Mobile Tech & Gadgets

from uranium to the plug, step by step

by

Do you remember Homer Simpson asleep in front of the control panel? For years, that has been the most popular image of a nuclear power plant: glowing bars, red buttons and donuts. Others, however, may think of sirens, black smoke, protective suits and names that continue to weigh: Chernobyl or Fukushima. Between fiction and collective fear, there is a much more normal story—and at the same time more amazing—that usually goes unnoticed: that of giant factories that produce electricity from the power of atoms. If you approach one, you will see towers that seem to breathe water vapor. And inside, hidden behind a heart of steel, millions of atoms splitting in two, releasing energy so enormous that a handful of uranium is enough to power a city for days. Although the debate is served with this type of fontthe truth is that it continues to be one more piece of the energetic present. So, leaving prejudices aside, let’s take a look inside a nuclear power plant: to discover how it works, how it differs from a thermal one, how many are still active in Spain and why it remains at the center of the energy debate. What is a nuclear power plant? A nuclear power plant is an industrial facility designed to produce electricity. At its core—literally—is the nuclear reactor, the place where the magic happens: the fission of atoms. Inside each atom there are protons and neutrons that remain united. When that nucleus breaks—when hit by a neutron—an enormous amount of energy is released in the form of heat. That’s where nuclear energy comes in: the same energy that holds those tiny particles together. Nuclear power plants take advantage of this nuclear fission process to obtain heat, heat water, produce steam and move turbines that generate electricity. It’s that simple. Or, if you look closely, that impressive. Difference between a nuclear power plant and a thermal power plant Confusion is common: “Aren’t a nuclear power plant and a thermal power plant the same thing?” In part, yes. Both use heat to drive a turbine and produce electricity. But the big difference is in the origin of that heat. In a thermal power plantthe heat comes from burning fossil fuels (coal, gas or fuel oil). This releases carbon dioxide (CO₂) and other polluting gases. While, in a nuclear power plant, heat is obtained from the fission of uranium atoms, without combustion or CO₂ emissions during electricity generation. Therefore, nuclear They are considered clean energy in emissionsalthough they leave a different challenge: what to do with radioactive waste? We could say that it is a smokeless energy, but not without questions and I will stop here because we will talk about it at the end. How it works: the process to generate electricity It may sound complicated, but the operation of a nuclear power plant can be explained in a simple way: Imagine a big kettle, like a teapot, only inside there are atoms splitting and releasing energy. Uranium fission. It all starts inside the reactor. Uranium-235 atoms break apart when hit by neutrons. Each fission releases heat and more neutrons, which continue colliding with other atoms, creating a controlled chain reaction. Water heating. The heat produced is used to heat water. This water circulates through pipes under enormous pressure or is transformed directly into steam, depending on the type of reactor. The steam drives the turbine. The force of the steam rotates the blades of a turbine connected to an electrical generator. That movement is what is finally converted into electricity. The electricity is sent to the grid. The generator converts the mechanical energy of rotation into electrical energy, which is transported to homes and industries. Cooling and recirculation. The steam condenses, cools, transforms back into water and returns to the circuit, repeating the cycle. It seems simple, and it is in concept. But behind it there are decades of engineering, thousands of security measures and constant surveillance so that this invisible and powerful energy is always kept under control. In Spain There are two types in operation: the pressurized water reactors (PWR)where water is heated inside the reactor and converted to steam outside, and the boiling water reactors (BWR)where steam is generated directly inside the reactor. How many nuclear power plants are there in Spain? According to the Ministry for the Ecological Transition and Demographic Challenge (MITECO)Spain has seven nuclear reactors spread over five sites: Almaraz I and II (Cáceres). In operation since 1981 and 1983, with a combined power of about 2,000 MW. It is one of the first that is on the list for closure: Almaraz I in 2027 and Almaraz II in 2028. Ascó I and II (Tarragona). Connected to the grid in 1983 and 1985, they total about 2,000 MW. Its closure is scheduled for 2030 Ascó I and 2032 Ascó II. Chests (Valencia). In operation since 1984; It is the only one with a boiling water reactor (BWR), with 1,092 MW of power. Its closure is scheduled for 2030. threshing (Guadalajara). In operation since 1988, with a power of 1,066 MW. It is scheduled to close in 2035. Vandellós II (Tarragona). In service since 1988, with a power of 1,087 MW. It is scheduled to close in 2035. In addition, there were three others that are already closed: Jose Cabrera (Guadalajara), the first Spanish nuclear power plant. Santa María de Garona (Burgos). Vandellós I (Tarragona), closed after a fire in 1989. In total, Spanish operational reactors generate around 20% of the country’s electricity, according to data from Nuclear Forum. And they do it constantly, 24 hours a day, without depending on the sun or the wind. What is the largest nuclear power plant in the world? If nuclear power plants had their own world ranking, Japan would be in first place. The central Kashiwazaki-Kariwa It has seven reactors and a power that exceeds 8,000 megawatts. Today it is stopped for revisions, but it is still the largest on the planet. The center follows … Read more

In Galicia some parents educated their young son at home. Now they are condemned for “irresponsible unschooling”

by

He homeschooling It’s news. And it is on account of a sentence issued by a court in Vigo that has decided to impose a fine of just over 2,000 euros (with severe reproach included) on parents who decided to educate their nine-year-old son at home, removing the little one from the school in which he was enrolled for the 2024-2025 academic year. The ruling is interesting not so much for its consequences (the penalty is not high: 1,080 euros per parent) as for its arguments and because it differs from other rulings on the same topic that yes they were acquittal. One moment,homeschooling? The term may seem strange, but it is not new. In fact, it connects with a movement that started in the US in the 70s. He homeschooling It is neither more nor less than an educational option that advocates educating children at home, far from conventional classrooms and schools. Often focusing educational responsibility on parents and those who practice it stand out above all its ability to adapt to the needs of each child, its personalization and flexibility of schedules, content and spaces. How many people practice it? Hard to know. In Spain it is estimated that there are between 2,000 and 4,000 families unschoolers. If we talk about the United States, there are calculations that indicate that 3% of students between five and 17 years old receive training at home. The variety of data is explained by the lack of censuses (case of Spain) and above all because the practice does not have the same lace in all countries. To better understand the regulatory differences, it is good to take a look at the web of Homeschooling. There are nations that clearly prohibit it, others that protect it and then there are cases like Spain, where there are those who consider that home education moves in “a legal ‘gray area’”. Swampy terrain. That is why sentences like the one just handed down by a judge in Galicia arouse so much interest, especially because not all cases end in the same way. What have they judged in Vigo? What the Criminal Court number 1 of Vigo has ruled on is a very specific case: some parents from Gondomar (in Pontevedra) who decided to take their nine-year-old son out of the public school in which he was enrolled last year to educate him at home. The ruling recalls that, despite the Educational Inspection’s resolution denying deschooling and the warning from the Juvenile Prosecutor’s Office, the parents continued with their plans, betting on homeschooling and a personal itinerary. The boy attended an academy for two hours a week, but the bulk of his education depended on the program decided by his parents. And what does the sentence say? The fundamental and key of the sentence, as precise Vigo Lighthouseis the educational bet chosen by parents. The ruling speaks of “flagrant deficiencies”, an educational project that “does not meet the minimum requirements” and “irresponsible unschooling” that compromises the child’s “academic progress” and, ultimately, will condition his life. “In this case, the education provided directly and almost exclusively by the parents, basically based on their own personal criteria and ideas, without an alternative educational method to the minimally solvent official one and without some evaluative objectivity, represents irresponsible unschooling,” the judge warns in his sentence, disclosed by Lighthouse and with a resounding tone. “There is negligence in education and failures in basic care duties.” Why is it important? Where you put the focus. The sentence focuses mainly on these shortcomings and the “irresponsibility” of the child’s family. In fact the ruling insists in which the parents did not even resort to an “alternative external educational system to the official one” and selected the subjects based on “their own criteria”, without any other reference. In practice this translated into training that, in the judge’s opinion“does not meet the minimum requirements established in the regulatory framework of compulsory education.” The minor participated in activities such as bicycle outings, excursions to the forest, sailing with a jet ski, collecting chestnuts or cooking, but he received “basic” skills in fields such as mathematics or language. For example, his parents did not teach him geometry. The expert who was in charge of examining the case in fact noted a “confusion” between the family routine and the school routine and also pointed out the “privatization of socialization” of the child. Does it explain anything else? Yes. The ruling conveys an interesting message. He explains that “home education may not be criminally reprehensible,” but it must meet a series of requirements, guaranteeing that the child will receive “sufficient” training thanks to an educational system that must be “responsible and competent.” In fact, this is not the first ruling issued by the Vigo court on the subject: at the beginning of 2024 spoke about another case involving parents who educated their son at home throughout the 2021-22 academic year. The Prosecutor’s Office found a crime of “family abandonment” (the same one that has been tried now) and requested five months in prison for the parents and six months of disqualification for parental authority. What was the result? On that occasion the judge acquitted them. Although he homeschooling was underlying in both cases, the judge noted clear differences in both cases. In the 2021-2022 report, it concluded that the parents did not show “carelessness” or “carelessness” and carried out “responsible unschooling.” The little boy continued to be educated with official books, he attended various extracurricular activities and the homeschooling It only really lasted one course. In fact, a year later the minor was already attending another center again, where he continued training “with absolute regularity” and obtained good grades. Is it legal or not legal? Europa Press assures that the ruling of the Court of Vigo warns that the homeschooling It is not a legal option in our country and it is “unquestionable” that parents who do not send their children to school violate an obligation contained in the … Read more

the light of its streetlights and the posters

by

Living in a big city is associated with living with a large amount of artificial light. LED street lightsscreens, signs, traffic… great light pollution that is undoubtedly one of the most visible traces of urban progress. But this is something that not only affects biodiversity or sleep, since a study presented in the American Heart Association Scientific Sessions 2025 suggests that it could also be linked to a higher risk of cardiovascular disease. Currently the towns are being left desolate, and the citizens are They are concentrating in the big cities because of the great job or training opportunities they have. But there are always different drawbacks, the noise being possibly very annoying (especially for sleeping and resting), but light is also something to explore. A few weeks ago the topic of the moment was undoubtedly the time change and why experts point out that winter time is the best despite the fact that it gets dark really soon. A debate that focuses on our circadian cycles and sunlightbut we must also take into account the great exposure to artificial lights that we have, especially in large cities. The study. The research, developed by the team of cardiologist Shady Abohashem, from Massachusetts General Hospital and Harvard Medical School, analyzed data from 466 Boston adults without active cardiac pathologies. The researchers crossed their brain scans (PET/CT) with satellite images of urban night glow from the New World Atlas of Artificial Night Sky Brightness. The result was quite clear in this case: the higher the levels of artificial light at night where they lived, the greater the stress-related brain activity and the greater the inflammation of the arteries. Two key indicators of cardiovascular risk, which undoubtedly set off all the alarms. If we put it into concrete figures, the risk of developing cardiovascular disease after five years is 35% higher than that of people who live in large cities. But looking at ten years, this risk stands at 22%. Its mechanism. In order to understand the reason for this data, we have to go to the brain. When this organ detects light at night (a time when in theory there should be complete darkness), the hypothalamic stress axis is activated. This response, maintained over time, can cause inflammation throughout the body and especially atherosclerosis, as has been shown in studies on chronic stress and cardiovascular disease. This phenomenon, Abohashem suggests, would explain how a seemingly harmless environmental factor could be integrated into the cardiac risk equation. Julio Fernández-Mendoza, expert in sleep medicine and author of the recent scientific statement of the American Heart Association on circadian and cardiometabolic healthsums it up like this: “Artificial light at night suppresses melatonin and disrupts the internal clock. This alters blood pressure, metabolism and endothelial function. This new study shows how this alteration can be observed even at the brain and arterial level.” More evidence. But this study is not isolated, since the idea is not new and there is already literature that supports it. For example, a cohort study with more than 400,000 people in South Korea already observed that living in brighter areas increased the risk of heart attack and stroke. The same thing happened in China with a study that detected that prolonged exposure to urban light increased the incidence of coronary heart disease in older adults. Besides, a published review in Sleep Medicine Reviews explains how artificial light can alter melatonin secretion, modify the circadian rhythm and activate low-grade inflammatory reactions in humans. In the end all the ingredients for us to expose ourselves to an alteration in our system. What can we do? Although the new work is observational and has not yet been peer-reviewed, its authors suggest concrete measures that could have a real impact: reducing unnecessary street lighting, implementing motion sensors in residential areas, choosing warm tones (less blue) and, at home, keeping bedrooms dark and free of screens before bed. In the end the idea is the one we have always repeated: maintain good sleep hygiene. Something that is achieved by avoiding looking at your phone minutes before going to sleep or even separating dinner from bedtime in order to maintain the best conditions for our brain. Images | Nao Ya DESIGNECOLOGIST In Xataka | Science has been trying to figure out the best time to go to sleep for decades. And you are already close to an answer

When we thought we had seen all kinds of rehearsals for an invasion, China makes science fiction: robots taking over an island

by

At the end of 2024, several military studies from Beijing were published outlining six different scenarios if future unification with Taiwan goes awry. So we tell that the Second World War I advised against all them because, in essence, there was talk of an invasion of the island. From then until now so much China as Taiwan have carried out all kinds of drills under the war scenario background. What you haven’t seen until now is that China has a plan B: robotic wolves. Mechanized herds. This week and through images and videosChina has shown to the world a new generation of autonomous combat systems in an exercise that simulated an invasion of Taiwan. On the landing beach, the traditional “human waves” of the People’s Liberation Army (PLA) were replaced by swarms of machines: suicide drones and well-known robotic quadrupeds like mechanized wolves. These units, developed by the state-owned China South Industries Group Corporation (CSGC), represent the first attempt to convert amphibious operations into a scenario dominated byor artificial intelligence. The broadcast images State television CCTV showed these metal “wolves” running across the sand ahead of human troops, detecting obstacles with LiDAR sensors, thermal cameras and autonomous navigation algorithms. Wolf specification. Of 70 kilos of weight and capable of carrying 20 more, these robots were divided into attack, transport and reconnaissance variants, managing to reduce the time between detection and destruction of the target to less than ten seconds. In fact, in one symbolic sequencea single human operator simultaneously directed nine robots and six drones from a 3D interface, while the devices cleared barbed wire and trenches for infantry. @elsa50356 “Breaking from China! The PLA’s latest amphibious landing drills—drones take the lead, and robotic ‘wolf packs’ rush the beach! The future of warfare is here!” 🚀🪖 #PLADrills #ChinaMilitary #Drones #RobotArmy #MilitaryTech ♬ 原创音乐 – Elsa Swarm intelligence. The training, called “Landing Operation in Taiwan” was part of an assault test coastal exercise carried out by the PLA 72nd Division, under the Eastern Theater Command, the unit operating in front of the Taiwan Strait. For the first time, quadruped robots performed as a spearheadfollowed by waves of FPV drones bombing simulated enemy fortifications. In total, the attack cycle was cfour times faster than that of a conventional square. This deployment is part of the EPL’s strategic shift from mass doctrine (the so-called human wave tactics) towards what Beijing calls “smart sea and land tactics,” a doctrine that prioritizes automation, cooperation between unmanned systems and data-driven decision making. The buts. However, the exercise itself revealed vulnerabilities: these wolf robots They lack armor, are easily detectable in open fields and one of them was destroyed by light fire. Chinese analysts they recognized limitations, but they stressed that the goal was not perfection, but rather to demonstrate that the army is willing to progressively replace human soldiers with swarms of coordinated machines. Ukraine in the shadows. The Chinese Army has incorporated direct lessons from the Ukrainian war into its maneuvers, where drones have redefined tactical and logistical effectiveness. According to Chinese military media like Daiwanthe PLA is applying the knowledge extracted from that conflict in its ground training, anticipating a future where hundreds of robots advance at 30 or 40 km/h in coordinated waves. The parallel is clear: if Ukraine demonstrated that a cheap drone can destroy a tankChina wants to prove that a network of smart machines can break coastal defenses in a matter of minutes. The current exercises, which until recently were limited to traditional landings, are already a general rehearsal of algorithmic warfare, where the human decision is reduced to an initial order and autonomous systems execute the rest. Strategic competition. Plus: The accelerated development of these systems occurs while the United States reinforces your deterrence strategy in the Indo-Pacific. According to the CIAan eventual Chinese invasion of Taiwan could occur before 2027, and the Pentagon has designed the so-called hellscape strategy: Saturate the strait with thousands of drones, submarines and unmanned vehicles to slow down Chinese forces and buy time for reinforcements to arrive. Beijing, aware of this, is creating units specialized in war against swarms, equipped with software capable of detecting, tracking and attacking targets without human intervention. Companies like Norincoanother state giant, have presented vehicles like the P60powered by the DeepSeek AI model, which can recognize targets, avoid obstacles and operate in logistics support or combat missions. A future of machines. He China’s advance towards an AI-powered war demonstrates both its technological ambition and its practical limitations. The images of robots breaching simulated beaches are as revealing as their failures in the face of enemy fire. However, beyond immediate effectiveness, Beijing’s message is unequivocal: the future of the war in the strait of Taiwan will be decided by the speed of the algorithms, not the number of soldiers. In that race, China seeks to transform mechanized warfare in smart warreplacing brute force with computational precision. The question is no longer whether robots will be present in the next invasion, but how many will be able to think, coordinate and eliminate before the first human makes landfall. Image | CCTV/China In Xataka | Less than 150 kilometers from Taiwan, the US does not stop accumulating missiles. It’s the closest thing to preparing for war. In Xataka | China has asked Russia for an airborne battalion and training. That can only mean one thing: they are preparing a landing

The most beautiful, exciting and hopeful thing about November has come out of England and it is a weather forecast

by

The most beautiful, interesting and hopeful thing about November has come out of a cold building in the British city of Reading and it is a weather forecast. In its latest seasonal reading, the European Weather Forecast Center has sounded the alarm: Your data points to a negative NAO. And that, as you may have guessed, is magnificent news for Spain. But let’s go in parts and explain what we’re talking about.. The ‘NAO’ is the ‘North Atlantic Oscillation‘: the ‘dance’ between the Azores anticyclone and the Icelandic low, the two major atmospheric phenomena that govern the meteorology of the North Atlantic. When the index we use to “measure who is winning” is negative, the Azores anticyclone is weaker than normal and, for this reason, it cannot block deep Atlantic storms. The direct consequence is that they circulate further south than normal: right at our latitude. That is, precisely, what the ECMWF has planned from its headquarters in Reading, England. Kristian Strommen et al. (2021) It is not an isolated prediction. On the contrary, experts they have been warning weeks of a 2025-26 winter conditioned by La Niña and a potentially weaker polar vortex; that is, with greater probability of cold bursts in southwestern Europe. This is the double confirmation we needed for a seasonal prediction that (seasonal as it is) remains too generic and uncertain. But how positive it is. Why is it important? As I say, for Spain this is, in aggregate termsgood news. By increasing the frequency with which cyclones deviate south (favoring the Atlantic storms that reach us), the direct positive impact is noticeable on the level of reservoirs and the price of energy. How much can we trust a prediction of this type? Moderately. There is no doubt that meteorologists have greatly improved their ability to capture trends, but let’s not fool ourselves: it is already difficult for us to predict 15 days ahead, the seasons and months are another matter. However, it is not a random prediction. We simply have to understand it as a risk or a probable scenario and not as a deterministic and closed forecast. Because, in the end, in the enormous set of possible scenarios, this begins to be the most probable. And, if confirmed, our reservoirs will thank us. Image | WeatherModels In Xataka | Ski resorts without snow at the end of the century: the most pessimistic models show what could happen in our high mountains

this is how you can get it

by

Although Google already launched, a few months ago, the Google Pixel 10he Google Pixel 9 It is still a good purchase option today. Now, at MediaMarkt you can get it at an unbeatable price: 499 euros in its 128 GB version. You just have to apply the code ‘TradeInPixel9100Nov‘ (without the quotes) when you have added the mobile phone to the shopping cart. Remember that this coupon will only be available until November 10 at 9 in the morning. Google Pixel 9 5G 12/128GB The price could vary. We earn commission from these links A high-end mobile now at a reduced price Although it was launched last year, the Google Pixel 9 It is a perfect mobile today within the high range. One of the things that makes it a winning horse (if you are thinking of buying an affordable high-end mobile) is because guarantee updates for six years of the operating system. The screen of this smartphone is 6.3-inch Super Actua OLED and with a resolution of 2,424 x 1,080 pixels. In addition, it stands out for being compatible with HDR10+ and present a refresh rate of 120 Hz. As for its battery, it supports fast charging at 45 W. The operating system under which it works is pure Android and its brain is the chip Google Tensor G4which is accompanied by 12 GB RAM and 128 GB of internal storage. Its photographic system is also outstanding and is made up of a dual 50+48 MP rear camera. Finally, it should be noted that it has IP68 certification. If you want a superior Pixel at a good price, you’re also in luck If the Google Pixel 9 is not enough for you, at MediaMarkt you can now get the brand’s most recent launch, the Google Pixel 10 with a 100 euro discount. On this occasion, if you apply the code ‘TradeInPixel10100Nov’ (also without quotes and which has the same validity as that of the Pixel 9), you can buy this mobile for 699 euros. The price could vary. We earn commission from these links You may also be interested in these accessories for this Google Pixel 9 Phoona Case for Google Pixel 9 with 2 Pieces Tempered Glass The price could vary. We earn commission from these links Xiaomi Redmi Buds 6 – Wireless headphones The price could vary. We earn commission from these links Some of the links in this article are affiliated and may provide a benefit to Xataka. In case of non-availability, offers may vary. Images | Pepu Rica (Xataka Android) and Google In Xataka | The best mobile phones, we have tested them and here are their analyzes In Xataka | The best quality-price mobiles. Their analyzes and videos are here

We already know which will be the most expensive data center in the world. If Bill Gates paid it, it would be almost zero

by

Already in 2024 we saw that infrastructure spending for AI was being insane. The trend has not relaxed, quite the opposite. Big tech continues to burn money as if there was no tomorrow (literally) and most of that spending is going to most valuable asset in the AI ​​race: data centers. How much do they really cost? Data centers in numbers Epoch AI has published Frontier Data Centersa complete database about data centers being built in the United States. Through satellite images, public documents and permits, they have obtained information about the estimated construction cost, as well as energy consumption and computing power. The award for the most expensive data center goes to Microsoft Fairwater, whose total cost It could reach $106 billion when completed in 2028. To put it in context, Bill Gates’ fortune is estimated to be 107 billion dollars. It would be fair to pay it. The forecast for Microsoft Fairwater even surpasses Meta Hyperion, the data center that It will be as big as the island of Manhattan which would cost 72,000 million. Next on the list is Colossus 2, by xAIwhose estimated cost is 44 billion dollars. It is closely followed by Meta Prometheus with 43 billion and the Amazon and Anthropic data center in New Carlisle with 39 billion. Epoch AI has collected more data, such as how much computing power each facility will have. This data is measured using the NVIDIA H100 GPUs for reference. They have also calculated the energy demand and who will be the main user of each of them. Below we leave you a table with the key information: Estimate DATE ESTIMATED cost ($) computing (EN gpUS H100) energy demand intended primary user microsoft fairwater September 2027 106 billion 5.2 million 3328 MW OpenAI meta hyperion January 2028 72 billion 4.2 million 2262 MW Goal xai Colossus 2 February 2026 44 billion 1.4 million 1379 MW xAI meta prometheus October 2026 43 billion 1.2 million 1360MW Goal amazon new carlisle June 2026 39 billion 770,000 1229 MW Anthropic oracle stargate July 2026 32 billion 1 million 1180MW OpenAI microsoft fayetteville March 2026 29 billion 920,000 1065MW OpenAI/Microsoft amazon ridgeland September 2027 32 billion 630,000 1008MW Anthropic Dizzying climb Looking at the case of Microsoft Fairwater, and always according to Epoch AI’s forecast, in March 2026 the investment will be $18 billion. A year later, in February 2027, it rises to 35,000 million, just four months later it shoots up to 71,000 million, to reach 106 billion in 2028. The price increase is dizzying and responds to several factors. The first is that the computational cost of training models has been increasing. For example, GPT 4 cost OpenAI over 100 million and rumors before the release of GPT-5 pointed to training rounds of 500 million each. Epoch AI also did an analysis on this and they estimated that the cost of training has multiplied by 2.6 year after year. On the other hand, there is the demand for GPUs, necessary for training the models and the most expensive component of all. An NVIDIA H100 GPU costs 25,000 dollars and its successor, the NVIDIA B200 also known as Blackwell, could be between 30,000 and 40,000 dollars. And this is just the GPUs, many are needed more components to get a data center up and running, such as power generators, high-speed networks or refrigeration, among others. The initial bottleneck was the shortage of GPUs, but it has been overcome by a more fundamental constraint: there is not enough power for so many chips. data centers They consume a lot of energy, Seriously, a lot. To put it in context, in 2024, data centers were already the 4% of United States electricity consumption and it is expected that Demand will double in the next five years. Nobody wants to live near a data center for one reason: mass consumption is raising energy prices up to 267% in nearby areas. Power supply has become a new choke point for the industry. Microsoft is already considering producing its own energy by creating nuclear power plants and others like Google and Amazon are considering taking data centers into space. Image | Microsoft In Xataka | AI data centers are an energy hole. Jeff Bezos’ solution: build them in space

We believed that no open model could outperform GPT-5. A Chinese startup proves us wrong

by

A Chinese startup called Moonshot just launched Kimi K2 Thinkinga gigantic open model with a trillion parameters that has done something that seemed almost impossible: surpass the best proprietary models from companies like OpenAI, Google or Anthropic. If we thought that “Open Source” models could never compete with GPT-5, Gemini 2.5 Pro or Claude, we were wrong. what has happened. This “AI laboratory” had already announced Kimi K2 in July with that gigantic size of one trillion parameters, but now they have released the “Thinking” version with that same size (32 billion active parameters, Mixture of Experts architecture). According to those responsible, the model is capable of maintaining stable use of agentic tools over between 200 and 300 sequential calls. Or what is the same: it can chain long sequences of actions autonomously and apparently without error. The best of all is not that: it is that it surpasses GPT-5 or Claude Sonnet 4.5 in various tests and costs much less than those models. The benchmarks. Those responsible for Moonshot explained how Kimi K2 Thinking achieves the highest scores in Humanity’s Last Exam (general knowledge, 44.9%) and BrowserComp (agent browsers, 60.2%). He is almost at Claude’s level in the SWE software development test, and is also almost the best in another of those benchmarks, LiveCodeBench v6. It is true that in some tests still slightly behind of its “western” rivals, but the achievement is spectacular. More benchmarks. Those responsible for Artificial Analysis have shown their first conclusions after evaluating it with various tests. Thus, they highlight its behavior in agentic tasks that simulate that the model is acting as a customer service agent. In this test it obtained 93% of the maximum, surpassing all its competitors by far (GPT-5 Codex High obtained 87%, for example). They will do more tests, but for now the prospects are fantastic. And on top of that, cheap. On CNBC indicate that training the model cost $4.6 million, a ridiculous figure considering that training proprietary models like GPT-5 It cost about 500 million dollars according to estimates. Using the Kimi K2 Thinking API is also very affordable: $0.6 per million tokens in and $2.5 per million tokens out. GPT-5 Chat costs $1.25/10 respectively, while Claude Sonnet 4.5 costs $3/15 respectively. The details. The model makes use of an INT4 quantization to improve its efficiency without compromising the precision and quality of its responses. Its context window—the “size” of the data we can enter when making prompts—is 256k, a relatively modest figure for large models but still notable. And as a good open model, we can download it to use locally… if we have a real monster at our disposal. The model weighs 594 GB, and for example joining two Mac Studio M3 Ultra It is possible to make it work locally relatively smoothly at about 15 t/s. Alibaba is behindyes. Although the model is developed by an independent startup called Moonshot, this firm has been financially supported by Alibaba, which is becoming an absolute powerhouse in this field. Already not only conforms with developing its own models, which are outstanding (Qwen is the clear example), but is also financing the development of other models such as Kimi K2/Thinking. China and its love for open AI models. During the last few months we have seen how China dominated in the field of open AI models —not “Open Source”—. The Asian giant has adopted an overwhelming philosophy with increasingly better models but which until now seemed to be several steps behind the large proprietary models of OpenAI, Anthropic or Google. This is no longer the case. The race is lively. This achievement represents a new vote of confidence for the open models coming from Chinese companies. It is true that they are huge and that makes it very difficult to use them in practice by end users, but they present an interesting alternative for companies. Image | idnaklss with Midjourney In Xataka | There are many “internal” races within the greater AI race. And Alibaba is winning Open Source

The danger is not when, it is the Arctic

by

The recent crossing of threats between Putin and Trump has revived a tension that seemed buried from the hardest years of the cold war. The Russian president ordered his senior commanders to prepare plans to resume nuclear tests after Trump’s statements on social networks, in which he announced that the United States would resume its tests “immediately.” If so, nuclear weapons experts are clear about how long it would take for Russia to carry out a “real” test. The nuclear ghost. Although the intention of the North American president seemed more political than technical (referring to tests of launch systems and not to real detonations), in Moscow the interpretation it was another: The Ministry of Defense assumed that Washington seeks to reopen the nuclear race and recommended Putin to be ready for “full tests” in the Arctic field of Novaya Zemlya. It we count: that gesture, accompanied by recent demonstrations of the Russian arsenal (since the Burevestnik missile nuclear propulsion to intercontinental torpedo Poseidon), symbolizes the disappearance of the last brakes in the atomic dialectic between the two powers. The end of the agreements. The current climate is the result of years of system erosion of gun control. Russia suspended its participation in the New START treaty in 2023, while the historic INF agreement, which banned intermediate-range missiles, had already been abandoned by both countries in 2019. Despite maintaining some technical respect for launch limits, the absence of verification and transparency has turned the arsenals of Washington and Moscow (5,177 and 5,459 warheads, respectively) in a field of permanent suspicion. The Putin’s orderMore than a technical step, it represents a political message: that Russia will not allow the United States to monopolize the symbolic gesture of resuming tests that, if carried out, would break the taboo in force since 1990 and the spirit of the Comprehensive Nuclear Test Ban Treaty. The Kremlin itself seems to have assumed that the return to “eye for an eye” logic It is part of the new post-Ukraine order, where shows of force count as much as victories on the battlefield. Satellite image showing tunnel construction at the Novaya Zemlya nuclear weapons test site in Russia. Russian viability: the Arctic. To the big question, nuclear security experts agree that Russia could carry out a real test within a margin of weeks or monthsdepending on the degree of instrumentation and preparation desired. Hans Kristensenof the Federation of American Scientistsestimates that an improvised detonation (without complex data collection) could be carried out quickly, although without significant scientific or military value. On the contrary, a complete, “real” test, with sealed tunnels, sensors and wiring, would require at least half a year of jobs in Novaya Zemlyawhere underground works have continued discreetly for years. Jon Wolfsthalfrom the American Federation of Scientists, gives the key: seasonal limitations, since the extreme arctic weather would allow trials of this caliber only in summer or early fall. However, both he and other analysts agree that the purpose would be mainly political (show parity with Washington) more than scientific. The great uncertainty. Most experts consulted on TWZ He stressed that neither Russia nor the United States have a technical or military need to resume nuclear testing. Both have extensive arsenals and advanced simulation programs that guarantee the reliability of their weapons without resorting to detonations. Daryl Kimballof the Arms Control Associationremember that Washington has made 1,030 historical tests and Moscow 715and that any new trial would be “purely for show,” an irresponsible act with no tangible benefit. Stephen Schwartz added that the United States maintains a structural advantage thanks to its arsenal maintenance program, valued in 345,000 million of dollars, and that Russia, although it could act with fewer environmental or political obstacles, would gain nothing beyond fueling the spiral of distrust. Still, Russian infrastructure on Novaya Zemlya, modernized in recent years, demonstrates a capacity to respond quickly if tension turns into action. A new deterrent. Beyond of personal confrontation between Putin and Trump, the real risk lies in precedent. A single test (even if it is underground and of low power) would be enough to break three decades of tacit consensus and open the door to new tests by, for example, China, North Korea or other actors seeking to legitimize themselves as nuclear powers. The gesture would have a huge symbolic power: demonstrate that powers can rewrite the rules of nuclear balance when they consider it necessary. In that sense, the experts’ warnings are clear: what is a rhetorical escalation today could become a tangible competition tomorrow, with unforeseeable global consequences. As Wolfsthal pointed out“this is what an arms race looks like: action, reaction, and a slope that costs much more to go down than to go up.” Echoes of the Cold War. The exchange between Moscow and Washington Not only does it resurrect the shadow of the nuclear confrontation, but it redefines its scenario: it is no longer fought in secret offices or under the logic of the balance of terror, but in televised broadcasts and social media posts. The threat of detonating atomic bombs again in the 21st century reveals a dangerous mix of geopolitical nostalgia and spectacle politics. Deep down, both know that no country can “win” a nuclear race. And yet, the temptation to show power, to regain influence and to project invulnerability to their respective audiences could be enough to reignite the powder keg. most feared on the planet. The silence of thirty years underground could be broken by a simple click on a social network. Image | Ministry of Defense of Russia In Xataka | The US and Russia have agreed on nuclear weapons: the time has come to take them out and see if they work In Xataka | In 1950 two scientists wondered if a 10 gigaton nuclear bomb was possible. Your results are hidden under lock and key

How to do it and what time limit do you have?

by

Let’s tell you how you can check the operation of your new V16 beacon without alerting the DGT. As you know, these beacons have an internal SIM card that connects with the DGT, so that when you activate them because you have an accident or breakdown, they can geolocate you. But what happens when you just want to check that everything works well on a beacon you just bought? Well, we are going to tell you the amount of time you have before this connection is established. Because in the end, when you have spent money on one, what you want is to see that it works to see if you should return it or keep it. 100 seconds margin As specified by the DGT in one of his postsV16 beacons and their SIM cards They have a margin of 100 seconds during which no information is sent. This is the time the device has to establish the connection with the mobile network, check its GPS and prepare to send data. This serves to provide a secure testing window, as well as to prevent false alerts or messages from being sent in error. This means that you can turn on the beacon and test that it works, since will not alert the DGT until 100 seconds have passed. This way, you can see that your light is working and everything is fine. Remember that you must activate these beacons on the roof of the car if you have an incident. And when they send your position to the DGT after those 100 seconds, it will be able to use the road information panels to warn other drivers, send notices to compatible navigation applications such as Google Maps, and make it easier to locate you. They are mandatory from January 1, 2026. In Xataka Basics | What to look for in a V16 beacon: requirements and how to check if the one you have or want to buy works for you

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.