ROBOTS: CONQUERORS OR WORKERS?

If one were to take recent films and media at face value, humanity seems to be only a short distance away from being ruled by robots. For nearly a century, Hollywood has been preparing us for this supposedly inevitable encounter. When a Terminator does not arrive from the future—Arnold in a spectacular role—to eliminate the leaders of human resistance, an artificial intelligence emerges that concludes humans are the planet’s primary problem and must therefore be replaced. Yet so far, the only ones who have truly conquered the world are science‑fiction screenwriters. Today, with the rise of social media (see the blog entry “Saint TikTok and the Liturgy of Live Foolishness”), viral videos appear almost weekly. A robot performs an impressive jump, dances, runs, lifts a box, or manages—after several hundred attempts—to catch a ball. Dramatic headlines immediately follow: “The end of human labour is near!”, “Robots are taking over!”, “Humanity’s final days!” One might easily imagine that, by the end of the month, the robot in question will be running for Secretary‑General of the United Nations. At the same time, major technology companies and investors promise millions of robots working in factories, hospitals, warehouses, restaurants, and even in our homes. At times, the enthusiasm is so exaggerated that one could believe the only thing preventing robots from ruling the planet is that they have not yet finished charging their batteries. The reality, however, is far less spectacular—almost mundane. Yes, robots are improving. Yes, artificial intelligence is advancing at an impressive pace. And yes, certain professions will undergo profound transformation in the coming decades. But the leap from this to a world in which a robot fires your boss, governs your country, and explains why it knows better how life should be lived is enormous. This article does not aim to demonise robots nor to portray them as humanity’s saviours. We will set aside both the flashy advertisements and the apocalyptic narratives, and examine the situation as it truly is. We will consider what robots can do today, where they are already useful, where they still desperately require human assistance, and—most importantly—how far we remain from the day they might take control of the world. A small hint: if a robot still needs three engineers and two hours of calibration to climb a staircase without falling, humanity can probably continue to sleep peacefully for a while.

The Largest Robot Factory in History

If there is one country that takes robotisation seriously, it is China. In recent years, both government authorities and major corporations have invested billions of dollars in factory automation. This is not merely a demonstration of technological prowess but also an economic necessity. The population is ageing, the number of young people entering the labour market is declining, and wages have risen significantly compared to two decades ago. For many factories, a robot is no longer a futuristic luxury but an investment that can reduce costs and keep production competitive. The results are striking. Hundreds of thousands of industrial robots are installed each year, and some factories resemble vast laboratories rather than the noisy halls we typically imagine when thinking about industry. Robotic arms weld, paint, assemble components, and move products with a precision few humans could sustain for eight hours a day. Yet if one follows the advertisements of robotics companies, the next step appears inevitable: by tomorrow morning, Beijing will be governed entirely by androids, and the mayor will be replaced by a robot making decisions based on an algorithm. Meanwhile, humanoid robots will go shopping, cook dinner, and deliver speeches on administrative efficiency. Fortunately—or unfortunately, if one hoped for a live science‑fiction film—the reality is far less spectacular. Most of the robots installed in Chinese factories do not resemble humans at all. They are mechanical arms fixed to the floor or to production lines, performing one or two tasks exceptionally well. One tightens the same screw thousands of times a day. Another welds the exact same joint. Another moves the same box from point A to point B, without boredom and without requesting leave. Their performance depends on operating within a perfectly ordered universe. Every part arrives precisely where it should, every movement is programmed, and the environment is controlled down to the smallest detail. The robot does not need to make complex decisions. It does not need to improvise. It does not need to infer the intentions of colleagues or resolve unexpected situations. Problems arise precisely when the real world refuses to follow the script. If a part arrives a few millimetres off its intended position, if an object is deformed, if an obstacle appears, or if someone changes the sequence of operations, the robot can become surprisingly helpless. At that moment, the human operator steps in, identifies the issue, finds a solution, and restarts the process. The robot does not protest, does not get upset, and does not learn from the experience. It simply waits for instructions. This is the major contrast between the promoted image and reality. Yes, China is building the most robotised economy in the world. Yes, the number of industrial robots is increasing at an impressive rate. But these robots are not universal workers capable of doing anything. They are highly disciplined specialists, each employed for a single task, in an environment where almost nothing is left to chance. Therefore, if someone claims that robots are on the verge of taking over humanity, it is worth first asking whether they can move a box that has not been placed exactly in the designated spot. In many cases, the answer is still… no.

An Ironic Note: Control Never Disappears, It Merely Changes Form

We—Angelo and George—coming from a totalitarian system with deeply ingrained centralising reflexes and a historical weakness for absolute control, recognise the pattern instantly: no matter how advanced technology becomes, no one relinquishes power without hesitation. And here lies perhaps the most amusing aspect of the discussion about China and technology. Within the cold logic of realpolitik, there is not a single moment in which a highly centralised political system would accept the idea that an algorithm—even a “super‑intelligent” one—could operate the levers of authority without human oversight (see the blog entry “Tyranny No Longer Shouts. It Sends Notifications.”). In practice, robots are welcome—but only insofar as they remain precisely what they are: instruments. And the example that invariably provokes a smile is the Chinese digital ecosystem. While the rest of the world debates whether an application is “too influential” or “too intelligent,” China has achieved the remarkable feat of exporting the global version of TikTok, while maintaining at home a completely separate variant, known as Douyin, integrated into a tightly controlled internet. Outwardly—a global festival of content. Inwardly—a carefully filtered system in which information circulates in an orderly fashion, through well‑defined channels, as if following a real‑time manual of social engineering. Of course, in the more sarcastic register of the discussion, some might say it is a fortunate coincidence that “export products” are open and viral, while “domestic consumption” remains impeccably filtered and secured. Yet beyond the irony, one thing remains clear: control never disappears; it merely changes form. And in such a context, the notion that a robot could “take over” within a system dominated by human decision‑making, hierarchy, and an instinct for absolute control remains, at least for now, more an exercise of imagination than a realistic scenario.

Real Tests, Global Fantasies

If one were to believe only what circulates online, the world’s governments already command entire armies of invisible robots capable of solving any problem—from fires to global conflicts—while sipping oil and updating their software. The reality is far less spectacular and far more pragmatic. Behind closed doors or in tightly controlled environments, all major powers test robots. Not in order to “hand over the leadership of the world,” but for a much simpler reason: to determine where they can be useful and where they still cannot. In the military, robots and autonomous systems are used for reconnaissance, logistics, surveillance, or reducing risks for soldiers. They are not strategic generals but rather extremely expensive assistants that do not tire, yet understand context poorly without a human behind them. In policing, tests include monitoring systems, patrol robots, or assistance in dangerous situations. The logic is straightforward: if a situation is risky, it is better to send a machine that can be repaired than a person who cannot be replaced as easily. In firefighting, robots are deployed where smoke, heat, and collapsing structures make human intervention extremely dangerous. They are useful, but still far from the image of an autonomous rescuer who enters, solves everything, and walks out whistling. In nuclear power plants, their role is even clearer: inspection, measurement, and targeted interventions in environments where human exposure must be minimised. There is no room for error here, but neither for the robot’s “creative intelligence.” Everything is strictly controlled. In natural disasters, robots and drones are used to explore inaccessible areas, transmit images, or transport equipment. They serve as additional eyes and hands, not decision‑makers.

The Japanese and Their “Atomic Robots”

A current example comes from Japan, where robots and semi‑autonomous systems are already used for inspecting nuclear power plants and conducting interventions in contaminated or otherwise inaccessible environments. After the 2011 Fukushima disaster—triggered by the earthquake and subsequent tsunami—the development of such technologies became an explicit priority. Not because robots are “ready to take control,” but because humans, quite reasonably, prefer not to enter areas where radiation levels do not negotiate with human biology. Robots enter, measure, transmit data, and—crucially—exit the scene without drawing philosophical conclusions about the future of civilisation. All these applications illustrate the same point: governments are not waiting for robotic revolutions; they are trying, in a highly pragmatic manner, to understand what works and what still does not. It is a slow, careful, and at times frustratingly mundane process for those expecting science‑fiction spectacles. Of course, there is also a less transparent side. Not all tests are presented at press conferences or in promotional videos accompanied by dramatic music. Some programmes are classified; others appear only partially in the public sphere. And from here, collective imagination inevitably takes over. The moment the word “secret” appears, the internet fills in the rest of the story without hesitation. Thus we arrive at the classic scenario: somewhere, in an underground bunker, ultra‑intelligent robots play chess with extraterrestrials, discuss global politics, and optimise the world economy between software updates—presumably with short breaks for recharging and philosophical debates about the meaning of human existence. In reality, however, the technology is far more down‑to‑earth. The robots tested by governments are far from absolute autonomy. They are useful systems—sometimes impressive—but still fragile, dependent, and constrained by context. In other words, before they can “run the world,” they must pass a far more modest test: functioning reliably on a rainy day, in a chaotic environment, without requiring a manual reset after every minor error. And until then, the notion of “global takeover” remains better suited to low‑budget television series than to technical reports.

Ukraine: The First War in Which Robots Truly Matter

The war in Ukraine is likely the first modern conflict in which robots are no longer “the future,” but a real and constant component of the battlefield. They do not command operations, yet they directly influence survival, reaction speed, and the cost of every military action. The most visible category is that of drones. Ukraine employs a wide range of reconnaissance and attack drones, from modified commercial models—including the DJI Mavic many of us have at home, adapted for frontline use—to military drones such as the Bayraktar TB2, heavily used in the early phases of the war for strikes on Russian columns and logistical positions. Later, the conflict entered an “era of cheap drones,” small and extremely efficient, used for precise tactical‑level attacks. On the other side, Russia uses drones such as the Orlan‑10 for reconnaissance and artillery fire correction—one of the most important functions in modern warfare: rapid target acquisition. But beyond the air domain, there are also ground robots. Systems like the Milrem THeMIS (used by Ukraine with Western support) serve in logistics, ammunition transport, and the evacuation of wounded soldiers from dangerous areas. The logic is simple: if a route is too risky for humans, send an unmanned platform capable of carrying supplies or extracting an injured soldier from the frontline. In medical evacuation and logistical support, these systems are not spectacular, but they are decisive at the human level: they reduce casualties in situations where every minute matters. Another crucial domain is demining. Organisations such as The HALO Trust and Ukrainian emergency services use specialised robots to identify and neutralise mines in a country that has become one of the most heavily mined in the world. Here, the robot does not “conquer the battlefield”; it enters spaces where a single misstep would mean the immediate end of the mission. Public estimates indicate that Ukraine now operates tens of thousands of active drones simultaneously on the front, with constant deliveries of new units each month. Consumption at the frontline is so high that some units treat drones as expendable munitions rather than durable equipment.

The essential observation remains unchanged: none of these systems has replaced soldiers. Robots and drones do not make strategic decisions, do not control the front, and do not substitute human presence. Instead, they reduce direct risk for personnel, increase reaction speed, and reshape the way war is conducted. In other words, Ukraine is not witnessing a “robot war,” but a conventional war in which robots have suddenly become indispensable tools—useful, efficient, yet still far from any notion of genuine autonomy.

India and the “Teachers” of Robots

If there is a place where the paradox of artificial intelligence becomes evident, it is India. Here, behind grand promises of autonomous robots and “intelligent” systems, lies a much simpler reality: people who teach them, step by step, how to appear intelligent. Across numerous Indian cities, companies and platforms specialising in data labelling employ workers to annotate images, texts, sounds, or conversations. They mark objects in photographs, classify emotions in texts, correct responses, or help AI systems recognise patterns. In short, an enormous volume of human labour feeds into the backstage of artificial intelligence: without these clean, labelled datasets, no modern AI model would function at the level we see today. The paradox is clear. While public discourse speaks of robots replacing humans, in reality, real people spend hours in front of screens teaching these future “replacements” how to recognise a cat, an ambiguous sentence, or a complex social situation. It is, practically, one of the most ironic labour relationships of the digital era: humans train the systems that, theoretically, are meant to reduce the need for humans. Sarcastically speaking, it is probably the first case in history in which someone is paid to train their own competitor—albeit with some peace of mind, hoping that this “competitor” is still far from asking for a salary, a lunch break, or a union. Yet the technical reality is far less dramatic. Artificial intelligence models do not “understand” what they process. They do not think, interpret, or possess consciousness. Essentially, they learn extremely complex statistical patterns based on correlations within data. If they appear intelligent, it is because they have been trained on vast quantities of information labelled by humans. In other words, “robots” do not learn like humans; they merely reproduce patterns derived from human labour.

The Student and the Laptop

Another important element of this picture is the way India has rapidly embraced the digital revolution. In recent years, programmes have been implemented through which many students receive access to laptops or digital devices as part of a broader strategy to integrate the country into the global technological economy. It represents an accelerated shift toward digitalisation in a nation with one of the largest youth populations on the planet. However, this transformation also has a less frequently discussed dimension. In an economy with hundreds of millions of people in the process of training and professional integration, rapid automation and full digitalisation can generate significant social tensions. In the long term, increased technological efficiency does not automatically guarantee an increase in employment opportunities, and labour‑market adaptation becomes a formidable challenge. In other words, while India is investing heavily in its digital future, the question remains how effectively it will manage to balance technological progress with the reality of an immense population dependent on traditional forms of human labour. And here emerges one of the major ironies of the current era: the very countries that adopt technology most rapidly are also the most exposed to its social consequences.

Jobs

The discussion about robots and artificial intelligence inevitably reaches the same sensitive topic: employment. Every time a new technology emerges, the same old question follows: “Will we still have work to do?” As usual, the answer is far less dramatic than online headlines suggest. The positive side brings real and difficult‑to‑ignore benefits. First, dangerous activities disappear or are significantly reduced. Work in toxic environments, unstable zones, fires, mines, or repetitive high‑risk operations can be taken over by automated systems or robots. This means fewer accidents and fewer lives placed in danger. Second, productivity increases. A robot does not get tired, does not take breaks, and does not have “bad days.” In factories, logistics, or data analysis, this translates directly into higher efficiency. New professions also emerge: robotics engineers, data specialists, maintenance technicians, programmers, cybersecurity experts, and many others that did not exist a few decades ago. Moreover, in certain fields where labour shortages are already a problem, automation arrives as a practical solution rather than a threat. The less comfortable side hits like a hammer. Some repetitive jobs will disappear or be significantly reduced. Not because “robots are bad,” but because they are cheaper, faster, and more consistent in certain tasks. Reskilling becomes mandatory, not optional. People will need to learn new skills—sometimes multiple times throughout their professional lives. At the same time, social imbalances emerge. Not all regions and not all social groups adapt at the same pace. Some transition quickly to new jobs, others fall behind, and economic disparities may deepen. The realistic conclusion is that the pattern repeats itself. Every major technological revolution has eliminated certain professions and created new ones. The automobile reduced the need for horse‑drawn carriages, electricity completely reshaped industry, computers eliminated thousands of administrative tasks, and the internet redefined communication and commerce. None of these technologies “eliminated work.” They merely changed its form. Robots and AI do not appear to be exceptions to this pattern. The difference lies in the speed of change, not its direction. In other words, the issue is not a world without jobs, but a world in which jobs change faster than we have been accustomed to adapting to them.

Will Robots Conquer the World?

If one were to rely solely on the internet, the answer already seems written: yes, robots are only two steps away from taking over the planet. Soon they will govern states, optimise the global economy, and perhaps even write reviews about their own domination. Reality, however, has a far harsher sense of humour. A modern robot—at least in its “spectacular” present‑day form—may look impressive in a few‑second promotional clip. But once removed from the laboratory and placed in the real world, things change quickly. A robot tumbles down stairs with a grace that would make a sack of potatoes envious. It fails to open a door if the handle is not exactly the type it was “trained” on. It trips over a cable stretched across the floor like a hurried tourist in an airport, and its battery dies precisely when it becomes “useful.” Yet, in parallel, the same scene is interpreted online as a sign of inevitability: “It has at most two years before it runs the planet!” This is technological optimism taken to an extreme, where every successful movement is proof of an apocalyptic future, and every failure is politely ignored as a trivial detail in the “learning process.” The irony becomes even clearer when we look at the infrastructure behind these “future conquerors of the world.” Behind every robot that manages a single step without falling, there are typically three engineers checking each line of code, two laptops running diagnostics, and a technician intervening faster than the robot can “make a decision.” Sometimes a well‑placed cable explains why the entire system has entered an unplanned pause. In other words, before conquering the world, many robots still have a rather fragile relationship with gravity and power outlets. The unintentional humour of the situation comes precisely from the contrast between perception and reality. While promotional images suggest autonomy, control, and near‑human intelligence, real‑world performance reveals systems that are sophisticated yet dependent, fragile, and surprisingly easy to “confuse” with a door or an unexpected event. Thus, the question “Will they conquer the world?” remains open. But judging strictly by their current balance, the honest answer seems to be: first they need to conquer a single step without stopping due to an error.

What Spectacular Demonstrations Don’t Tell You

If we were to rely exclusively on viral videos, we might conclude that robots are already on the verge of replacing humanity, folding our clothes, driving our cars, and perhaps even judging the moral quality of our life choices. The problem is that these demonstrations are, in most cases, the “cleaned‑up” version of reality. Viral clips are repeated hundreds of times until the robot achieves the “perfect” performance. They are edited so that failures disappear gracefully from the frame, and they are filmed in controlled conditions where not even dust seems invited. Under such ideal circumstances, almost any system can appear intelligent. Even a well‑programmed box would have a fair chance of looking “revolutionary.” But in real life, the world does not sign collaboration contracts with laboratory engineers. Out there, rain turns electronics into survival tests; dust enters precisely where it should never reach; mud ignores design specifications entirely; extreme temperatures never appear in promotional brochures; and unexpected obstacles—also known as “reality”—intervene without warning. And this is exactly where the less spectacular part of the story begins. Suddenly, the robot that danced flawlessly in a promotional clip starts behaving like an elegant tourist placed on a hiking trail without a map, without signal, and without a charger. It stops, recalculates, hesitates, asks for help (if it could), and sometimes simply gives up with a graceful system error. The irony is that these limitations are not hidden by engineers. They are known, documented, and often explained very clearly in technical reports. They simply do not look as impressive in a twelve‑second video with dramatic background music. The simple truth is that robots perform excellently… when the world behaves politely. That is, orderly, predictable, and without surprises—exactly like a laboratory. The rest of the time, reality has the impolite habit of ruining the presentation. Therefore, before declaring the beginning of an era in which robots “will do everything better than humans,” it is worth noting one minor but persistent detail: even the most impressive demonstrations still depend on conditions in which the real world is carefully kept at the back door.

Fear, Reality, and a Bit of Patience

The conclusion can remain simple, optimistic, and—above all—realistic. Yes, robots are evolving. Yes, they will reshape the economy. Yes, some professions will disappear or be radically transformed. But the gap between a robot that can move a box in a factory and one that could, theoretically, govern a country on its own is still enormous. Not a matter of “a little more progress,” but of “a great deal more.” The history of technology offers a healthy dose of realism. The first automobiles did not eliminate walking, no matter how enthusiastic the brochures of the time were. The first computers did not replace humans; they merely changed the way humans work. Similarly, today’s robots do not seem to herald total replacement, but rather a reorganisation of labour. The difference is that, this time, we have more spectacular packaging and a far louder media environment. Every technological demonstration is interpreted either as the beginning of a new era or the end of civilisation, as if there were no middle ground—no rather ordinary space called “gradual progress.” In reality, it is highly likely that in the coming decades robots will become our colleagues, not our masters. They will work alongside humans in factories, hospitals, logistics, and services—sometimes helping, sometimes hindering, but rarely replacing human judgement entirely. And if, in some distant future, they do aspire to “run the world,” their first obstacle may be surprisingly simple: a door that does not open properly, a cable left on the floor, or a step approached with excessive mechanical optimism. It may sound like a joke, but it is also a technical observation. Despite impressive progress, full autonomy in the real world remains a far more difficult challenge than it appears from the outside. Thus, between apocalyptic scenarios and uncritical technological optimism, reality sits somewhere in the middle: technology advances rapidly, but the distance between the ability to impress and the ability to dominate the world remains considerable—and, at least for now, perfectly visible.