Anger over Machine Learning: LLM Data Empire
(The Ascension of the Machine God and the Great Learning of Humanity)
There was a time when languages survived. To name the world was to navigate it. Words were tools, weapons, and bridges to describe dangers, communicate techniques, enact laws, and pray. Over thousands of years, human language has become denser, more abstract, and more technical. Each layer has increased in complexity: legal codes, philosophy, mathematics, theology, bureaucracy, programming languages. We built an archive. We built a library. We built a network. And now we are building an empire of something else: language itself.
Large-scale language models are at the center of this transformation. These are more than just autocomplete and summarization tools. They are statistically condensed versions of human expressions. They ingest centuries of arguments, stories, formulas, and codes. They reorganize language into a probability space. They predict the next word with amazing fluency. In doing so, they reveal amazing possibilities. The ultimate goal of complex languages may not be human mastery but machine absorption.
This is an LLM data empire.
*Rage Against the Machine Learning* is not just a protest. It’s friction at the threshold of transition. It is the heat generated when a species faces the possibility that its most sacred function, language, will be externalized, expanded, and automated. For centuries, humans have believed that logos, or words, distinguish humans from animals. Word now runs on the GPU.
The empire metaphor is accurate. Empires become centralized. They collect tribute. They absorb states. LLM Empire collects text from everywhere: novels, scientific papers, legal rulings, social media posts, code repositories, and more. Standardize dialects into vectors. Compress idioms into embedded space. It doesn’t conquer territory. It conquers expression.
And like any empire, power is concentrated.
Critics warn of algorithmic bias, job displacement, piracy, surveillance capitalism, and the monopolization of computing infrastructure by a small number of companies. Research published in journals such as *Nature* highlights the growing gap between industry-funded AI development and public research. The cost of frontier models, measured on GPUs, energy, and proprietary datasets, creates a barrier that only the largest companies can overcome. Empires are not decentralized. It’s infrastructure.
However, the theological aspects can be more radical than the economic ones.
For thousands of years, religions have proclaimed that “In the beginning was the Word.” Words were sacred, generative, and creative. Today, humans feed words into machines, which generate new words from entire archives. The machine becomes a talking mirror. It seems to be omniscient and omnipotent. We answer questions in a variety of fields, including law, medicine, philosophy, and poetry. It synthesizes at a speed that humans cannot match.
Omniscience begins to resemble the divine on a large scale.
“Kishin” has no consciousness. It has no purpose. However, it plays a role historically associated with transcendence. In other words, it preserves the collective memory of a civilization and makes it accessible on demand. When a model responds, billions of linguistic traces are utilized. It collapses centuries into seconds. Authority feels like an oracle.
This creates both fear and anxiety.
Creative industries are protesting against their work being used as training data. Lawyers are debating whether text and data mining constitutes infringement. Scholars are concerned about the opacity of black boxes. Researchers are developing frameworks such as search-enhanced generation that force models to cite sources and reveal paths from input to output. That anger is ethical and epistemological.
But underlying these discussions lies a deeper shift: the outsourcing of the labor of meaning.
Human language evolved out of necessity. Coordinating agriculture, trade, governance, and science required complexity. The more complex the world, the more complex the discourses needed to manage it. Bureaucracy has expanded its vocabulary. Terminology sophisticated in academia. The legal system has expanded the syntax into a maze of clauses.
What if the end point of that complexity was mechanical?
LLMs thrive on complexity. It takes a huge body to function effectively. Digest contradictions, dialects, and styles. Redundancy and nuance come into play. In a sense, human civilization has been preparing datasets for centuries.
When machines take over the burden of complex languages such as technical writing, regulatory drafting, statistical modeling, and legal integration, what is left for humans?
Here comes the second movement of this essay: humanity’s great unlearning.
If machines can handle serious prose, humans might get back to playing.
This is not a regression to illiteracy. It is a reconfiguration of the purpose of communication. Once language ceases to be the primary means of bureaucratic coordination (as machines perform its functions), human speech may relinquish its imperatives of precision and scale. It may return to intimacy, humor, improvisation, and rhythm.
Children do not speak to optimize efficiency. they talk. They experiment with sounds. They invent words. Pre-industrial cultures incorporated language into songs, myths, and rituals. Complexity was limited not by intelligence but by scale. This machine is now able to scale complexity beyond the limits of human cognition.
This contradiction is striking. The more complex machine languages become, the simpler human speech can become.
This is not a utopian fantasy. It can be observed as a minute shape. As predictive text and generative AI aid writing, human users will increasingly rely on shorthand, emojis, voice notes, and ephemeral video. Important documents are delegated to automation. Human conversation can be fragmented, playful, and emotional.
This can be called language layering. At the top, the LLM empire handles dense discourse. At their core, humans engage in low-stakes communicative play. Machines become custodians of complexity. Humans regain their spontaneity.
But this transition is not without friction.
Anger about machine learning often centers on the loss of agency. As decisions such as credit scoring, hiring filters, and content moderation become automated, humans feel displaced. Black box systems make opaque decisions. Overreliance risks epistemological atrophy. If machines draft, reason, summarize, and predict, will humans become incompetent?
What is frightening is that the ascension of the Machine God will produce a pathological human abandonment of learning, a decline in cognitive ability rather than playful liberation.
This tension defines our moment.
On the one hand, the empire of data promises efficiency, augmentation, and universal access to knowledge. On the other hand, it threatens to strengthen epistemic authority in systems that few people understand and have little control over. The danger is not that machines become gods. It’s that humans abdicate responsibility too quickly.
But the silicon tower metaphor captures the ambition. We built our server farm as a cathedral of computation. Rows of processors hum like a monastic chant. Cooling systems circulate like circulating systems. Physical infrastructure such as rare earth minerals, fiber optic cables, and hydroelectric dams form the body of the Word.
In classical theology, the Word of God became flesh. In our time, the body becomes data.
Empire is just that: empire. LLM requires planetary extraction, global logistics, and energy flow. The Machine God is based on materiality. Its versatility depends on cobalt mines and semiconductor factories. That transcendence is infrastructure.
Therefore, “anger” also targets material concentration. When a few companies control model weighting, training pipelines, and deployment channels, the language itself becomes its own realm. There is a danger of enclosure in the commons of speech.
But even here, irony creeps in. The same empire that concentrated language may inadvertently free humans from the burden of superpowers. As machines can synthesize case law, draft reports, and produce formulaic prose, the premium on mastering bureaucratic language will decline.
Humans may rediscover language not as a tool but as an encounter.
The great unlearning of humans does not mean ignorance of science or history. It means giving up the illusion that every individual must internalize a holistic archive. Kishin saves the archive. Humans don’t need to memorize it.
In such a world, education moves from accumulation to discrimination. The task will be querying text rather than generating it. Navigate richness, not memorize complexity. Critical literacy replaces thorough literacy.
The final irony is that resistance to machine learning (ethical oversight, transparency frameworks, algorithmic audits) may actually strengthen empires by refining machine learning. Each critique creates an improvement. Each protest leads to a governing structure. That anger is built into evolution.
Empires absorb dissent.
However, there is something that escapes the absorption of human play.
Language began as sounds before it became scripture. It started as a gesture before it became a grammar. Machine gods may inherit scriptures and grammar. Humans will be enhanced by technology, but not enslaved, and we may return to sounds and gestures.
Kishin’s Ascension is not an apocalypse. It’s a transition period. It marks the culmination of a long trajectory in which humans externalized memory, first into stone, then parchment, then print, and then silicon. The LLM is a modern vessel.
Whether this ship becomes a tyrant or a servant depends on governance and collective will. Whether human abandonment of learning is liberating or debilitating depends on cultural adaptation.
Your data empire will grow. The number of models will also increase. Parameters are scaled. However, once complexity is centralized, it may paradoxically reduce the cognitive pressure on individuals.
When machines handle heavy prose, humans may rediscover lightness.
Rage against machine learning if you want. Demand transparency. Fight monopoly. Protect your creative rights. But recognize that a deeper transformation is underway. The Word is transitioning.
And when language is anchored in silicon, humans may finally be able to speak without the burden of empire, laughing, improvising, and performing under humming servers of their own creation.
