OpenAI’s obsession was to train its models like crazy. Now it’s run them faster than anyone else

OpenAI has signed an agreement estimated to be worth more than $10 billion with Cerebras Systems, a startup that designs advanced AI chips dedicated to one thing: running AI models as fast as possible. It is a unique alliance not only because of that change of focus, but because there is a conflict of interests. what has happened. The firm led by Sam Altman has committed to purchasing 750 MW of computing capacity over the next three years from Cerebras. Sources cited in The Wall Street Journal indicate that this alliance has an estimated value of more than $10 billion. We are therefore facing an operation extraordinary in size, but peculiar in form and substance. What Cerebras does. The firm based in Sunnyvale, California, was founded in 2015 by former engineers from SeaMicro, purchased in 2012 by AMD. The startup designs artificial intelligence chips specifically aimed at the inference stage of AI models, that is, executing them. More tokens per second please. When we use ChatGPT or any AI model, what we are looking at is an AI model using inference. Some “write” faster than others, and that speed of displaying text in responses is measured in tokens per second. Typically NVIDIA chips are great for the training phase, but not so much for the inference phase. Chips from companies like Cerebras —or those of the well-known Groqwhich has just been “bought” by NVIDIA—are precisely designed to run those models at full speed and obtain very high token per second speeds. The AI ​​is already good. Now she wants to be fast. NVIDIA’s recent “purchase” of Groq makes it clear that Jensen Huang’s company wanted the ability to offer those ultra-fast inference chips, and now OpenAI seems to want something very similar with its deal with Cerebras. AI models have already achieved remarkable performance in many scenarios, and although they are not perfect, now companies want them to not only work well, but also work very very fast and their responses, even if they are long, appear almost instantly. OpenAI wants more computing power. This operation also helps Sam Altman’s company with another objective: to obtain (and reserve) as much computing capacity as possible in anticipation of the fact that demand for these AI models will grow non-stop in the coming months and years. According to WSJ OpenAI already has more than 900 million weekly users, and its managers have frequently commented that they continue to have computing capacity problems. Brains grow. This agreement reinforces Cerebras’ position in a market that clearly demands this type of solutions. The firm is negotiating a $1 billion investment round that would bring its market valuation to $22 billion, tripling the current valuation, which is around $8.1 billion. In the past it has raised $1.8 billion according to PitchBook. Conflict of interest. This agreement also draws attention for an important aspect: Sam Altman, CEO of OpenAI, is also an investor in Cerebras (he is at the bottom of this Cerebras website) and indeed your company At one point he considered acquiring Cerebras although in the end that operation did not bear fruit. We are therefore faced with an operation that theoretically benefits Altman on both sides, which is worrying. How will OpenAI pay for this party? This new agreement once again triggers the debate about OpenAI’s ability to meet its credit and debt obligations. In 2025 it generated about 13,000 million dollars in income, but that enormous amount remains minuscule if we take into account that the contracts it signed with OracleMicrosoft or Amazon They amount to about 600,000 million dollars that will have to end up getting from somewhere. Where from? It’s a good question. We’ll see if they can end up answering it. In Xataka | The alliance between Oracle and OpenAI is not just about data centers: it is about overtaking Google, Apple and Microsoft on the right

Science reveals that the weight returns four times faster than with a diet

The era of “miracle” drugs to treat obesity is entering a phase of crude scientific maturity, thanks to the time that has passed since its launch on the market. In this way, despite the years with big headlines pointing to great weight losses with Ozempic, science is now able to provide more answers to the key question What we should ask ourselves: what happens when we stop pricking ourselves? The problem. When a drug is newly released to the market, long term effects It is something that is not known exactly, since patients need to be taking them long enough to see the effects they cause. And above all the effect that exists when the drug is removed from the body. To answer this with Ozempic we have a study led by the University of Oxford which is not minor at all, since it has analyzed more than 9,300 adults in 37 different clinical trials. And the conclusion they have been able to draw is quite clear: patients regain weight when they stop treatment at a rate of 0.4 kg per month. The comparison. At first glance, this figure seems really low, but if we compare it with other methods to lose weight, we see that the magnitude of the problem is not minuscule. The study itself indicates that in behavioral programs, such as a diet and an increase in physical activity, the effect after its withdrawal is an increase of 0.1 kg per month. This way, the rebound effect of slimming drugs It causes you to return to your initial weight in approximately a year and a half, while a change in eating and sports behavior takes around four years. New generation drugs. But this is a simple average between the different medications on the market. This implies that within GLP-1 agonists we can see that the most powerful drugs also have a much greater rebound effect. For example, in the case of Wegovy or Mounjaro, where the initial loss was 14.7kg, the rebound was seen to shoot up to 0.8kg per month. An effect that tells us that the body tries to recover lost ground at twice the speed of previous generation drugs. Cardiovascular health. Beyond the aesthetic, science pointed out that these drugs had the ability to reduce the risk of heart attacks and improve the metabolic health. But it seems these effects are completely temporary. Specifically, the study has seen that approximately a year and a half after stopping the medication, the majority of cardiometabolic markers return to their levels before starting treatment. For example, blood pressure increases, diabetes markers reverse their improvement, and total cholesterol also returns to its risk levels. Why the rebound is so fast. The answer to this could lie in our own biology. Science believes that this effect is due to the fact that by injecting massive doses of GLP-1 agonists (a hormone that is produced in small quantities when we eat) we could be destabilizing our own cell receptors. Or we would even be blocking our body’s natural production of this hormone that gives us satiety. That is why when withdrawing the drugthe system does not have the ability to produce this hormone again in the same way as before (as if it had to turn the system back on) and that is why the body’s satiety system goes deaf. The result? Well, logically, the appetite returns with great intensity, causing the patient to eat much more food, since they are not satiated and in this way the weight increases again. The myth of the magic bullet. There are hardly any miracles in medicine, even though we say that these drugs are. And the reality is that these drugs are not the definitive solution for obesity, since real data indicates that the majority of patients stop treatment after 12 months due to its high cost, the fatigue of injecting or side effects. But in addition, there is no awareness that this treatment is a simple aid to self-regulation, but that logically it must be accompanied by a dietary change and physical activity that must be continued once the treatment is finished. If not, we can be sure that the injections will be of absolutely no use. A paradigm shift. This meta-analysis marks a turning point. Science tells us that GLP-1 is extraordinarily effective, but it is not a cure. If we treat them as a six-month “crash plan”, we are condemning the patient to a more aggressive yo-yo effect than any miracle diet of the past. The solution, according to Qi Sun and the Oxford researchers, is not only in the syringe, but in public policy: taxes on ultra-processed foods, aid in the purchase of fruits and vegetables, and urban planning that encourages exercise. Without a change in environment, the drug is just a temporary truce in a war the body is programmed to win. Images | David Trinks Towfiqu barbhuiya In Xataka | If you want a “miracle” weight loss drug, you no longer turn to Ozempic: the competition is beginning to surpass it

AI has allowed developers to program faster than ever. That’s turning out to be a problem.

Whoever has tried it knows it. Programming with AI can be wonderful. Especially if you have (almost) no idea about programming. This is where generative AI models have seen their first and probably only revolution. The developers were the first to be able to embrace this new technology. The appearance of GitHub Copilot in 2021 It showed us that it was no longer necessary to chop so much code, because the machine was already doing it for you, and since then the advance of generative AI in the field of programming has been overwhelming. The question is: has it been positive? The answer is not at all clear. It is evident that AI has allowed: That millions of people who were not programmers could turn their ideas for applications and games into a reality. That millions of professionals can save time by not having to write repetitive code (boilerplate) to focus on other more important and productive parts of your work The industry, of course, has been especially insistent with this vision of the transformation of this segment. Satya Nadella (CEO of Microsoft) and Sundar Pichai (CEO of Alphabet/Google) already boasted months ago that about 25% of the code generated by their companies is generated by AI. Meanwhile, Jensen Huang went further and made it clear that At this point no one should learn to program anymore because the AI ​​would do it for us. These are very forceful statements, but behind them lies another reality: that All that glitters is not gold in the world of AI for programmers. At MIT Technology Review they have spoken with more than 30 developers and experts in this field and have reached interesting conclusions. AI is a better programmer than ever. At least, according to the benchmarks In August 2024 OpenAI made a unique launch: presented SWE-bench Verifieda benchmark intended to measure the ability of generative AI models to program. At that time, the best of the models was only capable of solving 33% of the tests proposed by that benchmark. A year later the best models already exceed 70%. Current ranking of the best models according to the SWE-bench Verified benchmark. Several already pass 70% of the tests. Source: SWE-bench. The evolution in this area has been dizzying and we have witnessed the birth of that new modality programming called “vibe coding” and all the big ones have developed powerful programming tools to take advantage of the pull. We have OpenAI Codex, Gemini CLI, or Claude Code, for example, but they have been added startups like Cursor either Windsurfing who have also known how to take advantage of this fever for programming with AI. All of these tools promise basically the same thing: that you will program more and better. Productivity theoretically skyrockets, and while more code is certainly being written than ever thanks to AI, programmers They have gone from writing their own code to reviewing what machines generate. Recent studies reveal that veteran developers who believed they had been more productive actually they weren’t. Their estimate was that they had been 20% faster by being able to move forward without blockages, but in reality they had taken 19% longer than they would have taken without AI, according to the tests carried out. There is another problem too: code quality is not necessarily goodand as we say, developers must review that code before being able to use it in production. In the latest survey from Stack Overflow, one of the largest developer communities in the world, there was a notable fact: The positive perception of AI tools had decreased: it was 70% in 2024, and 60% in 2025. There are limitations, but even so everything has already changed Those interviewed by MIT Technology Review generally agreed with its conclusions. Generative AI programming tools are great for producing repetitive code, writing tests, fixing bugs, or explaining code to new developers. However, they still have important limitations, and the most notable is his short memory. These models are only capable of handling a fraction of the workload in professional environments: if your code is large, the AI ​​model may not be able to “consume” it and understand it all at once. For small projects, great. For large developments, probably not so much. The problem of hallucinations also affects the code, and in repositories with a multitude of components, AI models can end up getting lost and not understanding the structure and its interconnections. The problems are there, and they can end up accumulating and causing exactly the opposite of what they wanted to avoid. Several experts, however, explained in that text how it is actually difficult to go back. Kyle Daigle, COO of GitHub, explained that “the days of coding every line of code by hand are likely behind us.” Erin Yepis, an analyst at Stack Overflow, indicated that although this unbridled optimism towards AI has fallen somewhat, that is actually a sign of something else: that programmers embrace this technology, but they do so assuming its risks. And then there is another reality. One that is repeated day after day and that seems undeniable. The AI ​​we have today is the worst of all those we will have in the future. It may not be tomorrow or next week, but it is clear that the AI ​​you program will end up getting better and better. And there may come a point when those limitations disappear. Whether they do it or not, what is clear is that AI has changed programming forever. Image | Mohammad Rahmani In Xataka | OpenAI has turned ChatGPT into mainstream AI. In the business world the game is being won by its great rival

Nike wants to make slow runners faster. Your solution: powered sneakers

Nike has sneakers that you can put on without touching with your hands and even some that they tie themselves. The brand has just crossed a new frontier: that of motorized footwear that helps you walk and run faster. With a design reminiscent of an exoskeleton, Nike compares it to the operation of the electric bicycle. Project Amplify. This is how they have named this footwear system, which is currently in the testing phase. For its creation, Nike has partnered with the robotics company Dephy. It consists of a shoe and an ankle brace with a motor and rechargeable battery that transmits energy through a transmission belt. The shoe can be worn alone or with the ankle brace. Booster. What Project Amplify does is “increase the natural movement of the lower leg and ankle”, that is, it gives us a boost to be able to walk or run faster and for longer with the same effort. According to Nike, it is a system similar to that of electric bicycles whose motor assists us in pedaling and helps reduce the energy demand of the muscles. Nike says it’s like having “a second set of calf muscles.” Hacking speed. Nike has invested heavily in research and development to design footwear that drives athletic performance. Its previous innovations have focused on combining cutting-edge designs and advanced materials that offer a “rebound” or propulsion effect. An example is Vaporfly technology, which proved to be a determining factor in achieving records in major marathons. This is the case of the controversial record of Eliud Kipchoge, who ran a marathon under two hours. There is also the case of the Nike Super Spikes that were worn by several athletes during the Tokyo Olympics. The impact was undeniable: Up to three athletes broke the record in the 400 meter hurdles. Examples like this highlight the importance of footwear and opened the debate about the limits of technology in sports. Even It has come to be described as “mechanical doping”. For the slow ones. At the moment we will not see “motorized” athletes in the Olympics. Nike makes it clear that it is not a system designed for high-level athletes, but for amateurs. running with a rather low pace who want to go faster with less effort. We are talking about runners with a pace greater than 6 or 7 minutes per kilometer. Another use scenario for these motorized shoes is to be able to make urban trips more quickly, for example to go to work on foot. Images | Nike In Xataka | The Alicante sneakers that are succeeding in Silicon Valley and that have Zuckerberg as their best ambassador

It is not whoever shoots faster who wins, but whoever types better

The video game ‘Final Sentence’ is the perfect example that there are no mechanics that are too dry or complex: if the design is good and the packaging attractive you can have some of the content creators most relevant on the internet typing lapidary phrases without rest as if they were secretaries of some oil magnate. Typing for the masses in a title whose final version is not yet available but whose demo is already sweeping Steam. What is it. ‘Final Sentence’ is an independent video game with an overwhelmingly simple concept: a battle royale typing game in which up to 100 players compete to survive based on speed and precision when writingand where every spelling mistake is potentially fatal. Developed by independent Lithuanian studio Button Mash (actually just one person), the game creates an experience that some media have compared it to ‘The Squid Game’ or other games like ‘Buckshot Roulette’. The inspiration, according to your managercomes from your own clumsiness at the keys and the search for more entertaining ways to improve. Why has it been so successful? A series of factors have come together that have turned it into a bombshell. On the one hand, and above all, the launch of a playable demo as part of the latest Steam Next Fest in October. On the other hand, it uses a mechanic that we know well, the free-for-all, which is part of the essence of hits like ‘Fortnite‘, and washes his face. Being easy to understand and difficult to master (that is, anyone can start playing immediately), it also has an implacable brutality with errors: any mistake is severely punished. Which gives it both an addictive component and a viral character that has helped many content creators try it. The pressure of 100. He battle royale It is carried out against groups of rivals of between 40 and 100 players (although it is also possible to organize smaller private games). But the interesting thing is in the most populated ones: seeing how one after another the rivals finish their sentences, the players can feel the imminent elimination, since not only those who make mistakes are punished, but also those who are slower in finishing writing what they have been ordered to. Who has played. People as followed on the Spanish-speaking internet as IlloJuan, ElRubius, Genuine993, PNKeasy They have tried the game, but you just have to go around platforms like Youtube to check that there are thousands of videos with gameplays of the game, all squeezing out the component of tension and terror that this very peculiar concept has. The reason? The idea of ​​”write or die” is extremely juicy and practically any viewer, regardless of cultures and languages, can identify with it. Play to write. Since that legendary ‘The Typing of the Dead‘ that allowed you to connect a keyboard to the Dreamcast to undertake a literary version of ‘House of the Dead’, the very specific subgenre of “typing video games” has experienced multiple mutations, to the point of generating its own variants. There are, for example, online competitive titles (‘TypeRacer‘, ‘NitroType‘, ‘Ratatype Race‘), narrative and adventure games (‘Epistory‘, ‘The Textorcist‘, ‘Type to Continue‘) and games more oriented towards educational or casualas ‘TypingClub‘, the platform Typing.com or the hilarious Ztype. The important thing: that you have the keys well oiled, because you can see every work tool there… In Xataka | Transcribing at full speed with a keyboard with only 21 keys: the job of a stenotypist, according to someone who has been in it for 35 years

Nuclear fever goes faster than centrals

Uranium, for years relegated to a corner of the raw material market, Live a rebirth. His prices have shot themselves and investors chase him with enthusiasm, convinced that nuclear energy will be key in an electrified world. As Jennifer Hughes said in the Financial Times: “Investors in Uranium and scientists should have much in common: both seek great benefits from a small starting point.” However, this financial euphoria collides with an awkward reality: nuclear power plants are not built to the necessary rhythm and bottlenecks are huge. A small and overwhelmed market. The spot price is around 76 dollars per pound, After having exceeded $ 100 at the beginning of 2024. Much of this increase is explained by a small market: most uranium is sold under long -term contracts and the immediate market space is very narrow. At the same time, governments accumulate strategic reserves. In the podcast Stock Movers Bloomberg They detailed that The US Secretary of Energy, Chris Wright, wants to strengthen national inventories to reduce Russia dependence, which provides a room of enriched uranium that feeds the 94 US reactors. The result is clear: more uranium is purchased than the centrals can consume today, a sign that the geopolitical and financial appetite goes ahead of the real capacity. A spark that lights nuclear fever. Why this boom? The explanation is in global electrification. According to Bank of America analysts, by the end of this decade the world consumption of electricity will increase to 30%driven by the electrification of transport and the rise of AI data centers. According to the International Energy Agencythe data centers already consume about 415 twh per year – 1.5% of global electricity – and their weight will continue to grow with the expansion of AI. In fact, Nvidia, Microsoft, Google, among others need abundant, reliable and carbon free energy to support their operations. Hence, technology companies have taken an unusual step: Bet on nuclear. The turn of many countries. Nuclear energy has returned strongly, even in countries that had a firm predisposition to the closure. Germany He stopped his nuclear blackout plan and Belgium made the same decision. Indonesia, despite its wealth in coal, included nuclear in An energy investment plan of 235,000 million dollars. And the United States He has decided to quadruple nuclear capacity recycling uranium. Today there are about 440 reactors in operation in the worldwhich contribute about 10% of global electricity and are the second low carbon energy source after hydroelectric. The wall of reality: the deadlines. Political promises collide with industrial limitations. The projects are usually expensive and slow, with deadlines that do not fit with the climatic urgency. To this are added concerns about radioactive waste and fear of accidents such as Fukushima, Although even Japan is willing to return. In fact, in the US, only three reactors have been built in the last quarter of a century, two of them with exorbitant costs and significant delays. Today there is no plant under construction and to meet the objectives of Washington it would be necessary to initiate the works of 20 medium -sized reactors every year, According to Morgan Stanley calculations. Even China, famous for its speed When he decides to investit takes between five and ten years to design, approve and complete a new plant. Russia, the bottleneck of the nuclear cycle. The big problem is in the phase of the nuclear cycle that converts the mineral into useful fuel. There, Russia is the dominant actor. Although countries such as Australia (28%of world reserves), Kazakhstan (13%) and Canada (10%) large uranium deposits concentrateonly Russia Master the enrichment on a global scale. Canada emerges as an alternative. With mines in the Athabasca basin, the country not only extracts but can also enrich uranium, which makes it a “safe and reliable” supplier. His new mine, operated by Nexgen, could move to Kazakhstan as the world leader of production in the next decade. For its part, this last country accelerate your own nuclear plans. Kazakhstan has among his plans to build his first central in ülken, with financial support from Russia but technological alliances with France and South Korea, in an attempt to reduce the dependence of the Kremlin. Expectations ahead of reality. Uranium has gone from being a forgotten resource to become a central file of the energy and geopolitical board. Prices reflect it and investors bet strongly. But nuclear infrastructure slowly advance, the dependence of Russia in the fuel cycle continues to weigh and social resistances remain alive. As the energy expert warns in its columnwho invest in Uranium expect “too much, too soon.” The true nuclear energy boom, if it arrives, will take much more than a rebound in contributions. Image | Freepik Xataka | The largest nuclear fusion project on the planet has survived the setbacks. This is the date on which Iter should be ready

They have studied the effect of long -term sweeteners on our brain. His conclusion is that he ages faster

Little by little it is already becoming a daily gesture among many people: change sugar for a sweetener to avoid calorie consumption in excess. Whether in the morning coffee, in a yogurt or in a refreshing drink, sweeteners are attractive to respect the sweet taste and ‘be healthy’. However, a new and forceful study Posted in the prestigious medical magazine Neurology He puts this idea in check, suggesting that this substitution could have a long -term hidden cost for our cognitive health. A direct effect to thought. Research, which has established itself as one of the broadest and most prolonged to date on the subject, cooks that people with high consumption of sweeteners such as the aspartamosaccharin or sorbitol They experience a deterioration of their thinking and memory capabilities 62% faster than those people who consume. To put it in perspective, the researchers calculate that this accelerated decline is equivalent to aging 1.6 years suddenly. The details of the study. It is not a PSAJERA survey or a small -scale experiment. Scientists have been based on the Brazilian Longitudinal Health Study data of the adult (Elsa-Brazil), a mass and long-distance research project. They analyzed a cohort of 12,772 public officials with an average age of 52 years, which were followed for eight years, and with analysis at three different moments: 2008-2010, 2012-2014 and 2017-2019. Detailed questionnaires. Using food frequency questionnaires, the team quantified the combined and individual consumption of seven specific sweeteners: artificial ones such as aspartamo, saccharin and acesulfamo K, and sugar alcohols such as erythritol, xylitol and sorbitol, in addition to the tagatose. In parallel, the cognitive performance of patients with a six -test battery that focused on memory, verbal fluidity and global cognition was measured. The results. The consumption of sweeteners, both individually and combined, was associated with accelerated cognitive loss. The ‘suspect’ list includes some of the most common names we find on the labels of ‘Light’ products or ‘zero’: aspartamo, saccharin, acesulfamo k, erythritol, sorbitol and xylitol. Interestingly, the trend was more pronounced and statistically significant in participants under 60 years. This suggests, according to the authors, that median age is a critical window where the products that are chosen consumer may have direct consequences in brain health decades later. The researchers They point that until now the sweeteners without calories often “are seen as a healthy alternative to sugar.” But now it has been seen that great consumption of these has “negative effects on brain health over time.” There are limitations. The researchers themselves suggest that dietary data are based on self -reports, which can be inaccurate, and that, despite statistical adjustments, the “residual confusion” cannot be completely ruled out where other nutritional behaviors that may be interfering are not measured. Correlation is not causality. As expected, this study can generate a great debate, and the industry and the scientific community have called for prudence, remembering that correlation does not imply causality. Gavin Partington, general director of the British refreshing drinks association, and the International Association of sweeteners (ISA) They have pointed out that this is an observational study. That is, it finds a statistical association between two variables (consumption of sweeteners and cognitive impairment), but cannot demonstrate that one is the direct cause of the other. In Spain, experts such as neurologist Guillermo García Ribas, from the Ramón y Cajal hospital, They are cautious. He criticizes that it is difficult to isolate the effect of the sweetener of the rest of the diet. Often, a high consumption of these products goes hand in hand with a diet rich in ultraprocessed foods, which have already been linked in numerous studies to a worse cognitive aging. The defense of researchers. Anticipating this criticism, the Suemoto team offers two solid arguments. First, they observed that the association was also maintained for individual sweeteners, those that a person adds on their own to coffee or yogurt, and not only for the compounds used by the industry in the ultra -processed. Second, and perhaps more important, there is what scientists call “biological plausibility.” Previous studies carried out in animal models (mainly mice) have already shown that artificial sweeteners can trigger neuroinflammation processes and alter the crucial intestine-cerebro axis, mechanisms that could negatively affect brain function. The global context. This study does not arise in a vacuum. It adds to a growing wave of skepticism on the long -term benefits of sweeteners. In fact, in 2023, The World Health Organization (WHO) itself advised the use of these products to control the weight or reduce the risk of chronic diseases, arguing the lack of evidence on its long -term benefits and the existence of possible unwanted effects that had not yet been completed. The underlying problem remains the same: excessive sugar consumption. In countries like Spain, the maximum daily amount recommended by WHO is tripling. The sweeteners emerged as an apparent solution, but studies like this force us to ask ourselves if we are simply changing one problem for another. As Suemoto himself summarizes, his work “adds solid evidence that these compounds may not be harmless, especially when consumed frequently from the median age.” The conclusion is not that we should return to sugar, but that we must examine much more critically with what we are replacing it Images | Towfiqui Barbhuiya In Xataka | 9 questions and answers about Estevia, the fashion sweetener

Gas reserves advance faster than expected

Europe enters September with an unexpected energy mattress: gas deposits are already 76% of its capacity. Just a few months ago analysts doubted that the continent reached the legal threshold of 80% before November. Today, it not only seems safe, but it is even forecast that reservations could touch 90% if the weather accompanies. Short. According to Bloomberggas injections during spring and summer have been constant, thanks to a calmer market. The result is a level of reservations that, at least at least, the ghost of another energy crisis moves Like 2022-2023. Volatility has been reduced and European prices are maintained below those registered in August of the last two years. The Europe Infrastructure Gas (GIE) confirms that European storage was 76 %, equivalent to about 85 BCM, a level of less than 92 % last year, but aligned with the average of the last decade. A paradigm shift. First of all, Community regulations flexible. According to a June article by Bloombergthe EU will allow up to 2027 deviations of ± 10 percentage points compared to the objective of 90%, also extend in two months the deadline to complete it. This reduces the seasonal pressure of buying in summer at all costs, which in previous years fed speculation. For its part, the global LNG market played in favor of Europe. As an analysis of Ron Bousso for Reuters points outthe fall in Asian demand released cargoes that flowed towards the continent. China, traditionally one of the largest buyers, reduced its imports due to an increase in internal production and commercial tensions with the United States. In fact, Beijing It is promoting Massively the extraction of shale gas and deep perforations, which points to a lower dependence on external LNG in the coming years. Finally, the international offer grew. Bousso has pointed out that the worldwide capacity of LNG will expand from 550 BCM in 2024 to almost 890 BCM in 2030, led by the United States. Only in the first seven months of this year, their exports rose 22 % compared to the previous year, thanks to new plants on the Gulf coast. Germany, the weak link. However, the photo is not uniform. Germany, the largest European consumer, remains the most vulnerable link. A report from the anadolu agency points out that German reserves They are 67%, well below neighbors such as France or Belgium. The installation of Rehden – the largest in Europe – barely reaches 23 % filling, According to Bloomberg. The German delay has history: last winter left his deposits at 7%which forced to spend billions in their recharge. Berlin insists that his four floating terminals of LNG mitigate the risk, but the opposition warns of problems if a particularly cold winter arrives. The Russian unknown. The war in Ukraine continues to weigh on the energy board. Although Russian gas imports have been drastically reduced, LNG cargoes still arrive, sometimes under interposed flags. However, the European Commission presented in June a plan To gradually eliminate all Russian gas and oil imports before 2027, prohibiting new contracts since 2026 and ending existing ones no later than 2027. According to EurostatRussia’s quota in EU gas imports collapsed from 45% in 2021 to less than 20% in 2025, replaced by US flows, Norway and Algeria. The United States enters the equation. That turn has reinforced Washington’s dependence. A July of Reuters report explained that Brussels studies grouping the demand for European companies to negotiate joint purchases of American LNG, within the Aggregateeu mechanism. Under the objective of ensuring sufficient volumes and fulfilling the commitment to acquire up to 250,000 million dollars annually in American energy over the next three years. The Plan evidences EU’s strategy to replace Kremlin with Washington, although analysts cited by Reuters warn that the objective could be too ambitious and leave Europe vulnerable to political swings on the other side of the Atlantic. Forecasts The immediate panorama is more serene: traders expect a winter without shocks if the weather does not give surprises. But in the medium term, the dilemmas persist. As Ainvest has pointed outthe EU has added 70 BCM of new regasification capacity since 2022, but many terminals work at low capacity. There is the risk of assets stranded if the demand remains stagnant and the transition to renewables is accelerated. The possible output is in conversion: several projects seek to adapt gas infrastructure to hydrogen transport, within the decarbonization strategy. More robust reserves. Before a calmer market, at least for now. The continent has gone from the Russian dependence on a more diversified framework, with the US as a key partner and China releasing pressure in global markets. However, vulnerabilities persist: Germany as weak link, overcapacity in LNG infrastructures and the unknown of a climate that could alter the most optimistic forecasts. The 2022 nightmare seems distant, but the European energy transition remains a race against time. Image | Unspash Xataka | Natural gas has become essential in the AI ​​era, and this chart exposes countries with the largest reserves

How to use Gemini from the Chrome address bar to make faster queries

Let’s tell you How to use Gemini from the Chrome navigation barso that you can ask Google’s artificial intelligence questions saving the time to open a tab. This will be able to do it even if you do not have a payment subscription. You must understand that This is not Gemini’s complete integration in Chromeonly a first small part of it. The rest of the functions will arrive over time. We are going to focus on this concrete, which is basically a shortcut to AI. Gemini in Chrome’s navigation bar To be able to make Gemini a request at any time from Chrome, you have to click on the navigation bar and Write the command @gemini. This is the bar where you write the websites you want to visit, only in this case what you should write is exactly what we have told you. When you write @gemini and press on the space key on your keyboard, the navigation bar will change and The text will appear Ask Geminiand the writing field will tell you to enter a question. Now, what you have to do is write what you want to ask Gemini. Come on, you write @gemini, you press space and ask the question or write the prompt you want for artificial intelligence. Then, click intro to confirm the request. This It will open an automatically tab on Gemini’s website With the question you have asked, and without having to do nothing more, the AI ​​will begin to generate the answer to what you have asked. The bad news is that The rest of Gemini’s functions in Chrome are not yet available. This is a simple shortcut that does not even take into account the context of the tab from which you throw it. In Xataka Basics | Improve your Gemini images: 6 official tricks to create and improve the images you generate with artificial intelligence

It is less powerful than its previous version but it is faster and much more refined

The Tesla Model and Performance has been revealed. He has done it with a little surprise. The new version of the Model and Performance lowers the feet to the ground to deliver a little less power and, nevertheless, to manage them to be faster in 0 to 100 km/h. But that’s not, much less, the only thing that this new version brings. Why does an electric car have less autonomy than the announcing Tesla Model and Performance Technical Card Tesla Model and Performance Measures and weight 4,790 mm long, 1,920 mm wide and 1,624 mm high. 2,890 mm of wheelbase. Weight to be confirmed. Trunk 822 liters (Tesla does not specify if it measures to the ceiling). 116 liters in the front trunk. bodywork Five -seater SUV Maximum power 460 hp (dual motor and integral traction) MECHANICS Maximum speed of 250 km/h Acceleration (0 to 100 km/h) 4.8 s WLTP range 586 km WLTP consumption 15.3 kWh/100 km DGT environmental distinctive Zero emissions. Driving aids Autopilot system with adaptive cruise control with Stop & Go function, lane maintenance, dead angle sensor, emergency braking, front and rear cross traffic control. 360º camera and autonomous parking. Operating system Tesla software not compatible with Android Auto and Apple Carplay. Multimedia screen 16 -inch central screen. OTHERS Wireless load for mobile phone, two USB-C shots. Spotify, Disney+, Steam, YouTube applications, Netflix specifically developed. PRICE From 61,990 euros More refined at a very competitive price The tightening and tesla competition rises. The new Tesla Model and Performance is the most powerful and refined version of the company’s electric SUV that We already tried this year. The model, then, already presumed to refine the formula to offer a more mature version of the original concept. Now it seems to perfect this. The Tesla Model and Performance adopts the same formula as the Tesla Model 3 Performance. The car stays at 460 hp when it used to rise to 480 hp but is faster, lowering the acceleration 0 to 100 km/h in 3.5 seconds, two tenths faster than the outgoing model in this performance version. The maximum speed is 250 km/h. In the information sent by Tesla, the company talks about “an improved battery pack, equipped with new cells that offer a higher energy density, provides more energy without adding weight.” However, its size or capacity is not specified. Of course, an autonomy of 580 kilometers is advanced, according to WLTP cycle and an average consumption of 16.2 kWh/100 km. But the new performance is also more refined. To improve its dynamics, the electric car arrives with Achnid 2.0 forged tires of 21 inches staggered with integrated aerodynamic roofs for greater efficiency. They show the red brake tweezers. The car will also benefit from new shock absorbers and stabilizer bars that are coupled in an adaptive suspension As for design, abroad brings small novelties in its front and rear bumper that show that we are facing the most powerful version (like a small carbon fiber wing) but it is inside where it has received the most care. Now, the one who opts for the Performance version also takes the best screen, which scale until 16 inches and is offered with finer frames. The pedals are aluminum and the seats are also specific with extension on the thighs. All this with a highly competitive price. Their 61,990 euros They make it one of the most attractive options in the Power/Autonomy of the market if we take into account its size. In that price range we only have Chinese options in the market. The XPEng G7 for 56,500 euros and 476 hp of power but with a slightly lower size and the Smart #5 (of Geely) that with a price below 60,000 euros is the one that offers the most power, leaving the 500 hp, but has a higher consumption. On the contrary, if we look at the most powerful options of Hyundai Ioniq 5 or the Ford Mustang Mach E we find cars with much more power but a similar acceleration, a higher consumption and a slightly lower autonomy. All for a surcharge that goes well above at least 10,000 euros more than the version of Tesla. Photo | Tesla In Xataka | Byd wanted to have his hole in history. Solution: 3,000 hp to get the fastest electric car in the world

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.