The synthetic proletariat: The new class dispossessed by its own data-1

In the age of artificial intelligence, digital media has ceased to be a space where knowledge is democratized and has instead transformed into a massive liquidation center where collective intelligence is transferred to capital. While every piece of content we produce today becomes fuel for machines, the “Synthetic Proletariat” is building its own exploitation with its own hands.

The promise of digital media was the democratization of knowledge and the dismantling of hierarchies. However, looking back from 2026, we see the emergence of a synthetic proletariat class whose very existence has become commodified. This class begins its shift every morning the moment it picks up its phone. With every tweet posted, every photo shared, and every debate entered, it creates digital value. In this paradoxical process where content (the subject) is severed from the human and transformed into a purely algorithmic commodity, we offer our existence, our style, and our cognitive capacity as fuel to the massive data pools that train AI.

In the recent period (2024-2025), the multi-billion dollar data licensing agreements that OpenAI and Google made with platforms like Reddit and News Corp became the most concrete evidence of this exploitation. Platform owners sold this massive collective accumulation to AI companies as a package of raw material. The labor of users who spent years sharing information, debating, and building a collective intelligence on Reddit forums for free was transferred to corporations by the platform owners. What is happening here is one of the greatest transfers of cognitive surplus value in history. Every moment the user thinks they are socializing, they are actually building the intelligence of the algorithms that will replace them. The price of the deal did not go to the users, but to the platform's shareholders. We can call the result cognitive inflation. An article, analysis, or visual carries value due to the research process, intellectual accumulation, and time spent behind it. Now, far more synthetic content is being released into the market than human capacity to understand and process, leading to cognitive inflation.

Cognitive Inflation: The Devaluation of Meaning and Digital Enclosure

In economics, if the supply of a good approaches infinity, its value drops to zero. Today, generative AI (GenAI) has evaporated the marginal value of content by reducing the social labor time required for content production to seconds. A synthetic text produced in seconds by machine and deep learning has devalued organic labor. With the “click economy” revenue model, social media has evolved into an AI junkyard. Platforms train AI by exploiting the data created by users' free digital labor, and then use that same AI to pile “synthetic trash” (AI Slop) in front of those same users. In short, the synthetic proletariat is the mass of people who produce content—using or not using AI—in an environment managed by algorithms, and who are exposed to low-quality information filtered by that same AI. Meanwhile, in an internet of absolute content abundance, users are also being classified. While the masses are fed low-quality synthetic trash (AI Slop) produced by algorithms, cluttered with advertisements, and open to manipulation, a minority has emerged that has the luxury of escaping digital noise by paying subscription fees to access human-made, verified, and in-depth information (Substack, private newsletters, boutique publishing, etc.). The media industry's tendency to protect itself as a business and revenue model against technology-based platforms is undeniable in this regard.

In 18th-century England, the enclosure of common lands dispossessed the peasantry. Today, the public spaces of the internet (social media, Wikipedia, open forums, personal blogs) are being enclosed by global technology and AI companies to train AI models. In this state, the digital media economy makes us both the architects and the prisoners of a prison woven with our own data. The synthetic proletariat is a mass of people who work in the digital “media factory,” offering their own lives, ideas, and images to feed AI models and platform algorithms, yet having no control over the content they produce. And if we do not redefine the ownership of knowledge, the right to data, and the law of “human-labor cognitive products,” we will remain in the shadow of the algorithms we created ourselves in this synthetic noise where everything exists but nothing is valuable.

NEXT WEEK: We introduced the concept of the “Synthetic Proletariat” this week. In the next article, we will discuss the hidden geography of digital labor extending from “click farms” to AI data silos, the major data deals where the ownership of knowledge changes hands, and whether it is possible to exit this content noise.