factory

Elon Musk not only wants to reach the Moon, he also wants to turn it into a factory for satellites launched with catapults

by

What do they have in common ChinaSpaceX, Auriga Space and Electromagnetic Launch? Well, possibly several things, but one of them is that they have expressed interest in using mass cannons to launch from the Moon. Said to very roughlywant to use electromagnetic catapults loaded with satellites or construction materials to other points in space. Today, existing technologies are not optimized to launch large loads, but with so much power involved it would not be strange if it were finally achieved. The ethical implications would be many even if they only focused on science and communications. However, it could be even worse; Well, according to the analysis recently published by an expert, the arrival of the military applications It would only be a matter of time. From moon rocks to nuclear warheads. The idea of ​​the lunar mass cannon He initially proposed it a scientist named Gerard O’Neill in the 70s. His idea was to use them to extract minerals from the Moon and launch them into space to build space colonies with them. Over time, many space agencies, public and private, have become interested in its use. For this reason, the independent analyst specialized in cislunar security Andre Sonntag just published a report in which he recounts the risks that these catapults would entail. If the necessary technology is optimized to launch large payloads, they could be used to launch probes aimed at destroying satellites, inert projectiles and even ships loaded with nuclear warheads. Furthermore, if launches are made from the Moon they may be more difficult to detect by conventional early warning systems, so many attacks would go unnoticed. An ingenious design. Mass cannons are actually very interesting systems for space launch. They consist on a track in which one electromagnet is placed after another. A metal cart is circulated above and is attracted by said magnets. Each of these electromagnets is activated just when the car passes over it, giving it a new impulse, so that it accelerates more and more. The goal is to reach 2.4 kilometers per second, since this is the speed necessary to escape from the Moon. When this is achieved, the cargo on board the car is launched into space. In short, it is possible to catapult what is inside the car without having to consume propellant. Better on the Moon. This system has historically been proposed for use on the Moon for two reasons. To begin with, unlike what happens with a conventional rocketa very high speed is reached very quickly. If the process were done on Earth, the rocket would reach the atmosphere so quickly that it would catch fire due to friction when crossing it. On the other hand, there is no classical atmosphere on the Moon. On the other hand, since gravity on the Moon is much lower, a lower speed is needed than what would be needed to escape through this system on Earth. The case of Elon Musk. In his report, Sonntag has not mentioned any company or agency. However, it is well known that Elon Musk speak of the use of these mass cannons last February. He has been expressing his interest for a long time establish data centers in space and manufacture AI satellites directly on the Moon. This would avoid the energy, thermal management and launch logistics limitations that Earth poses for its ambitious plan. The vacuum of space would serve as a coolant and solar energy could be used to obtain electricity. The release waiting listIn addition, it would be much clearer than on Earth. In order not to have to carry large amounts of fuel to the Moon, their idea is to launch these satellites directly from our satellite. That is why he has already mentioned electromagnetic catapults on several occasions. There is legislation, but it is difficult to ensure. The United Nations Outer Space Treaty strictly prohibits the construction of military installations on celestial bodies. It also prohibits nuclear launches from space. However, Sonntag points out in his report that it would be quite difficult to verify. Therefore, he is concerned if these types of systems advance, given the interest of tycoons like Elon Musk. The payloads that could be launched with current technologies are minuscule. In fact, launching functional satellites directly with one of these cannons is science fiction. However, technology will advance. By then, we must be prepared, because those who have no scruples on Earth will not have them in space either. Image | SpaceX/xAI In Xataka | We knew there was water on the Moon, but not why some craters were empty. Finally we have the answer

In China they want humanoid robots to do household chores. The problem is that a house is not a factory

by

For years we have seen humanoid robots do somersaults, danceppractice martial arts or move through factories with increasingly striking capabilities. The next step seems almost natural: taking them home to do the laundry, prepare a bed or support elder care. The problem is that this transition is not as direct as it seems. A factory is designed to reduce uncertainty; A home, on the other hand, is full of small exceptions. And for a robot, those exceptions can be exactly the difference between a flashy demo and a useful product. The concept. SCMP account That GigaAI has introduced the SeeLight S1 as the country’s first general-purpose home humanoid robot model, developed in collaboration with the Hubei Humanoid Robot Innovation Center and the Hubei Humanoid Robotics Industry Alliance. In images released by the company, he appears performing very recognizable tasks: cutting vegetables, frying eggs, loading a washing machine, hanging clothes, making a bed or opening curtains. The company also plans to test it for free in homes in Wuhan in the first half of 2027. A house is not an assembly line. That is the fundamental difference. In a factory, the robot can work with known references, pieces always placed in the same way and movements that are repeated thousands of times with very few variations. In a home, on the other hand, nothing guarantees that the shirt is where it was yesterday, that the chair has not moved or that a pet does not cross in front of it just when the robot is trying to complete a task. Much movement, little understanding. Xinhua itself collects an idea that helps cool down the epic of the demonstrations and that does not only affect China, but humanoid robotics in general: humanoids have greatly improved in their “cerebellum”, the part linked to control and coordination, but they still have major problems in their “brain”. In other words, they can execute complex movements, but it is difficult for them to understand what a scene means and what function each object has within it. Home is also a data problem. Now, for these systems to work better in real homes, they need to learn from real homes, but the home is precisely one of the places where it is least easy to collect data. We are not just talking about room maps, but about objects, forces, angles, routines and physical decisions that are difficult to simulate. Advances and challenges. According to NSFCthe country expected to exceed 10,000 humanoid units sold in 2025, with a year-on-year growth of 125%, and there were already pilots in industrial manufacturing, delivery, catering and services. The important nuance is that none of this automatically turns this industrial career into a successful deployment within homes: the sector itself locates the path prudently, first industry, then logistics and commercial uses, and only later the home. A future easy to imagine, difficult to materialize. The difficult part is demonstrating that this can be done usefully, safely, and at reasonable cost outside of a prepared demonstration. There is the real border. China and other countries around the world can accelerate prototypes, pilots and production, but a home does not forgive clumsiness in the same way as a controlled stage. To get home, the robot will not have to understand human life better. Images | GigaAI In Xataka | In China there are already “schools” for robots. Its objective is the same as schools for humans: to teach them to work

space as a new factory

by

Two pharmaceutical companies have teamed up to launch an ambitious plan to synthesize drugs in space. It’s not an expensive hobby. It is more than proven that some medications They have added advantages if they crystallize under microgravity conditions. Until now, the few companies that had done so had worked alone or with the sole support of agencies such as NASA, but the fact that two of them are associated without the need for the space agency to intervene marks what could be the beginning of an era. Better to join forces. The two companies in question are Varda Space Industries and United Therapeutics Corporation. The first, founded by former SpaceX employees, has been synthesizing medicines in space since 2023. The second has never traveled beyond Earth, but it is a biotechnology company with enough potential for the union to be much stronger. The initial objective will be the crystallization in microgravity of drugs for rare lung diseases. However, in the future drugs could be produced for many other pathologies. It all started in 2019. In 2019, the companies Merck Sharp & Dome Corp. (MSD), in collaboration with the National Laboratory of the International Space Station (ISS), carried out experiments crystallization with the drug pembrolizumab (Keytruda). It is an anti-cancer drug that, as is common in chemotherapy, is administered by intravenous infusion, in a process that can last hours. By crystallizing it in space, a more stable form was obtained that allowed its administration in a single injection, making the treatment much more comfortable for patients. a matter of time. It has been seen that, when crystallized under microgravity conditions, many molecules assemble more slowly and constantly. The results are much more stable molecules that, once used as a drug, have a wide variety of advantages. For example, they dissolve better, do not require as much cold for storage, cause fewer side effects, and have a longer shelf life. Varda’s experience. The Varda company began its space pharmacology project in 2023. That year it launched the first of a series of unmanned capsules with chemical reactors into space. In these reactors, molecules crystallize and, after a few weeks or months of work, are returned to Earth. That first capsule was the W-1. W-6 is currently carrying out its mission and is expected to launch at least three more this year. Furthermore, after the merger with another pharmaceutical company, Varda is confident of being able to scale to 7 launches in 2027. Also for research. Molecules that crystallize in space give rise to larger crystals. This also makes your research easier. Therefore, with this type of project the aim is not only to obtain drugs. It is also expected to obtain candidate molecules to become medicines, to be analyzed more thoroughly by scientists on Earth. This is just the beginning. In the future, space travel will be much more widespread. The reuse of rockets will allow many more launches in less time, space tourism will become increasingly common and many public and private investigations can be carried out in orbit. If sufficient investment is achieved, the infrastructure to obtain drugs in space will become increasingly simpler. And, of course, the benefits for patients will also increase. Image | Varda/Magnific In Xataka | We knew that Mars has gravity. Now we have just discovered the unexpected effect it has on the Earth’s climate

20,000 million more will be spent on a factory with little water and labor

by

TSMC’s journey in Arizona (USA) continues. Yesterday the board of directors of this chip manufacturer, the largest on the planetapproved an injection of 20 billion dollars in what is already its most advanced semiconductor production plant of any it has in the US. The start-up of this factory It was full of setbacks.. In fact, it started production of integrated circuits almost a year late due to how much it cost TSMC. find qualified personnel that I needed. At the beginning of 2025 the first good news arrived. The plant had been producing semiconductors for several months at the N4 lithography node, which belongs to the 5nm FinFET family, and was ready to deliver to Apple the first batch of SoC A16 and SiP S9. This factory, known as Fab 21, made $514 million in profit last year according to Yeh Chun-Hsienthe minister of the National Development Council of Taiwan. This is not bad at all if we keep in mind that during the first year of operation, semiconductor plants do not usually deliver profits. In this scenario, the investment of an additional 20 billion dollars in the expansion of Fab 21, which is the purpose of this money, makes sense. In fact, this project is part of the $165 billion expansion plan that TSMC presented last year. However, not everything is going well for this company in Arizona. According to the newspaper Taipei Timesthe shortage of labor, and, above all, of water, is giving many headaches to the management leadership of this factory. And solving this last problem is not easy. Arizona’s water shortage is a huge challenge for TSMC Arizona is the second driest state in the US only behind Nevada. Semiconductor factories need a large amount of this resource, but it is not ordinary water like what comes out of our taps; They need a type of water almost impossible to find in nature. And its scarcity is getting worse. In fact, it is slowly becoming a systemic threat to the industry that sustains the artificial intelligencecell phones, electric cars and virtually any device that has an advanced chip inside. The water we are familiar with, such as that which comes from the tap, spring water, and even bottled mineral water, is full of impurities. It contains bacteria, dissolved gases, mineral salts and microscopic particles in suspension. This is not a problem for most of the everyday applications we usually use it for, but This water is not suitable for making chips. Even the slightest impurity invisible to the human eye is pure poison when involved in the production of cutting-edge semiconductors, such as the 2nm integrated circuits currently being manufactured by TSMC. The industry standard calls for water with an electrical resistivity of 18.2 megohms per centimeter The integrated circuit manufacturing process requires cleaning silicon wafers dozens of times. Every time a geometric pattern is transferred to wafers using lithography, they need to be cleaned. Also after pouring chemical reagents and photoresist fluids on them. However, the water used to remove any residue that may have deposited on the wafer cannot have the slightest impurity. It must be absolutely pure. In fact, the industry standard calls for water with an electrical resistivity of 18.2 megaohms per centimeter, which is the theoretical limit of water purity at room temperature. The problem is that producing ultrapure water is not easy. And it is not because it is necessary to subject it to reverse osmosis in multiple stages and ion exchange treatments. It is also necessary to degas it under vacuum, eliminate any microorganisms it may contain with ultraviolet light and filter it using membranes expressly designed to capture the slightest impurity. In this article we do not need to investigate these processes in detail, but there is something that we cannot ignore: this treatment consumes energy and requires the use of a large amount of chemicals. Furthermore, a significant part of the water that is processed is not transformed into ultrapure water, so it cannot be used. Once the water has been subjected to this demanding treatment, it acquires such a high purity that it becomes corrosive if it comes into contact with a very wide range of materials. Because it lacks its own ions, ultrapure water absorbs ions from virtually any material it comes into contact with. This is the reason why the pipes used to transport it must be made of materials immune to corrosion, such as PVDF (polyvinylidene fluoride), a fluorinated thermoplastic polymer similar to Teflon, non-polluting and extremely stable because it does not give up ions to ultrapure water. A single cutting-edge semiconductor plant consumes between 10 and 30 million liters of ultrapure water every day. This range is equivalent to the daily drinking water consumption of a city of between 50,000 and 150,000 inhabitants. Plus, there’s another challenge we haven’t looked into yet: ultrapure water degrades very quicklyso chip factories must have a very sophisticated production and distribution system capable of working in real time to deliver the ultrapure water required by the manufacturing process of advanced integrated circuits. Image | TSMC More information | Taipei Times In Xataka | Intel’s plan against an unattainable TSMC: beat Samsung and consolidate itself as the second largest chip manufacturer

Spain 1- Hungary 0. MG chooses Spain for its European electric car factory, according to Bloomberg

by

MG is one of the Chinese fashion brands in Spain. It is the company that dominates the market by sales volume, among other things thanks to its clear focus on the entry-level or mid-range. Its figures in the Spanish state rise like foam: In 2025 they sold 45,163 units and grew by 46.78%. In the midst of the global structural reconfiguration of the sector towards electric mobility enhanced by an EU that does not hesitate to try to regulate the market in favor of classic players located in its territory against the deluge in quantity, quality and price made in ChinaSAIC motor has made a move to protect its piece of the pie in the old continent: set up a plant to act as a European hub. Anonymous sources They have leaked to Bloomberg that the chosen destination is Spain. MG Choose Spain. SAIC Motor, one of the China’s largest automakers by volumeplans to establish its first electric vehicle production plant in Europe within Spanish territory. Anonymous sources linked to the sector they point out to Bloomberg that the decision is not yet final and that key details such as the size of the investment, production capacity or deadlines are missing. As explains The EconomistSAIC Motor has been studying options within the Spanish state for more than a year, which has experience in the deployment of electric vehicle platforms under the umbrella of government incentives. Why is it important. For SAIC Motor the answer is simple: investment is critical to mitigate the impact of the tariffs that the EU applies to Chinese manufacturers. By producing on European soil, MG escapes these taxes, which allows it to continue maintaining its aggressive pricing strategy, essential for its rapid growth. From a technical point of view, having a plant here would allow it to better adapt its models to the standards and tastes of the European consumer, in addition to reducing its logistical footprint. Spain It is the second European automobile producer and ninth in the world, with almost 2.3 million vehicles produced in 2025 according to ANFAC data. The arrival of a brand with the sales volume of MG represents traction for the entire automotive and associated industry, in the face of a panorama of forced transition from combustion engines and with an uncertain future of its most historic plants. That the industry shifts from fossil fuel engines to electromobility is key so that it does not lose relevance in the state GDP. Context. After eight months of investigation, Brussels determined that Chinese electric manufacturers receive state subsidies that distort competition and threaten millions of European jobs, so producing these cars on the old continent avoids those tariffs in the bud. The final tariffs approved by the European Commission applicable to BEVs imported from China are: BYD Group, 17%; Geely Group, 18.8%; SAIC Group, 35.3%, among others. To that we must add 10% base. Bottom line: SAIC Motor Corporation Limited is more interested than anyone in finding a solution. Despite this, they are selling cars like hotcakes: just look at any recent ranking of best-selling cars. Chinese brands have found two ways to avoid tariffs: establish factories within Europe or import vehicles in unassembled kits (CKD) to assemble them on European soil. SAIC aims to combine both phases. What Spain has that Hungary does not have. The eastern European country has until now been the favorite of Chinese manufacturers in terms of electric cars and batteries and there are no shortage of reasons to choose it: it offers lower labor costs, it has a geographically central position in Europe and direct links with China through the Belt and Road Initiative (Belt and Road). In addition, it already accumulates investments from BYD, CATL either NIOas well as other classic actors in the industry. Spain adds to its industrial muscle and its consolidated experience in the sector an advantage that Hungary does not have: a market where MG already leads sales and an ecosystem that is fully activated. In addition to having efficiency as a flag, in the words of the Spanish Minister of IndustryJordi Hereu. But there is a key milestone: the Figueruelas (Zaragoza) plant now has a date for make Leapmotor electric vehicles starting in October of this year, combining Chinese technology and Spanish production, consolidating a model in which SAIC can be reflected. Zaragoza is not only a precedent: it is proof that Spain can accommodate the Chinese electricity transition without industrial or political friction. All the pools point to Galicia. In the absence of official confirmation and knowing more details, the location for the future plant What sounds the most is Galicia: has extensive industrial experience linked to the sector, a consolidated chain of suppliers, the Salvaterra-As Neves logistics platform and the proximity to the port of Vigo, which opens the doors to maritime connections with the United Kingdom, northern Europe and Atlantic markets. The icing on the cake: the president of the Xunta, Alfonso Rueda, focused on his recent official trip to China in strengthening commercial relations in the automotive industry and in his busy schedule he had time to visit the SAIC factories and hold a meeting with the board of directors. First steps. Unveil Galicia Press that the model initially proposed for its Spanish plant points to CKD assembly, with the possibility of evolving towards full production according to sources in the sector. That is, the main components of the vehicle are manufactured at source and sent in pieces to the new factory for final assembly. Leo Zhang, head of SAIC for Spain and Portugal, has already spoken profitability of a possible European plant: from around 250,000 units per year. Although it is unknown which model they will start with (there are those who point to the utility vehicle MG2, direct rival of the Renault 5 and the future Volkswagen ID.2), if there is an orientation regarding deadlines: SAIC intends launch the first model manufactured in Europe in 2027. Yes, but. The news rests on anonymous … Read more

After visiting a Chinese factory, the CEO of Honda loudly admitted the noise of the industry

by

We are witnessing a great change in the automobile industry, led above all by the great presence of China in more and more global markets and a transition to electric which seems to still be difficult for him. The traditional automobile industry is going through a delicate point, and the president of Honda saw it clearly when visiting a supplier factory in Shanghai. The surprise. At the end of February, Toshihiro Mibe, president of Honda, visited the facilities of a large Chinese manufacturer of components in Shanghai. What he found was a completely automated plant, without workers on the production line, and capable of supplying parts to both Tesla and local builders, minimizing labor costs and operating constantly. “We have no chance against this,” counted Mibe when leaving, according to statements reported by the Nikkei Asia media. It is certainly not the type of statement that one would expect from someone who runs one of the most historic brands in world motorsport. Why does it matter? Honda is not an isolated case. It is the latest symptom of an industry that has been looking at China with concern for years. Chinese manufacturers have managed to compress the development time of a new model to between 18 and 24 monthsabout half of what the Japanese or Europeans need. And it’s not just speed: it’s cost, automation and software. It is a change that is costing the traditional automobile industry, and that is not easy to replicate either. Numbers. In 2020, Honda sold 1.62 million vehicles in China. In 2025, that figure fell to 640,000 units, a decrease of 24% in the last year alone and the fifth consecutive year of decline, according to data published by the media. Its factories in the country operate at 50-60% capacity, well below the 70-80% necessary to be profitable. By 2026, the planned production is less than 600,000 units. “It is an extremely disappointing plan,” acknowledged an executive from a Chinese supplier company to Nikkei Asia. “But it doesn’t surprise me either,” he continued. Honda is not alone in this. Jim Farley, CEO of Ford, warned in an interview with CBS Sunday Morning last October that China has enough production capacity to “supply the entire North American market and put us all out of business.” “Unless things change, we will not survive,” counted for his part, also the then president of Toyota, Koji Sato. And coming from Toyota, which is basically the largest automaker in the world, that says a lot. Vgo back to the past to go towards the future. Honda’s reaction goes through resurrect your R&D division as an autonomous entity, something that has already existed since 1960 and that in 2020 was dismantled in favor of centralized management. It was that independent structure that, in 1972, developed the low-emission CVCC engine (the first to meet US regulations) and turned the original Civic into a global success. Now, thousands of engineers return to a subsidiary with greater operational freedom. “Five or six years ago it was good for the headquarters to take the reins,” recognized a Honda executive to Nikkei Asia. “But now the world has changed drastically,” he continued. Doubts. The movement does not convince everyone. Takaki Nakanishi, chief analyst at the Nakanishi Research Institute, said to the media that “it is doubtful what will change just by restoring the organization.” Honda’s own management team admits that recovering the structure does not guarantee winning China. “But that doesn’t mean we’re going to raise the white flag,” added a company executive, according to Nikkei Asia. In parallel, Honda cancels two of its electric planned for the US, the 0 SUV and the 0 Sedan, and assumes losses of up to 15.8 billion dollars. Also have been left in the air the two vehicles under the Afeela brand, the joint project with Sony. The alternative bet: India. While Toyota and Nissan choose to ally with Chinese partners to learn from their speed and launch affordable electric cars, Honda prefers another path. The brand is betting on India as a manufacturing base for its next generation of electric cars. The Model 0 Alpha, its global strategic EV planned for 2027, will be produced there. In mid-March, the Indian subsidiary shared images of the Alpha in rolling tests, describing the moment as “a new milestone in Honda’s electrification journey.” Imbalance. The automobile sector is going through one of its most profound transformations. China has stopped being just a market to become the main global competitor, with brands like BYD already reaching 1.8% share in Europe in the first two months of 2026, according to data from the European Automobile Manufacturers Association (ACEA). Honda, with just 0.5% in the same period, illustrates this imbalance well. Cover image | Sling In Xataka | Sensors, luminous tires and fish scales: the crazy (and stinky) story of the first “autonomous” car

has tested its humanoid robot in a real factory (and there is video)

by

For years we have heard the same promise: humanoid robots working side by side with us in factories, warehouses or even in our homes. It’s an idea that appears again and again. However, when we go down from that stage to the ground of a real plant, the story changes quite a bit. There it is not enough to walk or grab objects; everything must happen precisely and be repeated many times without errors. In that context, each small advance begins to have a different meaning. The latest news from Xiaomi. Lei Junfounder, president and CEO, posted a message on his official WeChat account to update the status of the company’s robotics project. The executive explains that a humanoid robot developed by the company has already begun “doing practices” within one of its automobile factories. The manager also links to a published technical article describing the first tests carried out with the robot under factory conditions. Let’s see. What exactly has the robot done in the factory. According to the text, The humanoid robot has been put to the test in a very specific position within the automobile manufacturing process: placing self-tapping nuts on parts of the vehicle’s floor. In practice, the system collects these nuts from an automatic supply equipment and deposits them in the positioning tool where the automated screwing of the position is then carried out. The Chinese firm places this operation in the pressure casting workshop, on ground components after that process. Three figures that help understand the test. Xiaomi explains that the humanoid robot performed this work for three hours of continuous autonomous operation within that position. In that period, it achieved a success rate of 90.2% in the simultaneous placement of the nuts on both sides of the piece, a percentage that the company defines as the number of correct operations compared to the total number of attempts made. Another fact that stands out is the work pace, since the system managed to adjust to a production cycle of up to 76 seconds. This is an important fact because in an industrial line, each operation must fit into very specific times so that the process does not break down. behind the scenes. Xiaomi points out that its humanoid robot is based on the Xiaomi-Robotics-0 model, described as a VLA-type model that integrates vision, language and action within the same system. According to the company, this approach makes it easier for the robot to understand the tasks it must perform, perceive its environment and execute the movements necessary to complete them. The training is also complemented with reinforcement learning, a technique that allows the system to improve its behavior based on the experience accumulated in the physical world. The faults that the robot can find on the line. In its technical description, Xiaomi also points out several scenarios in which the operation may fail. One of the main problems appears during the alignment process between the self-tapping nut and the positioning pin, which must be well centered and seated before screwing can proceed. If this fit is not precise enough, a blockage may occur during the process and the assembly remains incomplete. Additionally, the orientation of the nut inside the robot’s hand can vary with each grip, and the company cites factors that complicate adjustment, such as the knurled structure inside the nut, the magnetic attraction force of the pin, and, in some cases, environmental interference or working angle limitations. The predecessor. To better understand this advance, it is worth remembering that Xiaomi has been exploring the field of humanoid robots for some time. In 2022 the company presented CyberOnea prototype that appeared at one of its events showing basic capabilities such as walking or holding objects. At that time the company itself made it clear that it was a project in an early stage of development. What we see now seems to be situated in another type of scenario: less demonstration on a stage and more tests within a plant, where the objective is to check if these machines can respond to the demands of a repetitive process. Looking to the future. The company also hints that this experiment is just one part of a larger project. Xiaomi points out that it is testing its humanoid robots in various jobs within the factory, including box transport tasks and operations related to the installation of exterior elements of the vehicle. In fact, in his WeChat post, Lei Jun states that the company wants to contribute to the deployment of humanoid robots in smart manufacturing and proposes a medium-term forecast. According to his estimate, in the next five years there could be large quantities of these machines working in his factories. Images | Xiaomi In Xataka | Huawei presents its AI supercluster to the world: it is a nod to Chinese Big Tech and a message to NVIDIA

RAM is in an “unprecedented” crisis. So much so that even Tesla is considering opening its own memory factory

by

Neither technological advances nor a revolution in devices: crises are what is defining the last years of the sector. He veto Huaweithe semiconductor crisis of 2020 and now, the RAM memory crisis. The difference between this crisis and the previous one is that, although the 2020 crisis was caused by a perfect storm, the RAM memory crisis is being caused by excessive interest in data centers and AI. And it is taking all sectors ahead. That there is no RAM memory for consumers is a symptom, but it implies something much bigger: although the main producers are investing millions to increase your RAM productionit is not memory for consumption, but for GPUs and data center systems. Only a few companies dominate the production of these chips, and if they cannot produce them, they do not produce the memory chips for SSDs –raising the price-. They dedicate all production to meeting the demands of AI. And, as we read in FortuneElon Musk, one of the owners of some of the largest data centers on the planethas shown that there are two ways to face this crisis: hitting the wall or taking action. And the translation is that Tesla is considering building its own RAM factory. The problem is that it is easier said than done. Tesla and Intel interested in biting the RAM biggies In recent weeks, some of the world’s leading companies have presented results and RAM has been the central topic. PlayStation, for example, has assured that they are very aware of their ability to continue manufacturing PS5 with the goal of not going upagain, the price. And NVIDIA has been stating for days that it needs TSMC – its main chip supplier – and Samsung – who provides them with new generation HBM4 memory – get the batteries. Meanwhile, the outlook is not good. own NVIDIA aims for seven or eight years of construction no brake on data centers. Intel assures that The crisis will extend beyond 2028 and Micron, one of the big three in DRAM memory, has cataloged the market bottleneck as “unprecedented.” In this technological tsunami, and during Tesla’s results presentation at the end of January, Elon Musk pointed out that the company could need to build your own memory manufacturing plant. The objective is the one that all companies have: ensure supply. Going from scratch to manufacturing RAM memory is easier said than done, however, here Tesla has an advantage: they are not new to chip manufacturing. Although they abandoned the project for a few months, at the beginning of this year Musk himself stated that They came back with their own chip for your data centers. Additionally, there is the fact that they are a company with enough muscle to create a clean chip manufacturing room next to some of its existing plants. Intel is another one looking to become one of the important voices in the RAM conversation. Together with the Japanese giant SoftBank, they are developing an evolution of stacked DRAM memory that have been baptized as ‘ZAM’ and that seeks to break the HBM memory monopoly of Samsung, Micron and SK Hynix. Now, things in the palace are going slowly, and if Intel (which is already in it) It will take between three and four years to have commercial productsTesla’s ambition may go into the next decade. Let’s hope we don’t continue in this crisis by then, but if more “players” are interested in producing RAM, it would mean that, in the event of subsequent crises, there will not be a few that dominate the sector, producing a bottleneck like the one we are experiencing. Domino effect of the AMR crisis and China taking action Because this is not just about RAM being more expensive for users: it goes much further. If companies do not have the capacity to satisfy the demand for AI, they pour all their manufacturing muscle into a single task, neglecting the others. This explains the rise in the price of SSDs, but also of other products that should not have a leading role in this conversation: hard drives or HDDs. It is a brutal domino effect because, as we say, it goes beyond the modules being more expensive: RAM is more expensive for companies and that implies mobile phones or more expensive or with less RAMconsoles that increase in price (like what is happening posing for nintendo switch 2), machines that are late and they will be more expensive (like the Steam Machine), car problems and even impacting the routers. And in this scenario, in which companies like Intel or Tesla are considering taking a bite out of the RAM sector, we have some Chinese companies that had no role in the conversation. positioning itself as an option to alleviate demand. We told it a few days ago: there were reports indicating that PC brands such as Asus, Dell or HP were considering purchasing memory from Chinese manufacturers such as CXMT. Their modules are not as advanced as those of Samsung, for example, and they do not have the production capacity of South Korean companies, but… they produce. And in lean times, that’s better than selling laptops without RAM. Anyway, as we have said on occasion, there are still more companies joining the production of RAM when the crisis has already had a full impact, but the goal is not to create more RAM for ourselvesbut for your data centers. It’s time to entrust ourselves to the most sacred thing: that our PC doesn’t break and we need to update. Images | Gage Skidmore, Intel In Xataka | The US has a problem with its AI data centers: more and more states are opposed to building them

Germany does not want to depend on Elon Musk for war. So the largest weapons factory in Europe wants a “military Starlink”

by

For decades, European security has rested on critical infrastructure controlled from the United States. But with the war back on the continent and space communications becoming a decisive military assetGermany is beginning to assume that it cannot afford depend on Elon Musk nor from Washington for something as basic as talking and fighting in case of conflict. A “military Starlink”. Rheinmetall and OHB are in preliminary talks to present a joint offer to create a satellite communications network in low orbit for the Bundeswehr, a system that in Berlin already is openly described as a “Starlink for the German army”. The initiative aims to capture part of the ambitious German plan for invest 35,000 million euros in military space technology, with the aim of providing a secure, sovereign infrastructure specifically designed for military use, reducing dependence on US services such as Starlink, owned by SpaceX. Technological sovereignty. The background of the project will be one of the great themes of this 2026, and it is both strategic and political, since the war in Ukraine has shown to what extent satellite communications in low orbit can be decisive when terrestrial networks are destroyed or degraded. Although Starlink (and its military version Starshield) became in a key asset for kyiv, many European countries distrust to base critical capabilities on a foreign private provider, which has accelerated plans to build national or European networks under state control. The weight of Germany. With this program, Germany aims to become the third largest investor world in space technology, only behind the United States and China, according to the consulting firm Novaspace. German military authorities have already defined the technical specifications and are preparing the tender, prioritizing coverage of NATO’s eastern flank, where Berlin deploys a permanent brigade of 5,000 soldiers in Lithuania as part of its defensive reinforcement. From armored to space. Traditionally associated with tanks, artillery and ammunitionRheinmetall is rapidly expanding its presence into new domains in the heat of German rearmament. At the end of last year it obtained its first major space contract, up to 2,000 million eurosto develop together with Iceye a constellation of radar satellites capable of operating at night and in bad weatherwhich puts it in a solid position to now aspire to a military communications system in low orbit. HBO and opportunity. For HBOthird largest European satellite manufacturer and navigation system supplier Galileothe project represents a key opportunity to strengthen its military business. The company faces the possible creation of a European space giant as a result of the merger of the divisions from Airbus, Thales and Leonardoan operation that its CEO considers potentially anti-competitive and that could leave OHB at a disadvantage if it does not expand its scale and capabilities. Boiling market. The simple announcement of the talks has OHB price skyrocketedreflecting the extent to which the sector perceives German military space spending as a catalyst for opportunity. That said, the project is still in an early phase, with no official comments from the companies or the Ministry of Defense, and is part of a growing competition for multi-million dollar contracts that will define who controls future critical military communications infrastructure in Europe. Image | Support Forces of Ukraine Command In Xataka | Germany is experiencing a new “industrial miracle” that it already experienced 90 years ago: that of weapons In Xataka | Europe’s largest arms factory faces an unexpected problem: earning an indecent amount of money

They kicked him out of the factory in 2020. Today, this tiktoker sells his company and an AI that replaces him for 900 million

by

Khaby Lame, the tiktoker with the most followers in the world (160 million), has sold his company Step Distinctive Limited to Rich Sparkle Holdings for between $900 and $975 million. The operation was closed in January with shares, without cash, and represents one of the largest transactions in the new creator economy. Who is it. He Italian of Senegalese origin25, rose to fame after being laid off from his job as a machinery operator in the pandemic. He began to publish silent videos, favored for their comical expressiveness, where he dismantled one of the fashions then in vogue on the Internet: life hacks that he knew how to detect as ridiculously complicatedand whose artificiality he exposed with a gthis distinctive that became his trademarkand that helped him propose obvious solutions to problems that, in reality, did not exist. On June 22, 2022, she surpassed Charli D’Amelio as the most followed person on the platform. What does the deal include? Dubai-based Step Distinctive Limited handled licensing, partnerships and sales. Lame becomes a shareholder of Rich Sparkle Holdings but loses operational control. A key fact: according to Celebrity Net WorthLame only had 49% of his company. 51% belonged to partners such as the Chinese Anhui Xiaoheiyang Network Technology, of the Three Sheep conglomerate. Lame’s case is not isolated. Mergers and acquisitions in the creator economy grew by 73% in 2025, reaching 52 operations. The sector, valued at more than 200 billion in 2024could exceed one billion before 2033. Other notable cases: MrBeast’s Beast Industries was valued at 5 billion and earned 473 million in 2024. Logan Paul earned 1.2 billion with Prime in 2023, valued between 2,000 and 3,000 million. The formula pursued is obvious: convert your followers into buyers. Digital twins. The most striking thing about the agreement is the transfer of the digital twin made by Lame’s AI. This technology creates digital replicas capable of speaking different languages ​​without recording new content. In China, platforms like Douyin employ AI-generated streamers who sell 24 hours a day, reducing costs by 80%. This technology allows a person to “work” simultaneously in multiple markets without a break. Who buys? Prior to the deal, Rich Sparkle was a financial printing company with no history in social media or AI, and the deal raises questions about the financial effectiveness of these types of deals. Now, for three years, Rich Sparkle has exclusivity over Lame’s business operations but despite his fame, is it a good move? Creators build value with their identity, but when they are controlled by outside corporations they risk losing what made them unique. The creator economy is no longer marginal: it has become a sector that operates with the same amounts as traditional industries. What started in March 2020 with an unemployed worker posting videos has ended five years later in a nearly billion transaction involving AI and Chinese conglomerates. But… is such a purchase capable of maintaining the spontaneity and freshness that characterized Lame? In Xataka | TikTok has dodged the bullet of the US veto. Although it has not been free

Log In

Forgot password?

Forgot password?

Enter your account data and we will send you a link to reset your password.

Your password reset link appears to be invalid or expired.

Log in

Privacy Policy

Add to Collection

No Collections

Here you'll find all collections you've created before.