Spain and Portugal have “free” energy right now. If we do not share it with Europe it is due to only one reason: France

While the Iberian Peninsula registers a surplus of unprecedented renewable energy at bargain prices, the rest of the continent continues to be suffocated by triple-digit bills. In the middle of these two realities a wall rises, not of stone, but of political and nuclear interests: France. The northern neighbor acts as a plug that prevents cheap energy from the south from flowing north, protecting its atomic industry at the expense of European consumers’ pockets. Two Europes disconnected. The data from February 11 are a blow to the table of European integration. According to the records of OMIE and ESIOSthe average daily market price in Spain has plummeted to €4.23/MWh, with hours in which producers have had to pay for injecting energy (negative prices of -€0.42/MWh). The situation in Portugal is even more extreme: the megawatt hour is paid at €0.34, that is, practically free. However, it is enough to cross the Pyrenees for reality to change drastically. The price map ESIOS turns central and northern Europe red: Germany pays electricity at €100.62/MWh, Belgium at €72.04/MWh and the Netherlands at €88.70/MWh. France, strategically located in the middle, enjoys a comfortable price of €13.61/MWh, benefiting from buying cheaply from the south without missing out on the flow to its northern neighbors. This disparity perfectly visualizes the concept of “energy island”: a peninsula overflowing with resources that does not have enough bridges to share them. The great uncoupling of February. What we are experiencing these first two weeks of February is what experts call a “total decoupling.” According to the analysis of Aleasoft Energy Forecastingthe arrival of several Atlantic storms has triggered wind and hydroelectric generation on the peninsula. By adding the solar contribution, the supply has far exceeded the internal demand. The Iberian market (MIBEL) has seen how their prices They fell by 43% in Spain and a staggering 74% in Portugal in just one week, reaching daily averages of €0.54/MWh, values ​​that had not been seen since April 2024. Meanwhile, the Energy Charts graphs show that Germany has continued with prices oscillating above €100/MWh for much of January and early February, still depending on non-renewable sources. The drama of throwing away energy. Having cheap electricity seems like excellent news for the domestic consumer, but it hides a serious systemic inefficiency. As there are not enough cables to export this surplus to a Europe thirsty for cheap energy, Spain is forced to carry out curtailment (technical discharges). As we have already explained in Xatakawe are literally throwing away around 7% of clean energy because it “does not fit” into the grid and has no outlet. This scenario causes zero prices that, paradoxically, can ruin renewable investors, who need profitability to continue deploying parks. Furthermore, the situation has uncovered the seams of the Spanish internal network. The network is administratively “collapsed”: the CNMC has had to delay until May 2026 the publication of the capacity maps because, under the new security criteria, 90% of the network nodes appear saturated. Only 12% of connection requests are being approved, which means that we have the energy, but the cables are missing to bring it to new industries and homes. The French nuclear “bunker”. If there is excess energy in the south and lack in the north, why not build an electric highway? The answer has its own name: nuclear protectionism. President Emmanuel Macron has declared that interconnections They are a “false debate”arguing that Spain’s problem is a “100% renewable model that its own network does not support.” However, the data refute the Elysée story. As expert Joaquín Coronado explainsSpain is not 100% renewable (it closed 2025 at 55.5%) and, in fact, it was Spain that came to the rescue of France in 2022 and 2025, exporting electricity through its combined cycles when the French nuclear park failed due to corrosion and heat problems. The reality, according to the CEO of RedeiaRoberto García Merino, is that the blockade “is not technical, it is pure geostrategy.” France needs to make profitable a pharaonic investment of 300,000 million euros in its nuclear park and fears that the massive entry of Spanish solar energy, much cheaper, will sink the prices and competitiveness of its reactors. Therefore, Paris has explicitly excluded of its 2025-2035 network plan the key interconnection projects for Aragon and Navarra, keeping the Iberian Peninsula as an island with only 2.8% interconnection, very far from the European objective of 15%. Any solution on the table? Brussels’ patience is running out. The European Commission has already issued an ultimatum to Francegiving him a period of nine months to unblock the situation and present a political declaration of commitment. Meanwhile, the only project that advancesalthough slow, is the submarine cable through the Bay of Biscay. Redeia confirmed that the laying campaigns will begin this summer of 2026, with an eye on its entry into operation by 2028. An unsustainable contradiction. Within the European Union, it is happening that while one member country desperately seeks energy autonomy and competitive prices for its industry, it allows another of its key partners to keep the door to the south closed. Spain could be Europe’s green battery, but without export capacity, that wealth is diluted in negative prices and technical waste. Everything happens while France acts as a strict customs officer that protects its atoms, preventing the European Union from truly being an energy union. Image | freepik Xataka | The great electrical jam in Spain: we have plenty of electricity, but there are no cables to build houses and invest more

OpenClaw is the most viral, fascinating and dangerous AI of the moment. For this last reason, it has joined forces with VirusTotal from Malaga

In 2025 we had a ‘DeepSeek moment’ and in 2026 we are having an ‘OpenClaw moment’. This AI agent is super powerful, but also super insecure. There is, however, good news, because the Malaga company VirusTotal has partnered with the OpenClaw project to try to mitigate one of the most important cybersecurity risks of this AI agent: its skills. what has happened. OpenClaw (formerly Moltbot, and before Clawdbot) has announced that it has begun a collaboration with the Malaga cybersecurity company VirusTotal, owned by Google. The agreement will see VirusTotal be in charge of “scanning” and analyzing the so-called “skills”, which work like OpenClaw plugins and add all kinds of functions. They do it, of course, but many take the opportunity to introduce malicious instructions that allow them to steal data and remotely operate other people’s AI agents. More security for disturbing AI. Peter Steinberger, creator of the project, has joined Jamieson O’Reilly, cybersecurity expert and founder of the company Dvulnand Bernardo Quintero, founder of VirusTotal, to offer that “additional layer of security for the OpenClaw community.” In it official announcement explain that “all the skills published in ClawdHub (the project’s official skills “store”) are now scanned through Virus Total’s Threat Intelligence system, including its new capability Code Insight (code inspection)”. Bernardo Quintero indicated on Twitter how the effort has already allowed 1,700 skillls to be identified as malicious. If the skill is malicious, it is blocked. This analysis carried out with the VirusTotal tools allows us to identify skills as malicious and block them immediately so that they cannot be downloaded. Not only that: those skills that have been classified as benign are analyzed again every day to detect scenarios in which for some reason they could end up becoming malicious. Still, be careful. Those responsible for OpenClaw warn: the VirusTotal scan helps a lot, but it is not a total guarantee that any skill can perform malicious actions on the machine on which we have our AI agent installed. The attacks of prompt injection Sophisticated skills can manage to cross that barrier, but of course this collaboration means that OpenClaw users can be much calmer regarding the skills available in the ClawdHub repository. OpenClaw wants to be much more secure. This first effort joins OpenClaw’s ambition to have a complete cybersecurity model which includes things like a public roadmap for your new developments in this area, a formal communication process, and details about full audits of your code. Plugging a problem that could kill OpenClaw. The OpenClaw project soon went viral due to its eye-catching options, but shortly after doing so a security audit initial 2,851 skills detected 341 malicious skills. Companies like BitDefender also joined these efforts to avoid problems with tools like AI Skills Checker to check whether a skill was dangerous or not. These malicious skills were, for example, capable of executing shell commands on the victim machine, which gave the attacker complete control of those resources. Attacking the machine is confusing it with natural language. Normally cybersecurity attacks are complex, but the problem with AI agents is that they work with natural language. This implies that to infiltrate these systems you do not have to use code, but simply “convince” and “trick” the AI ​​with natural language. That is where prompt injection attacks come in, which consist of giving instructions to those AI agents that can confuse them to obtain something that theoretically they should not allow them to obtain. Personal data, API keys of the models we use at OpenClaw, email accounts and passwords for all types of services… the possibilities are endless, and OpenClaw, which has access to all of this to operate autonomously, can end up being “tricked” into transferring said data. Beware of OpenClaw. These problems now seem a little less feasible thanks to the collaboration with VirusTotal, but those who are trying OpenClaw on their machines or any other platform should be very alert from the beginning. There are guides that help you install it with some barriers important security issues, and the project itself has a command (‘openclaw security audit –deep –fix’ to audit the most important problems and address them. In Xataka | OpenAI has a problem: Anthropic is succeeding right where the most money is at stake

He does it for a good reason.

In addition to being supported by one of the largest fortunes in the world, Blue Origin obtained a good part of its financing from suborbital trips, for which its millionaire passengers paid a fortune. However, Jeff Bezos is going to leave his exclusive space tourists on the ground for at least two years. In a statementBlue Origin has announced that it is now prioritizing its lunar plans over short New Shepard flights, putting on hold its luxury weightlessness experiences that have taken celebrities and great fortunes to the edge of outer space. The rich stay on land. Blue Origin has decided to suspend the flights of its reusable New Shepard rocket “for no less than two years” to redirect resources towards the Artemis project that must bring it back to life. humanity to the moon. This pause directly affects the space tourism business that depended on this reusable rocket launched from the Texas desert, where both the ship and the capsule return to be reused in future missions. Space tourism has been one of the great sources of income for Blue Origin, which has built a solid client portfolio and even has a waiting list for several years. The official statement does not clarify what will happen to people who already had flights scheduled for the coming months, leaving these millionaires in temporary limbo. It was a means, not an objective. The suspension of the space tourism program shows that the program was always a means of making profitable and financing Blue Origin’s developments and innovations, but the priority had always been reach the moon either establish orbital colonies. These suborbital flights are clearly reserved for economic elites, celebrities and select guests. Among his amateur “astronauts” is Jeff Bezos himself, his wife Lauren Sanchez, the singer Katy Perry, or to Spaniards like him media Jesús Calleja o Alberto Gutiérrez, founder of the platform Civitatis that just a few days ago reached the Kármán line. This stoppage shows that, although exclusive, suborbital tourism has not achieved sufficient launch frequency to maintain itself as a priority compared to more contracts. lucrative and strategic with NASA. Decisive bet on the Moon. Blue Origin maintains a multimillion-dollar contract with NASA to develop Blue Moonits lunar landing module for the Artemis programwhich seeks human return to the Moon and a sustained presence there. The pause in New Shepard launches is intended to “redirect resources to further accelerate the development of the company’s lunar human capabilities” by freeing up engineers, materials and budget to accelerate Artemis 5, planned for 2029 as descent module for astronauts. Blue Origin has underlined its “commitment to the national goal of returning to the Moon and establishing a permanent and sustained lunar presence.” Space tourism is left without competition. Blue Origin’s decision creates a new gap in the offer of suborbital flights intended for “space tourism.” In 2024, its main rival, Virgin Galacticsuspended its operations due to the high cost of the missions and the development of rockets to carry them out. However, Richard Branson’s company has announced the possibility of resuming its flights in 2026, although with higher rates. For the millionaires who longed for their moment of glory floating in the New Shepard capsule, this means at least two years of waiting without a clear resumption date, confirming that Jeff Bezos has chosen the Moon like the real prize. In Xataka | Jeff Bezos asked his parents for their life savings to found Amazon. They only asked him one question: “What is the Internet? Image | Blue Origin

Tesla is pivoting to turn its cars into a side business. The reason: their income falls by 61%

The Tesla Model S and Model X are incredible cars. Get them while they’re still available! With these phrases, Elon Musk, CEO of Tesla, has accompanied the company’s announcement in X in which they point out that during the next quarter they will reduce their production of the Tesla Model S and Model To its credit, the company will produce Optimus robots. by surprise. It was known that Elon Musk has been pushing for some time for Tesla to increase its investments in artificial intelligence and robots, either in humanoid form like Optimus or through its robotaxis for autonomous driving. But what we did not expect is that this bet would displace two of its most iconic models. And the company will stop producing its Tesla Model S, its first sedan, and the Model X, its first SUV, in Freemont (California) to make way for the production of Optimus robots. The company closes a chapter by recognizing that “Tesla would not be what it is today” without these cars. In Xataka Tesla wanted to make 20 million cars in 2030. The reality in 2025 is that Tesla has crashed and BYD is already leading A paradigm shift. The decision to invest in this factory to increase robot production is more than just a redistribution of its efforts, it is confirmation of a change in strategy in the company. Musk seeks invest $2 billion in xAIthe company dedicated exclusively to artificial intelligence. Intertwining your companies is one of the obsessions from the CEO of Tesla so that some feed each other. xAI is key to power and improve Grok which, in turn, is already included in Tesla vehicles as an artificial intelligence assistant. At the same time, xAI is also decisive for the functioning of its robotaxisthe cabin without wheels or steering wheel that Tesla wants to put on the street to offer a completely autonomous taxi service. In Xataka Tesla can’t wait for us to take our hands off the wheel. We have tried it and we have opinions More than complicated numbers. Optimus has left many doubts and Musk himself has confirmed that he expects a slow deployment. However, dedicating a plant that only manufactured a handful of cars is not only confirmation that the company does not care in the least about killing a product if it understands that it is not profitable or that its future is much less interesting than a new bet. Changing the use of the factory is also a necessity. And the numbers presented by Tesla are something much more than complicated: Net profit has gone from 7.1 billion to 3.8 billion dollars, 45% less. In the last quarter, turnover has fallen from $2.1 billion last year to $840 million. It is a drop of 61%. The company has delivered 1.64 million cars in 2025 in what is its second year reducing its sales. In the United States the drop in sales is 7%, according to Cox Automotive, reported in The New York Times.  In the same period, it is estimated that BYD has sold 2.25 million cars Purely electric. In Xataka The Tesla Cybertruck is such a sales failure that Elon Musk has only found one solution: buy them from himself Loss of identity. The Tesla Model S and Model X have become residual cars for the company since the Model 3 and Model Y occupied the bulk of sales. Both are very expensive cars that cost around or exceed 100,000 euros. Both the saloon and the SUV served the brand to boost your image and personality as unique cars. Over the years, that has been lost. And the huge screens that previously surprised now do not stand out in a market that has turned to trying turn the cabin into a multimedia centerespecially in China. Your own assembly line has been forced to keep its design unchangedwhich has made them lose freshness. The popularization of its Tesla Model 3 and Model Y has popularized access to the company, making them lose part of that desirable car aura. {“videoId”:”x9tnvi4″,”autoplay”:false,”title”:”Why YOUR NEXT CAR WILL SURELY BE CHINESE”, “tag”:”Webedia-prod”, “duration”:”614″} A cut production. The decline in sales has led to declining production of both models. To give us an idea, nothing is better than the data provided by the company itself: 2022: 71,777 units produced and 66,705 deliveries 2023: 70,826 units produced and 68,874 deliveries 2024: 94,105 units produced and 85,133 deliveries* 2025: 53,900 units produced and 50,850 deliveries* Starting in 2024, Tesla accounts for the production and deliveries of the Tesla Model S, Model X and Cybertruck in the same item. That’s whyCybertruck sales are estimates outside of Tesla The Tesla Model 3 and Model Y Standard confirms a story. The story of what I want and I can’t of Tesla’s 25,000 euro car In Xataka The Tesla Model 3 and Model Y Standard confirms a story. The story of what I want and I can’t of Tesla’s 25,000 euro carThe limits . Tesla is in a stagnant situation with its electric cars. The company stepped on the accelerator in 2024 to remain the best-selling electric car brand in the world and improve the previous year’s data. But it did not succeed, going from 1.85 million cars produced and 1.81 million cars delivered in 2023 to 1.77 million units produced and 1.79 million cars deliveredin 2024 . Year in which, in addition, They increased their range with the Cybertruck which started at a very good pace. The company, therefore, needs to kill some very expensive cars that are barely generating a positive impact on its accounts no matter how high the profit margin obtained with each unit. To begin with, because the company needs a boost from its investors, who seem to support these decisions. And, second, because we have to see if the company has not already peaked in its vehicle sales. At leastwith its particular way of producing cars with huge presses that are only profitable by manufacturing millions and … Read more

All viewers believe that the trailers spoil the movies too much. But there is a reason: fear of lawsuits

The paradox of trailers: they serve to encourage the public to see upcoming releases, but more and more people decide to literally cover their eyes or start talking to their neighbor, because the feeling that the trailers reveal too much is widespread: plot twists, climatic scenes that should be a surprise. There is a more or less intuitive reason: the market is increasingly competitive and it is important to show the public what each film offers that the others do not. But there are more prosaic reasons why trailers reveal more and more about movies. The trailer as a marketing tool.For decades, trailers were considered pieces with their own narrative: small works that condensed the spirit of a film, not mere advertisements. That premise was shaken in December 2022, when a court ruling questioned the legal limits of film marketing. The case pitted Universal Pictures against two viewers who claimed to have been misled by the ‘Yesterday’ trailer. The ‘Yesterday’ case.Two viewers had rented the film after seeing the trailer, in which Ana de Armas appeared in an apparently relevant role; but in the film he had disappeared: his character had been completely eliminated after test screenings. The plaintiffs alleged that they would never have paid for the film if they had known that de Armas was not in it. In Xataka The AI ​​trailers for ‘Avengers: Doomsday’ are indistinguishable from the real thing. In the end, Scorsese was right Universal Pictures requested the case be dismissed, arguing that the trailer conveyed the film’s theme in three minutes, but the judge rejected this line of defense. Although trailers involve creativity and editorial decisions, these elements do not nullify their fundamentally commercial nature: they must be treated as advertisements, and the sample they show of the film must correspond to the final product. The judge specified that his resolution was limited to the presence or absence of interpreters, excluding subjective assessments of tone, quality or generic expectations, but set a precedent. It’s not the first. The friction between what the trailers promise and what the movies deliver has generated some attempted litigation. None went so far as to establish firm jurisprudence, but all illustrate a recurring tension between public expectations and studio marketing practices. {“videoId”:”x88pexn”,”autoplay”:false,”title”:”Yesterday Trailer”, “tag”:””, “duration”:”180″} Drive (2011).a spectator sued the distributor claiming that the trailer presented the film as an action film in the style of ‘Fast & Furious’, when in reality it was an atmospheric drama with few chases and a practically mute protagonist. The case dragged on for years without success for the plaintiff: the film did contain driving scenes, but the discrepancy lay in the tone and pacing, not in objective matters. Suicide Squad (2016).The trailers had highlighted Jared Leto as Joker, but his presence in the final cut turned out to be less than fifteen minutes. a scottish fan announced his intention to sue Warner Bros. for false advertising. Leto himself fueled the controversy by confirming that the deleted material It was so extensive. what would make for an independent film. The lawsuit was unsuccessful, but it highlighted the problem of trailers edited before final editing. Dune (2021).Zendaya featured prominently in promotional material: posters, trailers, and a press tour placed her on the same level as Timothée Chalamet. However, his screen time barely exceeded seven minutes of a total footage of 155and most of his appearances were dream sequences. There were no legal repercussions: Zendaya had previously warned that her presence was reduced and that she had only filmed for four days. {“videoId”:”x88q6ut”,”autoplay”:true,”title”:”Dune Trailer”, “tag”:””, “duration”:”208″} The Castaway case.Robert Zemeckis, with his usual ability to anticipate the rest of the industry, had already traveled this path years before. The trailer for ‘Castaway’ (2000) was criticized at the time because it revealed the eventual rescue of the protagonist. Zemeckis defended himself with an argument that is still valid in the industry, beyond the current legal precautions: market studies indicate that the public wants to know exactly what they are going to see before paying for a ticket. The problem of outsourcing.Trailer production rarely falls to the films’ creative teams. Studios hire specialized agencies (Buddha Jones, Trailer Park or Mark Woolen & Associates, only in Los Angeles) that work with raw material, often months before there is a final assembly. These agencies operate fromdailieseitherrushesthe raw footage that comes directly from filming. The process of creating a trailer can take up to a year, a calendar that forces you to work without knowing the final cut. The case of ‘Yesterday’ is a direct consequence of this dynamic. In Xataka Good series are a journey that no spoiler can ruin The pressure for difference.When a franchise accumulates multiple installments, marketing teams face an additional dilemma: how to convince the public that this film offers something different from the previous ones? The answer often involves revealing the differentiating element. The trailer for ‘Terminator: Genesis’ (2015) told that John Connor, traditionally the leader of the human resistance, had been turned into a machine, a twist that constituted the dramatic core of the film. Director Alan Taylor acknowledged that the decision responded to a complex calculation: how to signal to the public that this installment was not a mere repetition of the previous ones? A dilemma that promises to continue giving us headaches for a long time. In Xataka | Disney is looking for a successor to Bob Iger as CEO and has only one condition: that he does not look like Bob Iger’s previous successor as CEO (function() { window._JS_MODULES = window._JS_MODULES || {}; var headElement = document.getElementsByTagName(‘head’)(0); if (_JS_MODULES.instagram) { var instagramScript = document.createElement(‘script’); instagramScript.src=”https://platform.instagram.com/en_US/embeds.js”; instagramScript.async = true; instagramScript.defer = true; headElement.appendChild(instagramScript); – The news All viewers believe that the trailers spoil the movies too much. But there is a reason: fear of lawsuits was originally published in Xataka by John Tones .

Windows 95 had a little secret that made rebooting faster. The reason was in its more chaotic architecture

If before Windows 95 If you used other operating systems, it’s hard not to remember the feeling of being faced with something completely new. That proposal introduced elements that we take for granted today, such as the Start menu, the taskbar or Plug and Play, and it did so at a time when starting a PC was almost a small ritual. But beneath that familiar interface a complex architecture was hidden, the result of the forced coexistence between DOS inheritances, 16-bit Windows and the first 32-bit layers. That design, as inelegant as it was effective, gave rise to unexpected behaviors that still surprise today. Few users knew that Windows 95 hid an alternative route to the classic reboot. It was enough to hold down the Shift key during the process started from the graphical interface for the system to display the warning “Windows is restarting”, instead of following the path of a cold restart, as described by Raymond Chen. The difference was not spectacular, but it was noticeable at a time when every minute of starting counted. That small gesture activated an internal mechanism designed to avoid, whenever possible, starting from scratch. The shortcut that did not restart completely Behind this behavior there was a precise technical decision. Chen details that Windows 95 used a flag called EW_RESTARTWINDOWS when invoking the old ExitWindows function, still 16-bit. With that instruction, the system did not order a cold restart of the computer, but rather something more limited: close Windows and restart it. The objective was to save steps, as long as the internal situation allowed it, although this optimization depended on everything fitting correctly. Once that alternative route was taken, the process followed a very specific sequence. The 16-bit Windows kernel was shut down first. The 32-bit virtual memory manager was then turned off and the processor returned to real mode, the most basic state of the system. At that point, control returned to win.com with a special signal asking for something very specific: restart Windows in protected mode without going through a full computer boot. With control back on win.com, the most fragile part of the process began. The program had to simulate a clean boot of Windows, as if it had just been run from scratch, which involved, in Chen’s words, resetting the command line options and returning some global variables to their original values. Although the work was largely clerical, it was especially complex because win.com It was written in assembly. There were no abstractions or modern conveniences. The decisive point was in memory. When win.com was executed, like any .com file, it received all available conventional memory. However, it freed up almost all memory beyond its own code so that Windows could load a large contiguous block when entering protected mode. If during the session a program reserved memory within the space that win.com had left free, the memory was fragmented. In that scenario, win.com could no longer recreate the original map it expected, and, Chen explains, it was forced to abandon the fast reset and fall into a hard reset. When everything fell into place, the process continued without turning back. win.com jumped directly to the code responsible for booting Windows in protected mode, recreating the virtual machine manager and llifting the 32-bit layers again. From there, the graphical interface loaded as usual and the user returned to the desktop. The difference was subtle but real: Windows hadn’t had to reboot the entire system to get to that point. This type of shortcut was only viable in a system built on cross-compatibilities. Windows 95 had to coexist with DOS software, 16-bit Windows programs and Win32 applications, and that mix forced us to accept inelegant but very practical solutions. The developers took advantage of this complexity to introduce hidden optimizations that could speed up restarts, although they could sometimes end in crashes. The obsession with saving memory led to very imaginative solutions. Chen explains that in assembler it was common to recycle code that was no longer going to be used as if it were free memory. On win.com, the first bytes of the entry point were reused as a global variableunder the premise that this code was only executed once. Since the quick restart did not return to that initial point, the system could allow that shortcut without affecting the process. That shortcut also showed its seams. Chen recalls that some users detected errors after performing several consecutive quick reboots, something that he was unable to consistently reproduce. Their hypothesis is that some driver wasn’t rebooting properly, leaving the system in a weird state, and that weirdness ended up taking its toll later. It’s no surprise that this type of behavior wasn’t presented as a documented feature, but it sums up the spirit of Windows 95 well: inventive, ambitious, and full of compromises. Images | Microsoft In Xataka | Schrödinger’s Office: at this point it is impossible to know if Microsoft keeps it alive or if everything is AI and Copilot

There is a word that has multiplied exaggeratedly in scientific articles for a reason: ChatGPT likes it

That there are academic articles written by AI is something that has been proven beforethe question is how serious it is. To know the magnitude of this practice, a group of researchers has reviewed millions of paper abstracts published in PubMed and have found something interesting: there is a word that the AI ​​loves and the reason why it likes it so much is quite murky. Delve. Its translation is ‘go deeper’ and its use multiplied by 28 between 2022 and 2024, which coincidentally coincides with the boom of ChatGPT and language models. Other words such as ‘underscore’ or ‘showcasing’ are also cited, with a frequency increase of x13.8 and x10.7 respectively. None of them are a noun or a word related to the content, but rather have more to do with the style of writing and are very characteristic of the flowery language that LLMs usually use. flowery language. Does this mean that if we see one of these words in a paper it was written with AI? Not necessarily, but the increase is brutal. Researchers have compared the rise of ‘delve’ to other keywords, such as pandemic, which had a huge peak in 2020 and began to decline in 2021. The increase in the frequency of use of ‘delve’ is much more pronounced than all the others. It’s not coincidental. There is a stage in the process of creating a chatbot like ChatGPT that requires human intervention to fine-tune the responses; This is what is known as reinforcement learning from human feedback (for its acronym in English). RLHF). It turns out that most of the workers who are dedicated to this refining work are in African countries, such as Nigeria. guess where The use of these words in formal English is quite common. Exactly, in Nigeria. African style. ‘Delve’ is a fairly common word in business English in Africa, especially in Nigeria, and it is not the only one. There are also others like ‘leverage’, ‘explore’ or ‘tapestry’ that are more common in African English. According to 311institutealthough human feedback is very small compared to the enormous amounts of training data, it has a great impact since it is what defines the tone of the model when responding to us. Data labeling. It is a key step for training large language models and requires humans to be behind it. The problem is that the majority of workers who dedicate themselves to this are from impoverished countries such as Nigeria, Kenya or India, among others. In case the endless days and the ridiculous salaries were not enough, many times workers must review violent and very explicit imagesall without any type of psychological support. In Xataka | Being a porn moderator is not fun at all. He was exposed to “extreme, violent, graphic and sexually explicit content” Image | National Institute of Allergy and Infectious Diseases in Unsplash

ASUS just killed its phone line for good reason. Goodbye to the Zenfone

ASUS, the Taiwanese giant known for its computers, had been competing in the smartphone field for years. The Zenfone family is one of the ones that has convinced us the most at Xataka: phones with a ROM similar to that of the Google Pixel and top-of-the-line specifications. The proposal, solid on paper, has had low commercial success for years. And given the great opportunity in formats that AI allows, ASUS is clear: there will be no more Zenfone for a while. A harsh goodbye. ASUS has made a stoppage official in its smartphone division. It not only affects the Zenfone family, but also ROG Phone, its division of phones aimed at gamers. The company is now emptying its release calendar to leave a division whose future is uncertain. “ASUS will no longer launch new mobile phone models in the future.” Jonney Shih, president of ASUS. What’s wrong with my ASUS. If you have a company device, you will be covered. ASUS will maintain both guarantees and support for existing mobile phones, but there will be no room for new launches. And now what. The company’s strategy is clear: they are leaving the field of mobile phones to focus on new solutions and formats linked to artificial intelligence. The group’s revenue from its AI server business has doubled expected resultsand the robotics and smart glasses divisions will be the main beneficiaries of the death of Zenfone. Why is it important. ASUS’s strategy gives us a clue about where the industry wants to move in AI beyond the smartphone. The race in this field is being won by an unstoppable Gemini that has forced Apple to take its handand with Google distributing Android, competition beyond OpenAI does not seem possible. But in smart glasses Meta operates with its own technologies, Google has its proposals based on Gemini, and There is still room for participants who want to develop their own solutions. Even fiercer is the war in robotics, Why outside the smartphone. The mobile phone business represents a very small part of ASUS’ strategy. Their smartphones are niche products, and they never had it easy to conquer the general public. Abandoning a territory that is high in costs and difficult to obtain profits offers additional room for maneuver to invest in new product categories. Image | Xataka In Xataka | Artificial intelligence guide: main characteristics of the main AI models, points for and against, and comparison

We know that the Earth has been hit by 80,000 meteorites. For some reason, most end up in Antarctica

If we look at the global statistics of finds of meteorites on our planet We may think that they are distributed homogeneously throughout the territory, but the reality is very different. Official data indicates that of the approximately 80,000 meteorites cataloged all over the world, more than 50,000 have been found in Antarctica… And this raises a big question: does Antarctica have something special about having so many meteorites? A contradiction. Although we talk about 60% of the meteorites that have been found on Earth come from Antarcticacollision theory tells us another. Specifically, physics, which tells us that meteorites fall randomly and uniformly throughout the planet, so Antarctica does not receive more impacts than the Sahara Desert or the Pacific Ocean. So… Why do we find so many meteorites on the frozen continent? The answer lies in a perfect combination of glaciologyvisual contrast and a natural “trap” that is now, ironically, being sabotaged by climate change. The conveyor belt theory. To understand why Antarctica is the great archive of the solar system, you have to understand how ice moves. And the secret is not in how the rocks fall, but in how the ice delivers them to humanity. To do this, we must go to glaciological models and studies from programs such as ANSMET, where they point out that Antarctica It is a real meteorite conveyor belt. The process. In this way, a meteorite when it falls inside the frozen continent buried deep in the ice sheet. Once here, the natural flow of the glaciers will push the ice that stores the rock inside from the center towards the coast. At certain points, the ice encounters barriers beneath the glaciers, such as hidden mountains that slow its flow and forces the ice to return to the surface. And this is where the famous katabatic winds come into play, which are truly fierce and dry with a force capable of eroding the upper layers of the ice from solid to gas. The result. It is what scientists call the ‘Meteorite Stranding Zone’ (MSZ) or blue ice areas. It is nothing more than the part of ice that has been worn away, but has not affected the rock it stored in any way. That is why over time, meteorites that fell thousands of years ago and traveled trapped in the depths of the ice now appear on the surface as if someone had put them there. A contrast trick. Logically, finding a meteorite among a pile of red ones can be somewhat complicated in our environment. But when we talk about a black rock on a white sheet like ice, the truth is that visually it is easy to find it. That is why this contrast is the best ally that meteorite searchers have. The preservation. But beyond the fact that finding a rock the size of a walnut in the middle of the jungle is a really complicated task, it must be taken into account that humid climates degrade the meteorite quickly. Something that does not happen in Antarctica, which is technically a polar desert. The dry environment it has acts like a real freezer which preserves the samples almost intact for millions of years. This allows scientists to recover not only the rock, but pristine information about the origins of the solar system. And that is why all these factors together make it more common to find more meteorites in this location than in others, and not because there is a predilection for falling here. An invisible threat. As pointed out a study published in Nature, we have a serious problem on the table: We are losing about 5,000 meteorites a year. Intuition would tell us that if the ice melts due to climate change, more rocks would emerge. But the opposite is true due to the thermal properties of the meteorites themselves. Being dark rocks (and many of them metallic with high thermal conductivity), they absorb solar radiation much more efficiently than the surrounding ice. Even at subzero temperatures, the rock heats up enough to melt the ice just below it. This causes the meteorite to sink and create a small pool of water that refreezes them, burying the rock out of sight of researchers or satellites. Thermal models suggest that this disproportionately affects iron meteorites, which are especially valuable for understanding planetary cores, causing us to have many more chondrite or rocky meteorites. Race against time. Humanity has so far managed to recover 23,000 meteorites, giving us a large cosmic library that allows us to better understand everything around us. The problem is that the clock is ticking, and the most important part of the archive is beginning to sink, so now the most important thing is to hurry up to get the most valuable meteorites for us. Images | Kamran Abdullayev henrique setim In Xataka | In 2011, a collector bought a meteorite in Morocco. It has turned out to be direct evidence of thermal water on Mars

There is a reason why Germany allows driving at 300 km/h and it is not history or politics: it is the asphalt

If you like to step on the accelerator, you will have already seen firsthand that cornering at 100 km/h is better than at 130 km/h. I don’t need to remind you that the maximum speed allowed on state roads is 120 km/h. Although there are quite a few countries within the European Union with higher limits, Germany is the only state where there are sections without speed limit. 300 km/h without breaking a sweat. Obviously, this poses a danger to driving as cars such as a Porsche at 322 km/h. At these speeds, the risk of the car jumping or losing control is notable. But the “recipe” for manufacturing the German Autobahn has its particularities that allow it to offer enviable flatness and a road surface with high load capacity. And it can be found in the regulations and standards of the FGSV (Research Society for Highways and Transportation) and the BASt (German Federal Authority for Road Safety and Traffic). Blessed sandwich. While in Spain Flexible or semi-rigid pavement predominates with thicknesses of 40 to 60 centimeters. In Germany they use a standardized layer system called RStO 12 (Guidelines for the Standardization of Pavement Structures). That is, with a total thickness of between 70 and 90 centimeters with an antifreeze base composed of highly permeable gravel and sand so that water does not remain trapped (in case of freezing, it would generate large cracks as a result of expansion). About this, layers of gravel mixed with concrete or asphalt to provide sufficient rigidity to prevent collapse under the passage of heavy trucks. cwhen concrete and when asphalt. In the intermediate section the Germans use two materials, highlighting the concrete for those stretches of free speed and high truck traffic thanks to its rigidity and durability. In more detail: The 25 to 30 centimeter high-resistance concrete pavements longitudinally integrate plastic-coated steel bars. Thus, they allow some thermal expansion but do not allow them to move independently, causing steps. The transition between the concrete slabs is barely noticeable. Asphalt with stone matrix (S.M.A.), a combination with crushed stone and cellulose to offer extreme resistance to deformation and maximize the tire’s grip. The “superstructure” of German roads. Von Susan from Bielefeld, Deutschland – Straße, CC BY 2.0 Extreme plain for safety and by law. If you hit a speed bump at a certain speed, your car will go away. If you go 300 km/h in a sports car, the loss of aerodynamic load is such that it could be fatal. So Germany takes the plain very seriously by regulations: the maximum allowable deviation three millimeters in four meters. They achieve it with controlled pavers by global navigation and laser sensor systems. Auf wiedersehen, aquaplaning. Once the risk of steps, cracks and unevenness has been minimized, there remains another staunch enemy for speed: water on the asphalt. And they fight it in two ways. For starters, autobahns have a slope of at least 2.5% on the sides to evacuate the water as soon as possible. For concrete pavements, it is used waschbeton or washed concrete, a technique that brushes the surface to expose the aggregates, thus creating a rough, non-slip area that breaks up any water film that may form. In Xataka | Germany, Austria and Switzerland have plenty of roads. So they have started covering them with solar panels In Xataka | The Autobahn are the only roads in Europe without a speed limit. More and more Germans want to end them Cover | Wes Tindel and Nick Fewings

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.