Mobile Tech & Gadgets

Salmon have never taken so long to reach Asturias. And yes, it is as worrying as it seems

by

14 days. That is, two weeks after the season opened, we are still waiting for the capture of the ‘campanu’the first salmon to be traced by any of the five Asturian salmon farms. Whatever appears when it appears, it will be the latest campaign in history. No one is surprised that the Ministry have requested an expert report to decide whether to classify the species as “endangered.” What has happened? On Saturday, April 18, 2026, two weeks later than the traditional date, the season opened. The counseling deliberately delayed the start to “accumulate more salmon entries” (in the same way that reduced the number of specimens that can be caught to 154): after all, last year only 472 specimens were detected in the entire Principality. And the problem is not only Asturian. In Cantabria (where 38 salmon can be caught) none have been caught either. Although it must be pointed out that in the Cantabrian rivers the ban opened this May 1st. And in Galicia a total ban was declared for 2026 (although some exceptions have been made). Why is all this happening? The causes are well known: we are talking about things like river fragmentation, pollution (agricultural, industrial and urban), the effects of aquacultureuncontrolled repopulation, invasive species and, of course, climate change. So, are we going towards a total ban? The truth is that no one can know. Especially because we have a very close precedent: Spain has repeatedly refused to ban eel fishing despite the fact that all scientific reports say that it must be done. In favor of the salmon it plays that, thanks to the farms, the money that the wild moves is rather symbolic. But the paths of agricultural regulation in an election year are inscrutable. What is clear is that these are not good times for wild salmon. Not in Spain, not anywhere. In Norway, for example, only 323,000 wild salmon were observed in 2024. The previous year, the figures They amounted to 481,463 copies. In Scotland, another of the great salmon-growing countries, the population of wild specimens has fallen by 80% since the 70s; and, in the Faroes, the total ban is on the table. As we said recently, there have never been more salmon in the world. And, for that reason, this species has never been closer to disappearing in the wild. Image | Brandon In Xataka | We are drugging the salmon with cocaine and anxiolytics. And that’s causing them to behave strangely.

They came to believe that the horoscope predicted true love. It was actually a bug in the 2001 British census.

by

When searching a new lovethere are many people who are looking for the best possible compatibility, but not only in terms of tastes or hobbies, but also to the zodiac sign that corresponds to each of the people. Nowadays there are dating applications that allow you to filter by zodiac sign, since it is not uncommon to see or hear that being a Leo or Scorpio sign is something catastrophic. But science has something to say here. Demonstrating it. When asked if the zodiac sign affects the compatibility that exists in a couple, researcher David Voas wanted to give him an answer to see if it is a definitive tool or not. And to do so, he analyzed a sample of 20 million people from the census of England and Wales, which is equivalent to 10 million couples. This way, if there is any relationship in such a large sample, it has to be seen somehow. And here we didn’t want to see if the fire signs get along with each other, but rather something more basic: is there any combination of signs that occurs more or less frequently than pure chance dictates? The first results. At first glance at the different couples analyzed, it seemed that the astrologers were right and that people were grouped according to the affinity of their zodiac signs. But the truth is that the researcher did not stop at this and continued digging into the data in order to see if there was something more. The real results. Upon further investigation, he discovered that this “astrological sign” did not come from the cosmos, but from imputation errors and biases in the collection of census data. Here he observed that in records where the exact day of birth was missing, officials used to assume that it was the 1st day of the month by default, which generated artificial accumulations of people in certain signs and created patterns that were not real. With all this, it was seen that many apparently compatible couples were simply sharing registration errors or rounded birth dates, which is something quite common within the administration. And once these statistical artifacts were corrected and matches by birth month were separated from matches that actually crossed the boundary between zodiac signs, the effect disappeared completely. In short, there was no trace of a “force” that united the signs. It’s always the same. This new research sits in a long tradition of scientific attempts to validate astrology in some way, and the truth is that it always comes to the same conclusion: there is no relationship. As Carlos Orsi explains in his work published by Columbia University Press, the problem with couple astrology is that not even astrologers themselves agree, since there is no homogeneous theory about which combinations are “good.” And given this lack of consensus, the Voas test is the fairest possible: look for any deviation from chance. And chance won by a landslide. Images | lookstudio in Magnific freepik In Xataka | Esperanza Gracia had been explaining the horoscope to the Spanish for three decades. Its closure illustrates a deeper change

Murcians and Castilian-La Mancha have been fighting for nothing for years. Whatever happens with the transfer, what we are really losing is time

by

Ultimately, this is the story of a deception. Since 2019, the Supreme Court has been saying exactly the same thing: the application of the European Water Framework Directive forces Spain to change the way it manages its transfers. And he hasn’t said it once, no: if we talk about the transfer of the Tagushe has said it, at least, six times. Despite this, the different administrations have been interpreting a political melodrama for years that has prevented the design of a system that minimizes the problems that the directive may create. And the result is that Murcians and Castilian-La Mancha They have been fighting for nothing for years. Fortunately or unfortunately, this race forward seems to end on May 5. What happens on May 5? If everything goes as planned, on May 5 the Supreme Court will decide the future of the Tajo-Segura Transfer and the Tajo Hydrological Plan 2022-2027. That day, the high court will decide what happens to the appeal of the Central Union of Irrigators of the Tajo-Segura Aqueductthe last major judicial process that remains open against the changes that the Government approved in 2025 to adapt to the regulations. It is, so to speak, the last legal bullet left for the irrigators of the eastern peninsula. And what can we expect? Bit. The president of the union himself, Lucas Jiménez, has publicly admitted ‘cold spirits’ and ‘without great expectations’, given the meaning of previous pronouncements. And at this point, the issue being debated is whether the new ecological flows (which, according to the University of Alicante, will entail an average loss of 105 hm³/year from 2027) come into force now or may be staggered. But, the unpopularity of the measure in large areas of the country has caused everything to be postponed. To the point that the National Court just admitted to processing Castilla-La Mancha’s appeal for the Ministry’s inaction in publishing the new rules: in fact, if Scrats’ appeal is overturned tomorrow, there will be no rules to apply the transfer. And then? The conflict will enter a new phase: given the eventual rejection and with the transfer cuts legally consolidated, all that remains is to discuss technical details and compensation measures. We must not forget that the Transfer supplies almost 150,000 hectares irrigation in Murcia, Alicante and Almería. This is water that is already de facto granted to irrigators and the State will have to compensate them. Although, of the 1,450 million euros that Moncloa committed to cushion the blow, it seems that only around 5% has been executed. The story that never ends. We have been fighting over water in Spain for decades and we have been unable to create a system that reorganizes the country (and adapts it to real water). Almost the opposite: for more than 30 years, it has never been like this. As explained in Datadista“since the deep drought of the 1990s, each dry period has served to implement emergency measures (…) or allow practices that were not eliminated when the rains returned, they were used to expand irrigation, increasing the problem of overexploitation and contamination of aquifers and the wetlands they feed.” And the bill for all that is what we are paying now. Image | Trent Haddock In Xataka | The Tagus reservoirs have reached their maximum level. The response of the authorities has been to empty them immediately

US sanctions have left its current market share at 0%

by

In just two years Nvidia has gone from leading the GPU market to artificial intelligence (IA) in China to not sell practically anything in this country. In fact, Jensen Huang, the CEO of this company, has confirmed just a few hours ago now Its market share in this Asian nation is 0%. This dramatic scenario for Nvidia is the result of two decisive strategies of the governments of China and the United States. Early October 2024 the Chinese Administration sent to its AI companies a recommendation asking them to use chips produced in China as much as possible. Ten months later this recommendation became a requirement. And the Chinese Government is already forcing state-owned data centers throughout the country to use at least 50% Chinese integrated circuits in their servers. The Administration led by Xi Jinping has made this decision because it can afford it. And it is that It already has three very clear alternatives to Nvidia: Cambricon Technologies, Huawei and Moore Threads. This panorama has led Jensen Huang to decide to openly criticize the US export policy. AND has done it on several occasions during the last few years. The head of Nvidia is not enough to be the most attractive option for his Chinese customers; It also has to deal with the decisions of the US Government. US sanctions on China are destroying Nvidia Jensen Huang holds that the US will not protect its technological hegemony by blocking AI chip exports; According to this executive, what Donald Trump’s Government must do is ensure that the American AI ecosystem is dominant worldwide. The current scenario proves him right, but at the moment nothing indicates that the US Administration is going to bet on its strategy. At least not in a consistent way from a practical point of view. And the US Department of Commerce does not give the slightest respite to American AI chip designers. When these companies receive an order from one of their Chinese clients must apply for an export license to this government entity and indicate which GPU they intend to send to China, their specifications and which client is going to use them, among other relevant information. The Office of Industry and Security is also responsible for carrying out investigations into the tariffs deployed by the Administration led by Donald Trump. Once the bureaucracy has been put in place, Department of Commerce technicians analyze export applications in the framework established by current regulation and approve or deny the sale of integrated circuits to China. This is the usual procedure, so there is nothing new up to this point. However, Nvidia, AMD and other American AI chip designers face a very serious problem: the Commerce Department takes several months to process their export licenses. The staffing of the Department of Commerce has been drastically reduced in recent months, and in the current context this scenario represents a very serious problem. The Industry and Security Office of this entity is not only responsible for processing export licenses linked to AI chips; is also responsible for carrying out tariff investigations deployed by the Administration led by Donald Trump. And with fewer personnel than in 2024 and 2025 it cannot cope. According to Bloombergthe Office of Industry and Security has lost 101 employees in recent months, which represents a 19% reduction in staff compared to what it had in 2024. Curiously, the staff specifically dedicated to developing standards linked to the semiconductor industry and reviewing applications for export licenses has decreased by 20%, although at the moment it has not emerged what is the reason for this staff drain. Be that as it may, during 2025 the Office of Industry and Security took an average of 76 days to resolve export requests, but this period is increasing in 2026. Very bad news for Nvidia and AMD. Image | NVIDIA More information | Tom’s Hardware In Xataka | The US remains committed to stopping China. Now it has targeted the second largest Chinese chip manufacturer

If the question is how to stop the bleeding of emptied Spain, in Cantabria they are clear: subsidizing festivals with bulls

by

If you have learned anything ‘Spain emptied’ At this point there are no magic recipes against depopulation. In recent years, the administrations of that rural Spain that is gradually emptying have tried everything, often without success: from offering free employment and housing to assume the management of basic services, such as gas stations and stores. Now in Cantabria they have decided to add a new strategy against rural exodus to that list: subsidize bullfighting celebrations. At the moment the measure does not seem to have served to attract new neighbors. What it is generating is controversy. A cape for bullfighting. To understand the controversy we must go back to April 28, when the Official Gazette of Cantabria (BOC) published a call of subsidies from the Ministry of the Presidency. It basically announces a sum of 41,000 euros to “promote bullfighting in rural areas.” For this purpose, the Government offers to cover up to 90% of the expenses of the festivals that revolve around the bull, with amounts that range between 2,000 and 14,500 euros, depending on whether they are bullfights, bullfights, bullfights or other “popular celebrations.” It matters what… and it matters where. So far nothing exceptional. Spain has been immersed for years in a debate (sometimes bronco) on bullfighting and whether or not it should be supported with public funds, but the Cantabrian initiative does not stand out precisely for its budget. In 2025, without going any further, the Community of Madrid approved a game of 1.7 million of euros to support the Bull Festival. What is striking about the Cantabrian case is that its objective is not only to support the ‘national holiday’. In fact, that is not even the main argument made in the call of the BOC. Its purpose is another: to fight against depopulation. The subsidies are specifically directed at the 41 municipalities of the community at “risk of depopulation” and their objective appears clearly described in the official bulletin: “Encourage the aforementioned local entities to have resources that energize their social and economic life, such as bullfighting shows.” In short, use the bullfights, bullfights, bullfights, bullfights and other shows with bulls to revitalize the economy and establish population. “Put them on the map”. In case there were any doubts, the counselor of the Presidency, Isabel Urrutia, recalled a few days ago that last year the Cantabrian Government financed bullfighting celebrations in four small municipalities of Cantabria, which in his opinion allowed “to put them in the focus of the bullfighting world.” “We help with small aid to fight against depopulation and put these municipalities on the map, many of them with a great tradition of bullfighting. The aid is fulfilling its objective,” argues the counselor after remembering the case of Pesaguero: the town has 400 inhabitants, but in 2025 1,800 fans attended its bullfighting show. What exactly do they subsidize? Low the argument that bullfighting can become a “stimulator of social and economic life”, the Cantabrian Government offers to assume up to 90% of the expenses of organizing the festivities, as long as they do not exceed certain limits: 14,500 euros in the case of bullfights or bullfighting, 10,000 if we are talking about bullfights with picadors or bullfighting of bulls, 6,000 for bullfights without picadors, calves or festivals and 2,000 for similar shows. “To award aid, the Government will take into account the classification of the municipality as being at serious risk of depopulation or special and differentiating treatment for this reason, and also the type of show or celebration,” duck the regional executive. Although it recognizes that there are 41 municipalities that meet the depopulation requirements to qualify for aid, in 2025 only bullfighting shows were subsidized in four town halls of the region: Pesaguero, Tudanca, Rasines and Bárcena de Pie de Concha. In 2024 there was one more, Molledo. This would be the third consecutive year in which subsidies have been announced that, they insist From the regional government, they are fulfilling the objective with which they were set. Opinion division. Not everyone thinks the same. The Franz Weber Foundation has questioned that the initiative really serves to strengthen the economy of those localities or combat rural exodus, and provides data as proof: the number of residents who have won the subsidized town councils can be counted on the fingers of one hand. Literally. “Four inhabitants in three years”, resume the organization, which estimates the funds mobilized between all the calls at 132,000 euros. The tables from the INE confirm that Bárcena de Pie de Concha only gained one neighbor between 2023 and 2025 and Rasines another five. Pesaguero and Tudanca lost population. “They neither fix population nor suppose a real dynamic activity,” ditch the foundation. “The autonomous Executive has dedicated around 132,000 euros since 2024 under this excuse, but the population reality in the municipalities awarded in different calls shows an evident inability to have a positive impact.” In your opinionthe real purpose of the Cantabrian Government is another: “Support bullfighting using subterfuges such as depopulation.” Images | Alex-David Baldi (Flickr) and Arild Andersen (Flickr) In Xataka | The great debate about the future of bullfighting is not in Spain, but in an unexpected country: South Korea

In the year of the World Cup, the brand is betting everything on RGB MiniLED

by

Hisense reaches 2026 playing a good part of its range on a single card: the RGB MiniLED. The Chinese brand has renewed everything your television catalog and has presented it to society exactly where it should be done in 2026: at the FIFA headquarters in Zurich, the same year in which the entire world will be glued to a screen watching the World Cup. This movement is not coincidental. According to Omdia data provided by Hisensethe manufacturer ranks second in the world in total television sales and first in the 100-inch or larger segment between 2023 and 2025. From that position, the brand has designed a catalog that ranges from a 116-inch giant to the most affordable models with MiniLED, through two high-end series that are going to be a lot to talk about in 2026. 116 UX RGB MiniLED Evo UR9 UR8 panel RGB MiniLED Evo (4th subpixel cyan), VA, 4K UHD, 8-bit + FRC RGB MiniLED VA 4K UHD, 8 bits + FRC, 180 Hz and 16:9 RGB MiniLED VA 4K UHD, 8 bits + FRC, 180 Hz and 16:9 resolution 3,840 x 2,160 points 3,840 x 2,160 points 3,840 x 2,160 points size 116 inches 65″,75″,85″ 55″, 65″, 75″, backlight RGB MiniLED Evo, FALD up to 8,000 nits, 3,584 dimming zones RGB MiniLED FALD, up to 4,000 nits RGB MiniLED FALD, up to 3,000 nits hdr Dolby Vision 2, Dolby Vision IQ, HDR10+, HDR10, HLG Dolby Vision IQ, HDR10+, HDR and HLG Dolby Vision IQ, HDR10+, HDR and HLG processor Hi-View AI Engine RGB Hi-View AI Engine RGB Hi-View AI Engine RGB operating system VIDAA U9 VIDAA U9 VIDAA U9 sound 2 x 15 watts + 2 x 10 watts + 2 x 5 watts + 2 x 15 watts + 2x 10 watts Dolby Atmos, DTS 2 x 15 watts + 2 x 10 watts + 20 watts + 2 x 15 watts + 2x 10 watts Dolby Atmos, DTS 2 x 10 watts + 2 x 5 watts + 20 watts Dolby Atmos, DTS connectivity 3 x HDMI 2.1 3 x HDMI 2.1 4xHDMI 2.1 wireless connectivity Wi-Fi 6 Bluetooth 5.0 Wi-Fi 6 Bluetooth 5.0 Wi-Fi 6 Bluetooth 5.0 price Not available Not available Not available FIFA, the World Cup and TCL in the background Hisense has renewed for the third time consecutively its official sponsorship with FIFA after the 2018 and 2022 editions, and in 2026 it goes further: it will be the official and exclusive supplier of RGB MiniLED TVs for VAR Video Operations Rooms throughout the tournament. Romy Gai, FIFA’s chief business officer, said the organization “partners with Hisense to welcome the best display technology to deliver an unprecedented World Cup experience for billions of fans around the world.” Hisense is not the only one bet on the king of sport as a sales driver by 2026, its competitor TCL has been official sponsor of the Spanish Soccer Team from 2023 and expanded that agreement in October 2025 to include new products and a renewed contract. The television market anticipates one of its best years in volume precisely due to the World Cup effect, and the big Chinese brands They are well positioned to take advantage of that momentum. The pie that manufacturers share in the Soccer World Cup is not small. It is estimated that the match played between France and Argentina in the 2022 World Cup in Qatar had a hearing of 1,420 million viewers and the tournament registered a average of 2.9 billion viewers from the different television signals. All glued to a television. That’s a lot of televisions. The 116UX and the fourth color that changes the image The most advanced model in the 2026 catalog is the 116UX, a 116-inch television that incorporates the RGB MiniLED Evo panel that Hisense already advanced at CES in Las Vegas. Its particularity compared to the conventional RGB MiniLED is the incorporation of a fourth cyan subpixel, an intermediate color between the blue and green that the three traditional colors cannot reproduce accurately enough on their own. Thanks to this fourth subpixel, the 116UX manages to expand the coverage of the BT.2020 color space, the standard used in professional film production. The change applied by Hisense has certain parallels with what LG has been doing for years in its WRGB OLED panelswhere a white subpixel is added to improve brightness. The difference is that in the case of the RGB MiniLED Evo the objective is not brightness (something that is necessary in OLED) but rather to expand the volume of color available, covering complex tones that were previously only approximated by combining the three RGB subpixels. The result is an image with greater fidelity in skies, vegetation and skin, exactly the elements that make a sports broadcast look more natural. With this new panel, the 116UX is positioned as the brand’s flagship for the domestic field of large-inch models with 116″ and 100″ diagonals. UR9 and UR8: the flagships in one size below 85 inchesHisense proposes two models within its high range: the UR9 and UR8. Both models are a technological showcase of what Hisense is capable of offering in its 2026 catalog, lowering RGB MiniLED technology to more accessible price ranges for users. Both mount RGB MiniLED panels (without the last name Evo and the fourth pixel, which is reserved for the UX), the Hi-View AI Engine RGB processor, and sizes of 65, 75 and 85 inches in the case of the UR9, and from 55 inches for the UR8. One of the arguments of the panels Hisense RGB MiniLED is the improvement in color representation and increased brightness. However, here is the main difference between the UR9 and UR8, depending on the brand, the UR9 can reach peaks that exceed 4,000 nits with 1,056 local dimming zones, while the UR8 would have its ceiling at 3,000 nits. Beyond that difference, both the UR9 and the UR8 share some elements that place them in a different category from the … Read more

Meta and Google talk about nuclear fusion for the future; The short-term reality is that they are pulling natural gas

by

Silicon Valley has an undeniable gift for selling the future. If one listens to the great technological leaders, Artificial Intelligence will soon be powered by energy sources worthy of a science fiction novel. Goal just signed an agreement to obtain solar energy directly from satellites in space, while figures such as Sam Altman, CEO of OpenAI, They assure that nuclear fusion It is the great “silver bullet” that will save the sector. However, it is enough to look down from the stars to the earth to find a much smokier reality. To feed the insatiable “energy monster” that AI has unleashed, big technology companies are turning to the technology of the past. As explained from Axiosthe race to dominate artificial intelligence is accelerating at such a dizzying pace that the industry’s ambitious climate goals are taking a discreet backseat. Today, the world’s most sophisticated cloud is being built on a foundation of fossil fuels. The numbers speak for themselves. Far from nuclear fusion laboratories, the actual infrastructure being built in the United States tells a story based on natural gas. Meta’s case is perhaps the most graphic, as detailed in Bloomberg, US utility Entergy Corp. has had to increase its capital spending plan by almost a third, reaching $57 billion, to build 10 new natural gas plants dedicated exclusively to powering the new data campus Hyperion of Meta in Louisiana. This gigantic complex will require more than 7 gigawatts of power, the equivalent of the output of seven large nuclear reactors. Google, the historic champion of clean energy, is not far behind either. An investigation by the market intelligence firm Cleanview has brought to light Google’s partnership with the company Crusoe Energy to develop a huge data center in Texas named “good night“. The project includes a 933-megawatt gas plant built outside the traditional electrical grid. The end of the green utopia? The environmental impact of this installation is not minor, how to explain Guardianthe plant will emit up to 4.5 million tons of carbon dioxide per year. To put it in perspective, this exceeds the annual emissions of the entire city of San Francisco or is equivalent to putting 970,000 additional gasoline cars on the roads. Given this, Google’s official position is cautious. Chrissy Moy, company spokesperson, does not deny the project before the mediaalthough it clarifies that, although they are linked to the campus, they still “do not have a contract in force” to acquire energy from said gas plant. How have they developed in oil pricethe origin of this sudden gas rush is that data centers are putting local power grids under unprecedented pressure, causing consumers to bear the cost of this increased energy competition. To overcome the slow expansions of the public network and the endless waiting lists for permits, Wired points out that data center developers They are choosing to generate their own energy “behind the meter” (off-grid). And in that fast and private strategy, gas is king. Their green mask falls off. This is a serious blow to Silicon Valley’s green image. As you remember GuardianGoogle was once a pioneer in promising net zero emissions by 2030. However, the company itself has had to admit that its carbon emissions have increased by 48% in the last five years due to data centers. Now, those environmental objectives have been internally downgraded to the category of climate moonshots (speculative projects very difficult to achieve). The underlying problem is purely physical. As he reflects Impakterenergy—not chip shortages—is emerging as the real bottleneck for AI. Traditional renewable sources are intermittent, and large language models require devouring electricity 24 hours a day. A systemic problem that is already raising blisters in Washington. The return to natural gas is not an isolated anecdote of a couple of companies. There are currently about 100 gigawatts of gas-fired power in development in the United States destined for data centers alone. Microsoft just signed a deal with oil giant Chevron in Texas, and permits for OpenAI’s Project Jupiter in New Mexico suggest it could emit up to 14 million tons of greenhouse gases annually (triple that of Google’s project). Faced with this fossil avalanche, Democratic senators such as Whitehouse, Van Hollen and Heinrich have sent letters demanding formal explanations from leaders of Meta and OpenAI for putting the country’s climate commitments at risk. The industry defends itself by arguing that it is a necessary evil. Cully Cavness, president of Crusoe, explained that natural gas it is a critical “bridge” and the only power source available today capable of scaling at the pace AI demands. Next-generation clean alternatives will take decades. Meta’s promising agreement to receive solar energy from space will not have a pilot satellite until 2028and its commercial viability is not expected, at best, until the 2030s or 2040s. The same happens with commercial fusion reactors: they will not dump a single watt into the grid well into the next decade. The great paradox of AI. Business magazines celebrate the financial success of this revolution. In their profiles of the most influential companies, TIME relates how Google, under Sundar Pichai, has reached a $4 trillion market value driven by its advances in AI, while Mark Zuckerberg celebrates record ad revenue on Meta by promising systems that will soon “understand the unique personal goals” of each user. Silicon Valley promises that this same Artificial Intelligence will one day help us solve humanity’s great challenges, including climate change itself. But the current paradox is inescapable: in the real world of 2026, to train the most brilliant and avant-garde artificial mind ever created, human beings still inevitably need to set natural gas on fire. Image | Photo by Tasos Mansour on Unsplash Xataka | Solving the mystery of the red balls on high-voltage cables: a simple way to save lives

Mistral has a new AI model. The good news is that it is absolutely European; the bad one, which is absolutely mediocre

by

The French startup Mistral has just launched Mistral Medium 3.5an open-weight AI model that is the great European exponent in an industry absolutely dominated by China—which competes directly with this type of projects—and by the US. And if this is the best they can do, it seems Europe has a problem. Mediocre. This is a “dense” model with 128 billion parameters and a context window of 256,000 tokens. While models with Mixture-of-Experts (MoE) architecture only activate a subset of the total parameters to achieve enviable efficiency and capacity, Mistral activates them all. That makes it much less efficient, but theoretically it should make its performance promising. And that’s the problem. Which it is not. Benchmarks. Pedro Domingos, professor of deep learning at the University of Washington, he expressed it very well: “Mainstream AI companies brag about how their model is much better in benchmarks. Soo Mistral brags about how their model is much worse.” It is true that the models with which it is compared are larger in total number of parameters, but as we will see later, even taking that into account, they are cheaper and theoretically more efficient thanks to the use of that MoE architecture in many of them. The model, however, unifies the previous catalog and follows the market trend of being able to establish the desired level of reasoning (reasoning_effort) as a parameter. Bad results. And he is somewhat right: Mistral does not seem to have problems showing the results of various benchmarks in which it performs poorly, but it also performs poorly with models that are by no means the most recent or powerful on the market. Thus, it is compared with Claude Sonnet 4.5/4.6, with Kimi K2.5, with GLM-5.1 or with Qwen 3.5 397B. In almost all cases (except GLM 5.1) there are already more recent and powerful versions of all of them. Not so far from local models. In fact Medium 3.5 scored 77.6% in SWE-Bench Verified, a programming test in which Qwen3.6-27b It reaches 72.4% with a fundamental difference: you can run it “for free” (with the appropriate hardware, and you paying the electricity bill) with a relatively affordable machine. More expensive (and somewhat more restrictive). If we use it via API, Mistral Medium 3.5 costs $1.50 per million input tokens and $7.5 per million output tokens. GLM-5.1 costs 1.4/4.4 respectively, and Kimi K2.5 costs 0.5/2.8 respectively. Its recent successor, Kimi K2.6, costs 0.95/4, and it is significantly better than Mistral being cheaper. There is a curious fact: Mistral uses a “modified MIT license” instead of the traditional Apache 2.0, and indicates that this model can be used commercially or non-commercially except for “high-income” companies. Chasing Anthropic. In addition to the model itself, the company has presented the so-called remote scheduling agents using Mistral Vibe CLI to, for example, send pull requests to GitHub in an automated way. It also has the so-called “Work Mode” in LeChat, allowing multi-step tasks to be managed autonomously. These are tools clearly intended to strengthen Mistral’s role as a base for scheduling agents, which is the path that has worked fantastically for Anthropic. Your advantage: being European. The only great strength of this model is that it has been developed by a European startup, and that gives it clear visibility at a time when many EU countries they talk about digital sovereignty. It is the only Western model that seems to want to compete with China in the field of open weight models, which is good news, but the truth is that in terms of performance it does not seem that the Mistral Medium 3.5 is going to perform competitively. The geopolitical security network. That, together with the fact that it costs more than its competitors, makes the decision to use it difficult unless for those who prioritize clearly that European origin. That is Mistral’s ace in the hole, and they are taking advantage of perfectly. The company has recently obtained financing to create data centers in Europe, and is nourished and fed by this new obsession with minimize dependency of North American Big Tech. In Xataka | The CEO of Mistral sends a message to Europe: the end of being the technological vassal of the United States

guide them to the exit one by one… with a surprise

by

In the late 80s, during the call “tanker war”the United States even escorted civilian ships in the Persian Gulf under flags changed in a hurryin an operation so delicate that any error could have triggered a direct conflict between powers. Even so, the simple transit of each ship became an almost surgical operation. A blocked strait that paralyzes half the planet. Since the start of the war in Iran, the Strait of Hormuz has become a critical bottleneck for the global economy. It is estimated that at this time there are around of 1,000 ships who remain trapped since the beginning of the conflict and about 20,000 sailors They remain on board with increasingly limited supplies. Not only that. The blockade imposed by Iran after the outbreak of war with the United States and Israel has reduced traffic at minimum levels and has stressed energy markets. The situation is such that analysts warn of a possible breaking point if the situation continues, since a fifth of the world’s oil normally circulates through this route. The US plan. In this scenario, Washington has launched the so-called as “Project Freedom” with an idea as direct as it is risky. Free trapped ships guiding them one by one outside the strait, yes, through coordination between countries, insurers and logistics operators. However, after the announcement it was learned that the key to the plan is what it does not include and that has become its great “surprise”: in reality there will be no naval escorts traditional boats accompanying each ship, according to Washington officials. Although the United States deploy destroyersaircraft and thousands of troops in the region, traffic will rely more on traffic management and indirect deterrence than on direct armed protection. Diplomacy, threats and an extremely fragile balance. The movement also occurs in parallel to indirect negotiations between Washington and Tehran. The messages at the moment are rather mixed, combining diplomatic optimism on the one hand… with strong warnings from both sides on the other. On one side, the United States insists that it will respond forcefully to any interference in the process. Iran, for its part, has just made it clear through its news agencies that any foreign military presence in the strait will be considered a direct threatso the United States “plan” threatens not to occur. A military deployment present but contained. There is no doubt, despite the absence of escorts direct, the environment remains heavily militarized. The United States maintains a wide presence with aircraft carriers, aircraft and unmanned platforms ready to act if the situation escalates. At the same time, naval mines have been detected and specific attacks against ships in the area. In other words, in the current scenario, transit has become an operation of calculated risk where coordination replaces direct protection. The real pulse: control, pressure and global economy. Beyond the movement of ships, what is at stake is the control of one of the most important energy arteries of the world. Iran uses the blockade as a pressure tool in the midst of negotiations, while the United States tries to unblock the situation in its own way and with one caveat: without legitimizing that control. The result is an intermediate solution that summarizes well the tension of the moment: intervene just enough so that the boats leave without causing an escalation that blows everything up into the air. Image | Iranian Media In Xataka | After gasoline, the war in Iran is about to skyrocket the price of something just as painful: your Zara clothes In Xataka | Iran has pulled out a “trick” to sell to China while avoiding the US: turning the ocean into its secret gas station

BlackBerry seemed dead and buried. You have no idea how important it still is (but not for your mobile)

by

Let’s take a little trip back in time. It’s 2014 and we’re in Silicon Valley. An executive named John Wall, president of QNXa division of Blackberry, has just arrived in the world’s technology mecca, fresh from Waterloo, in Ontario (Canada). He goes to a meeting with managers from Audi, one of his company’s best clients. The automotive firm has just announced the arrival of Android Auto to its cars. Meanwhile, Apple he doesn’t stop signing QNX engineers to create their own operating system for cars. Blackerry’s shortcomings continue to grow: after losing in mobile phones, it now also had a really bad time in the infotainment systems business. And then, something happened. A beer. The beer of the resurrection. Audi’s head of engineering went to have a beer with him and confessed that Audi was going to use Google’s infotainment systems. However, he told him, his next generation of cars would still need safety features that didn’t exist yet. So Wall came up with an idea. Instead of trying to control the car’s screen, I would try to conquer the software that is the backbone of that entire experience. “The circumstances that caused us to lose infotainment caused the company to pivot in the right direction, whether we knew it or not at the time,” counted Wall. The truth is that they did not have many alternatives, especially after experiencing one of the most famous boom and bust phenomena in the technology industry. From everything to (almost) nothing. In 2008 BlackBerry I was on top of the world. Its market capitalization at that time reached $83 billion, but from then on it plummeted mainly due to the iPhone. Today its capitalization round the 3,000 million dollars, but the surprise is that its great treasure is an almost mythical software that has turned out to be a success in the automotive world. Its name: QNX. How BlackBerry ended up making car software. In 2010 Research In Motion (RIM)—BlackBerry’s former official name— acquired QNX. This real-time operating system that appeared in the mid-1980s and in the 2000s completely changed its focus to target the automotive industry. When RIM bought it, it tried to take advantage of it for its own operating system, blackberry 10but we already know how that ended. QNX’s other big business. The curious thing is that while the company was sinking in the mobile market, QNX engineers who had not moved to the smartphone team continued working on car software. John Wall, president of QNX, has been with the company since graduating in the early ’90s, and in an interview in The Wall Street Journal he recalled how “no one paid attention to us.” That was precisely what changed the course of the company. A crucial operating system. QNX is the operating system that operates collision alerts, blind spot notifications, adaptive cruise control, pedestrian detection or lane correction systems. Not only in cars, be careful: also on motorcycles. It is invisible to the user, who never sees the QNX logo, but certainly sees that everything works. Wall compared his engineers to plumbers and electricians: “What makes QNX virtually irreplaceable is its reputation for never failing.” In Fortune a user commented years earlier how “the only way to make this software fail is to shoot a bullet at the computer running it.” A flourishing business. For years QNX was an overlooked division within an ailing company, but today it accounts for about half of BlackBerry’s total revenue. The software that failed on mobile phones and tablets has ended up being almost the only business that matters for the company, but it has not been limited to cars. QNX is integrated into surgical robots and dozens of medical devices in hospitals around the world. It is also used in industrial plants and automation systems that depend on the security and reliability that QNX provides. It is not without problemsof course, but its software is still key in many critical systems. Prize for being late. BlackBerry was late to the smartphone revolution and lost. It was also late to try to conquer the infotainment segment in the automobile industry and lost. But upon losing that second battle, it adapted and managed to reconvert its operating system into something for which it was precisely designed: a real-time operating system that does not fail and whose latency and response time was (and is) extraordinary. If QNX had continued trying to compete with CarPlay or Android Auto it would probably have disappeared completely, but now it is an absolute benchmark in a niche where its reliability is much more valuable than the flashy new features that infotainment systems usually sell. Today its systems are installed in more than 275 million vehicles. BlackBerry is doing well. BlackBerry shares are up 50% in the latest financial results, and the company has four consecutive quarters in profits. That has caused BlackBerry CEO John Giamatteo to declare that his company “is now a growth story.” These data must be taken with caution, because BlackBerry is very far from where it was almost two decades ago, but the path this company has taken seems the right one. We may not see it compete in the mobility arena anymore, but it has become a fundamental element of an automobile industry that is only doing one thing: growing. In Xataka | There are still those who insist on resurrecting the keyboard and the jack: this is the mobile phone that brings them back with the scent of BlackBerry

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.