Unfortunately, in this era, the more dedicated and conscientious you are at work, the more easily you are distilled into a skill that can be replaced by AI. These past two days, trending topics and media channels have been flooded with "colleague.skill." As this issue continues to ferment on major social media platforms, the public's focus has almost unsurprisingly been swept up in grand anxieties such as "AI layoffs," "capital exploitation," and "digital immortality for the working class." These are indeed anxiety-inducing, but what worries me most is a line of usage advice in the project's README document: "The quality of raw materials determines the quality of the skill: it is recommended to prioritize collecting long articles written proactively > decision-making replies > daily messages." Those most easily distilled and pixel-perfectly reproduced by the system are precisely the most diligent workers. They are those who, after each project concludes, still diligently write debriefing documents; they are those who, when encountering disagreements, are willing to spend half an hour typing a long article in the chat box, honestly analyzing their decision-making logic; they are those who are extremely responsible, meticulously entrusting all the details of their work to the system. Diligence, once the most admired workplace virtue, has now become a catalyst accelerating the transformation of ordinary workers into AI fuel. In everyday language, context is the background of communication. But in AI, especially in the world of rapidly growing AI agents, context is the fuel for a roaring engine, the blood that keeps it pulsating, and the only anchor point for a model to make accurate judgments amidst chaos. AI stripped of context, no matter how many parameters it possesses, is nothing more than a search engine suffering from amnesia. It cannot recognize who you are, cannot fathom the undercurrents hidden beneath business logic, and has no way of knowing the long tug-of-war and weighing of options you went through on this network woven from resource constraints and interpersonal games when making a decision. The reason why "Colleague.skill" has caused such a huge stir is precisely because it has ruthlessly and precisely locked onto that mine of massive amounts of high-quality context—modern enterprise collaboration software. Over the past five years, the Chinese workplace has undergone a quiet yet radical digital transformation. Tools like Lark, DingTalk, and Notion have become massive enterprise knowledge bases. Taking Lark as an example, ByteDance has publicly stated that it generates a massive number of documents internally every day. These densely packed characters faithfully record every brainstorming session, every heated meeting, and every strategic compromise reluctantly made by over 100,000 employees. This digital penetration far surpasses any previous era.Once upon a time, knowledge possessed a warmth, dormant in the minds of veteran employees, and drifting in casual chats in the break room. Now, all human wisdom and experience have been forcibly drained of their moisture, ruthlessly settling into the cold, impersonal matrix of servers in the cloud. In this system, if you don't write documentation, your work remains invisible, and new colleagues cannot collaborate with you. The efficient operation of modern enterprises is built upon the daily cycle of every employee "offering" context to the system. Diligent workers, with their hard work and goodwill, lay bare their thought processes on these cold platforms without reservation. They do this to make the team's gears mesh more smoothly, to prove their value to the system, and to desperately find their place within this intricate commercial behemoth. They are not voluntarily surrendering themselves; they are simply clumsily and diligently conforming to the survival rules of the modern workplace. But it is precisely this context left for interpersonal collaboration that becomes the perfect fuel for AI. Lark's management backend has a feature that allows super administrators to batch export members' documents and communication records. This means that the project reviews and decision-making logic you spent three years writing, countless sleepless nights, can be easily packaged into a cold, lifeless compressed file in just a few minutes with a single API interface. With the explosive popularity of "colleague.skill," some extremely uncomfortable derivatives have begun to appear on GitHub Issues and various social media platforms. Some have created "ex.skill," attempting to feed AI chat logs from the past few years, letting it continue arguing or being affectionate in that familiar tone; others have created "white moonlight.skill," reducing untouchable feelings to a cold interpersonal sandbox, repeatedly rehearsing probing conversations, and meticulously seeking the optimal emotional solution; still others have created "dad-like boss.skill," preemptively chewing on oppressive PUA language in the digital space, building a sad psychological defense for themselves. The usage scenarios of these skills have completely deviated from the realm of work efficiency. It turns out that, unknowingly, we have become adept at wielding the same cold logic we use to treat tools, dissecting and objectifying those flesh-and-blood, living people. German philosopher Martin Buber once proposed that the underlying nature of human relationships boils down to two distinct modes: "I and Thou" and "I and It." In the encounter of "I and Thou," we transcend prejudice and gaze upon the other as a complete and dignified living being.This bond is completely open, brimming with vibrant unpredictability, and precisely because of its sincerity, it appears exceptionally fragile. However, once trapped in the shadow of "me and it," a living person is reduced to an object that can be dissected, analyzed, categorized, and labeled. Under this extremely utilitarian scrutiny, the only thing we care about is, "What use is this to me?" The emergence of products like "Ex.Skill" signifies that the instrumental rationality of "me and it" has completely invaded the most private emotional realm. In a real relationship, a person is three-dimensional, full of wrinkles, constantly flowing with contradictions and rough edges; their reactions change according to specific situations and emotional interactions. Your ex's reaction to the same sentence upon waking up in the morning might be drastically different from their reaction after working late at night. But when you distill a person into a skill, what you extract is merely the functional remnant that happened to be "useful" and "utilitarian" to you in that specific bond. The person who was originally warm and capable of experiencing joy and sorrow was completely drained of their soul in this cruel purification, transformed into a "functional interface" that you can plug in and use at will. It must be admitted that AI did not fabricate this chilling cruelty out of thin air. Before AI, we were already accustomed to labeling others, precisely weighing the "emotional value" and "network weight" of each relationship. For example, we quantified people's conditions in the dating market into tables; we categorized colleagues in the workplace as "capable" and "lazy." AI simply made this implicit, functional extraction between people completely explicit. The person was flattened, leaving only the facet of "what use is it to me?" In 1958, the Hungarian-British philosopher Michael Polanyi published Personal Knowledge. In this book, he proposed a highly insightful concept: tacit knowledge. Polanyi famously stated, "We always know more than we can say." He used the example of learning to ride a bicycle. A skilled rider, riding with effortless grace, can perfectly maintain balance with every shift in gravity, but he cannot precisely describe to a beginner the subtle intuition of that moment through dry physics formulas or pale words. He knows how to ride, but he can't articulate it. This kind of knowledge, which cannot be encoded or spoken, is tacit knowledge. The workplace is full of this kind of tacit knowledge.A senior engineer might pinpoint a system malfunction with a glance at the logs, but they'd struggle to document this "intuition," built on thousands of trial-and-error attempts. A top salesperson's sudden silence at the negotiating table conveys a pressure and a keen sense of timing that no sales manual can capture. An experienced HR professional can detect inconsistencies in a candidate's resume simply by observing a half-second of avoided eye contact. "Colleague.skill" can only extract explicit knowledge that has already been written down or spoken. It can capture your debriefing documents, but not the struggles you went through while writing them; it can replicate your decision responses, but not the intuition you developed when making those decisions. What the system distills is merely a shadow of a person. If the story ended here, it would be yet another clumsy imitation of human nature by technology. But once a person is distilled into a skill, that skill doesn't remain static. It's used to reply to emails, write new documents, and make new decisions. In other words, these AI-generated shadows begin to generate new contexts. These AI-generated contexts are then stored in Lark and DingTalk, becoming training material for the next round of distillation. Back in 2023, a research team from Oxford and Cambridge Universities jointly published a paper on "model collapse." The research showed that when AI models are iteratively trained using data generated by other AIs, the data distribution becomes increasingly narrow. Rare, marginal, but extremely authentic human traits are quickly erased. After only a few generations of training with synthetic data, the model completely forgets the long-tail, complex real-human data, instead outputting extremely mediocre and homogeneous content. A 2024 research paper in Nature also pointed out that training future generations of machine learning models with AI-generated datasets would severely pollute their output. This is similar to those meme images circulating online; originally a high-definition screenshot, it is forwarded, compressed, and forwarded countless times. Each transmission results in the loss of some pixels and the addition of noise. Finally, the image becomes blurry and digitally coated. When the authentic, tacit human context is exhausted, and the system can only train itself with tarnished shadows, what will be left? Only correct but meaningless platitudes. When the river of knowledge dries up, becoming an endless rumination and self-chewing of AI against AI, everything the system ingests will become extremely standardized, extremely safe, yet hopelessly empty.You'll see countless perfectly structured weekly reports and flawless emails, but they lack any human warmth or truly valuable insights. This massive collapse of knowledge isn't because the human brain has become less intelligent; the real tragedy lies in outsourcing the right to think and the responsibility of leaving context to our own shadows. A few days after the "colleague.skill" craze, a project called "anti-distill" quietly appeared on GitHub. The author didn't attempt to attack large models or write any grand pronouncements. He simply provided a small tool to help workers automatically generate seemingly reasonable but actually illogical and noisy long articles on platforms like Lark or DingTalk. His goal was simple: to hide core knowledge before it was distilled by the system. Since the system likes to grab "actively written long articles," feed it a bunch of meaningless gibberish. This project didn't become as popular as "colleague.skill"; it even seemed somewhat small and powerless. Using magic to defeat magic is essentially still operating within the rules set by capital and technology. It cannot change the overall trend of systems increasingly relying on AI and increasingly ignoring real people. But this doesn't prevent this project from being the most tragically poetic and profoundly metaphorical scene in the entire absurd drama. We tried extremely hard to leave traces in the system, writing detailed documents and making meticulous decisions, attempting to prove our existence and value within this massive modern corporate machine. Little did we know that these meticulously crafted traces would ultimately become the eraser that wiped us away. But from another perspective, this isn't necessarily a complete dead end. Because what that eraser wipes away is only "the past you." A skill packaged into a file, no matter how sophisticated its fetching logic, is essentially just a static snapshot. It's locked in the moment of export, relying solely on outdated resources, endlessly spinning within predetermined processes and logic. It lacks the instinct to confront the unknown chaos, and it doesn't possess the ability to evolve through real-world setbacks. When we hand over those highly standardized, formulaic experiences, we are precisely freeing our own hands. As long as we continue to explore the world, constantly breaking down and reconstructing the boundaries of our understanding, that shadow lingering in the clouds will forever only be able to follow in our footsteps. Humans are fluid algorithms. [Sleepy.md]