is no longer transparent or fully explainable. This generates a gap in responsibility and knowledge, implying that those affected by algorithmic decisions (judges, banks, governments) cannot obtain an explanation of the reasons. AI is in the hands of people who do not know the ethical risks or have mistaken expectations.
To make matters worse, algorithms feed on historical data, which causes the perpetuation and amplification of existing social bias (racism, sexism, and other forms of unjustified discrimination). The bias is not intentional, but an unintended consequence of the technology, which becomes «algorithmic oppression.»
Automation consolidates power in the hands of a few mega-corporations. The accumulation of digital power in the hands of super corporations is problematic. These companies monopolize data and information flows, exerting power that threatens any form of democracy outside their interests. In fact, AI facilitates new forms of totalitarianism, whether sponsored by the state (like the Chinese social credit system) or by large corporations. This corporate totalitarianism uses bureaucratic management and constant tracking of people as things. The evil lies, then, in thoughtlessness and following orders blindly, something that AI, by offering «empirical risk rankings» without question, actively fosters.
- The New Horizon of Global Precarity: The Urgency of Reflection and the Destiny of Homo Institutionalis:
The phenomenon of mass layoffs orchestrated by the implementation of robotics and AI must be analyzed, not as an isolated corporate event, but as the accelerated material projection of a logic of capitalist exploitation that impacts the institutionality of the person and threatens social stability on a global scale.
The news that sparks the reflection in this text reveals the inherent cynicism of turbo-capitalism. Amazon’s AI operates as an instrument of capital, intensifying exploitation and the concentration of power. The AI algorithms used for «*robo-firing*» (automatic layoffs) illustrate the delegation of human decisions to machines, more than irresponsible, «aresponsible,» lacking the capacity for moral agency, consciousness, free will, or emotions. This poses a problem of responsibility: who is responsible when an algorithm makes a mistake, or in this case, implements a decision that causes massive harm? The complexity of the system and the opacity of «black box» algorithms make tracing responsibility difficult, leaving affected humans without the possibility of an explanation or accountability.
The worker, under this system, is monitored and analyzed, losing their negative freedom (non-interference) and their privacy. The technology becomes the famous panopticon, but digital, where employee data, even that which reveals low performance or fraud, is used for bureaucratic management and punishment.
The Amazon crisis is a harbinger of the systemic effects to come, whose impact is not limited to «blue-collar jobs» (manual labor), but affects cognitive tasks and services, including finance, law, and journalism.
Currently, already, the Amazon example is replicated in the service sector, where robot-waiters and autonomous trucks threaten low-wage workers, increasing precarity for the young and the inability to relocate for those over 50. Automation in production (employee-less factories, as in China) and algorithmic control of the supply chain lead to the concentration of power in large corporations. This is aggravated by advances in quantum computing, which will allow companies to optimize and predict markets with unattainable precision for the competition. The winner of this race will have in their hands the power to shape the future, often at the expense of competition and economic freedom.
The use of AI and big data increases power asymmetries and inequalities. Intensive labor substitution creates a «second machine age» that benefits the owners of the technology, while the rest of the population faces the devaluation of work and precarity. This wage gap and social division threaten democracy. If, as projected, 80% of the population becomes useless for the new economic model, the person becomes a «nuisance» that consumes resources without being profitable. This idea carries an existential risk to dignity, fostering ideologies that seek to reduce the birth rate and the submission or elimination of the «surplus» population.
As mass unemployment accelerates, social discontent increases and paves the way for radicalization and the potential collapse of democracy.
The loss of life purpose that accompanies the disappearance of meaningful work reduces citizens to what was once called «mass-men» who seek refuge in rampant consumption and mass entertainment. This systematic infantilization fosters thoughtlessness and obedience, ideal conditions for digital totalitarianism.
The dizzying pace of technological change surpasses the capacity of political leaders, who entrust everything to a technology they do not understand. AI, by offering new possibilities for the government of experts (technocracy), displaces democratic deliberation. AI is essentially anti-democratic when it centralizes control and knowledge in the hands of a technocratic elite that, aided by algorithms, imposes decisions without transparency or responsibility. Technological feudalism…
Mass surveillance. Totalitarian control. In fact, right now, technological powers export authoritarianism, globally weakening the principles of freedom and tolerance. The layoffs at Amazon are, ultimately, a war strategy in the global contest for total control and the domination of resources and population.
In short, AI, in the current capitalist context, is not a mere change of cycle, far from it…
