Raw LLM Responses

The speed of data center expansion around the world is simply fantastic. It is estimated that in 2015 there were 8,500 data centers worldwide. This number grew to 11,800 operational data centers by 2025. By 2030, there will be 15,000 data centers. They consume immense amounts of electricity and cooling water. Some of them depend fundamentally on energy generated by fossil-fueled power plants. The environmental problems caused by data centers, which are increasingly larger and have more computers, are immense: pollution, lack of drinking water for the population, and, of course, noise. In general, the press pays a lot of attention to Big Tech propaganda about the benefits of technology investments and pays little attention to the communities affected by data centers. The negative externalities of this sector of the economy must necessarily extend to the need for raw materials for the manufacture of microchips and other electronic components, conventional and optical cables, etc. The same can be said for mobile electronic devices (smartphones, tablets, drones, robots, and laptops) that users will inevitably use to connect to the internet, producing data that will circulate and be stored and analyzed in a data center. Old copper mines are being expanded. New mines for various other materials are also being developed. Rare earths are being assiduously sought, even in nodules on the seabed. The appetite for this sector of the economy is likely to expand even further now that new AI-powered weapons, autonomous or not, are beginning to be developed and incorporated by the armed forces of the richest countries (and the poorest that can afford them, too). These weapons rely on microchips and geolocation devices, and are and/or will be connected in real time to data centers that provide services to the military. All of this costs money. Much of the internet infrastructure is financed with private money. But public funding is important, more important than most people realize. A greater allocation of public resources to this branch of the economy (including for reasons of national security) means less money for investment in other critical areas with a major humanitarian impact (health, education, unemployment insurance, housing, public transport, social welfare and public pensions, for example). In the foreseeable future, in the event of war, the enemy's electrical, communications, and internet infrastructure will be legitimate targets. Enemy data centers will also have to be destroyed, whether owned by public or private institutions. The war will be fought both in cyberspace and in the real world, because supercomputers will be as valuable, if not more valuable, than the weapons factories they direct and/or control in real time. The race to the bottom has already begun. This can be seen in the enormous size of the US military budget. China and Russia have also increased their investments in this area. Below is a table provided by Perplexity:: Summary table of estimates for 2025 (approximate values) Region Estimated investment Key details USA US$50-90 billion Military AI, drones, data centers, digital infrastructure China US$30-70 billion Robotics, AI, drones, military cloud, chips Russia US$8-15 billion Military AI, robots, autonomous systems, exports Europe US$35-60 billion AI, drones, startups and military infrastructure Brazil will also invest in this area, but the amounts spent (estimated at less than US$100 million in 2025) are more modest. Irrelevant considering the data mentioned above. More sophisticated weaponry has never been synonymous with military victories. At the height of their military, industrial, and economic power, the Americans were defeated by the South Vietnamese in the 1960s. Two decades earlier, the poorly armed Chinese communists had defeated the sophisticated and well-armed Imperial Japanese Army using guerrilla warfare. Russia suffered a terrible defeat in Afghanistan in the 1980s, but the Afghans received thousands of modern, "made in the USA" surface-to-surface and surface-to-air missiles that nullified the supremacy of Russian attack helicopters and tanks. The future of war is always uncertain. But the future of societies embarking on the adventure of total digitalization is predictable. The new barbarians who attacked the Empire are not outside it. They will be created by the contempt that American and European politicians show for the basic needs of their populations. The capture of the state by arms manufacturers was a problem. This problem becomes even greater now that they are joining forces with Big Tech, whose appetite for public funding seems insatiable. What will be the response of Western states to the internal conflicts generated by reduced investment in essential humanitarian areas? More police repression and increased digital surveillance using AI (this is already happening in the US, England, and Germany). This is ironic. American and European governments and the press love to criticize the Chinese model (described as an example of AI-empowered totalitarianism), but the truth is that it is becoming increasingly difficult to see any difference between East and West. The only difference, perhaps, is that, unlike its adversaries, China continues to allocate considerable sums of money to healthcare, education, unemployment insurance, housing, public transportation, social welfare, and public pensions. In this context, the desire to bring about "regime change" in China is nothing more than a childish dream. The most likely scenario is a "regime change" in the US and Europe, generated internally by the contradictions that Western leaders themselves are creating by failing to care for growing segments of their populations at a time of structural unemployment and low-paying precarious work. Donald Trump may even send the Army to remove the poor from Washington, DC, but that certainly won't reduce poverty in the US. However, in addition to the low-paying, bullshit jobs familiar to Brazilians, Europeans, and Americans (UBER drivers, app delivery drivers, Amazon inventory handlers, etc.), there are monotonous, lower-paying, remote jobs. These are the jobs of invisible African, Venezuelan, and other workers who perform microtasks like correctly identifying photos and merchandise, moderating abusive, grotesque, or violent content, etc., to train AIs or reinforce their machine learning. There's a good DW documentary on this subject. These miserable workers, barely able to earn a few dollars a day to buy food, are the modern version of the slaves who worked in and under the Roman Baths. They too were indispensable and invisible, forced to perform arduous, oppressive, and tiring tasks without being able to enjoy the benefits of those who frequented the Baths daily. One of the most remarkable things about Roman history is that it provides a well-documented model of how an Empire was slowly corroded by internal contradictions until it became so weak that it was eventually decapitated by barbarian hordes. After Rome was invaded and sacked, the provinces did nothing to save the Empire or to regain control and administrative centrality of the imperial capital. The provinces of the Western Roman Empire simply went their own turbulent ways. New waves of barbarians exploited the vulnerabilities of the former Roman provinces. Conflicts between masters and slaves multiplied and were exploited by the invaders. And at some point, the major Roman cities began to fall into ruins. Many of them were completely or partially abandoned. Something that strikes anyone who observes the Roman legacy in Europe, and even in Italy, is the grandeur of the surviving aqueducts, temples, bridges, and gladiatorial arenas. Many of these structures remain nearly intact. Others are in a deplorable state, which is understandable. But it is surprising that not a single Roman Bath has been preserved intact. They were all destroyed. And there were many of them. Every Roman city had Baths. In large urban centers, there were two or three. The Baths were the most visible symbol of Roman opulence and oppression. Within one of them, Roman citizens had access to several distinct environments (a steam room with heated floors, a pool with water heated to room temperature, a pool with cold running water, a gym, a locker room, etc.). The slaves who kept everything running, especially those who stoked the fires that heated the boilers and maintained the underground heat transmission systems, lived in hell. But they knew that above them and within the Baths, a paradisiacal luxury prevailed. We can assume that these slaves intensely hated their workplace and, when the opportunity arose, they went to great lengths to destroy it. But this is only a plausible hypothesis. Another hypothesis that may ultimately prove true is the destruction of data centers in the event of a civil war. Big Tech is already intensely hated. And they will become even more hated in the coming decades. When the opportunity arises, will data centers be invaded and vandalized by the barbarians the West is creating? It remains to be seen. One thing is certain: they are the most obvious symbol of the uncivil civilization being created by data barons and AI owners.