Month: November 2025

When you bought a car you were supposed to control it 100%. The industry has managed to make us lose our desire

by

An owner of a Hyundai Ioniq 5 N recently discovered that he couldn’t perform one of the most basic maintenance tasks on his electric car: changing the brake pads. The reason has nothing to do with how complex or not the task was mechanically, but rather with Hyundai’s proprietary software and the professional-level credentials necessary to access it. The episode has reopened the debate on the repairability of electrified vehiclesin an era in which we increasingly need professionals specialized in software and electronics in workshops, and carrying out maintenance on our own is increasingly more complicated. The underlying problem. Brake pads are wear components that any car needs replacing periodically, although in electric vehicles they last longer thanks to the regenerative braking. On most cars, this job can be done at home with more or less basic tools and moderate mechanical experience. However, the Ioniq 5 N incorporates an electronic parking brake that must be fully retracted by a computer to allow changing pads, and then recalibrated to adjust to the thickness of the new parts. What it cost the owner. According to shared user SoultronicPear on Reddit, no conventional diagnostic scanner worked on his 2025 Ioniq 5 N. After trying several options, he purchased a subscription to Hyundai’s J2534 software (costing $60 per week) and an approved adapter (about $2,000). Still, the system didn’t work. After contacting the software developers, he discovered that the Windows version was not updated for the 2025 models, while official dealers use a completely different program based on Android. a barrier. When I finally received the updated version of the software, a new obstacle appeared: the system requested credentials NASTF (National Automotive Service Task Force), a US organization that validates professional mechanics and regulates access to sensitive vehicle functions in the country. According to TheDriveHyundai’s technical documentation states in red that “access to two-way tests and special functions requires NASTF Diagnostic Professional or Vehicle Safety Professional credentials.” Therefore, the owner could not access this adjustment firsthand. Hyundai’s position. The middle consulted to the firm, which defended its procedure arguing reasons of security and functionality. “The official repair procedure requires placing the rear calipers into service mode using our Global Diagnostic System or the J2534 app. This ensures proper functionality and customer safety,” a spokesperson explained. The company he added that it is exploring ways to make routine maintenance easier by “balancing convenience with security,” and that its official tool is available for anyone to purchase, although it is worth mentioning that its price is around $6,000. Beyond legality. Technically, Hyundai does not violate the laws of law to repair because it offers access through systems compatible with the J2534 standard, not only through proprietary equipment. However, what has always been a task accessible to individuals with moderate mechanical knowledge who wanted to do it on their own, has been relegated exclusively to professional workshops, at least in this case. A growing problem. Although the case focuses on Hyundai, the Korean brand is not the only one that makes repairs on modern vehicles difficult. The electrification and digitalization of automobiles is creating new barriers for owners and even independent workshops, who also cannot access these functions. For many enthusiasts, this takes away autonomy over their own vehicles and creates confusion, especially for something that should be as accessible as routine car maintenance. Cover image | Tekton In Xataka | In 2001, Renault launched a car ahead of its time: it was a miserable failure that now has another chance

What is this subscription, what does it include and how much does it cost in Spain?

by

Let’s tell you what it is the price and features of the subscription AI Plus from Google. This is a lower subscription than Google AI Pro, which retains things like access to the most advanced models of Geminibut cut back on the use of other additional tools. This is a movement similar to that of ChatGPT Gowith which Google tries to make the artificial intelligence payment. Because 20 euro subscriptions are excessive for most users, so they have decided to try a much lower one. We will tell you the characteristics and comparison of the other subscription. What is Google AI Plus Google AI Plus is the name of the cheapest subscription to use paid Gemini. Gemini is the company’s artificial intelligence, and its free version is quite limited in terms of access to more advanced and powerful models. The solution if you want greater access to Google’s most powerful AI is to pay, and this is a subscription that seeks to be more affordable. Google initially created two paid subscriptions. For “ordinary” people it offered AI Pro with a price of 22 euros per month, and for more advanced users another price of 275 euros per month. Clearly, even the “cheapest” one was excessive for most users, so they decided to create one more step, a cheaper subscription. By paying this subscription you will have access to Gemini 3 Pro or the new models that arrive in the future. Access will not be any more limited than the most expensive subscription, but wider than in the free version. You’ll also have access to AI interactions with Gmail and other Google services, video creation, and more. Google AI Plus vs AI Pro Google AI Plus Google AI Pro Gemini App Access to Gemini 3 Pro, and Deep Research in 3 Pro. Videos are also generated with limited access to Veo 3.1 Greater access to Gemini 3 Pro, and Deep Research on 3 Pro. Videos are also generated with limited access to Veo 3.1 flow Access to the AI ​​tool for film creation, being able to create scenes and stories, with limited access to Veo 3.1 Greater access to the AI ​​tool for film creation, being able to create scenes and stories, with limited access to Veo 3.1 Whisk More complete access to creating videos from images with Veo 3. Expanded access to creating videos from images with Veo 3. Credits 200 monthly AI credits for Flow and Whisk 1,000 monthly AI credits for Flow and Whisk Notebooklm Access to the research and writing assistant with expanded access to audio summaries, notebooks, and other resources. Greater access to the research and writing assistant with 5x more audio summaries, notebooks, and other resources. Gemini in google apps Access Gemini directly in Gmail, Docs, and Google apps. Access Gemini directly in Gmail, Docs, and Google apps. Also to Vids. Google Home Premium Standard No 30 days of Gemini event and feature history Gemini Code Assist and Gemini CLI No More model requests per day on Flash and Pro models in Gemini CLI and Gemini Code Assist IDE extensions Jules No Higher limits when using Jules, the asynchronous scheduling agent for software developers created by Google Storage 200 GB of total storage in your Google account, to use in Drive, Gmail or Photos 2 TB of total storage in your Google account, to use in Drive, Gmail or Photos Price 7.99 euros per month 21.99 euros per month As you can see, the 8 euro subscription already gives you access to Gemini 3 Pro and Deep Research, as well as video generation with Veo 3.1. The more expensive version has broader limits of use, but the economical one is already an interesting gateway, also for creating videos from images. One of the differences is seen in the number of monthly credits for Flow and Whiskwhich is five times higher than the most expensive subscription. The same thing happens with NotebookLM, which has five times more limits in the Pro version, but in the Plus you already have access. The same thing also happens with the possibility of using Gemini in Google apps, such as Gmail or documents, you have it in both subscriptions. The main difference comes with three additional tools. The most economical subscription does not include the additional benefits of using Gemini on your smart speakersuch as a greater history of events and functions. You also don’t have access to special options for creating code or scheduling agents. For this you will have to go to the Pro subscription at least. There is also a difference in storage that is added to your Google account. In the Pro version it is still 2 TB, but in the Plus you already have 200 GB, which is more than the 15 GB that Google gives you for free. Google AI Plus Price Google AI Plus is priced at 7.99 euros per month. There may be times when there is a temporary promotion, such as two months at 3.99 euros and then return to the normal subscription. But the full price is eight euros per month. In Xataka Basics | The best prompts to save hours of work and do your tasks with ChatGPT, Gemini, Copilot or other artificial intelligence

The fear is that once again we will not be prepared

by

If we were thinking that autumn was being kind to us, the State Meteorological Agency has bad news for this weekendsince this season with days of high temperatures is coming to an end. The culprit? A arctic air mass which will work as a switch for the peninsular climate, since in 48 hours we will go from autumn weather to a mid-winter scenario with usually low snow levels in the north of the peninsula. This is something that reminds us a lot of the last ‘Beast from the East’ that we saw at the beginning of the year in the European center and that affected us with a significant drop in temperatures. But in this case the truth is that the polar cold is going to be very present in our peninsula. The arctic corridor. The meteorological situation is defined by the entry of a very cold air mass coming from very high latitudes. This is something that will begin to be noticed from the afternoon this Wednesday, November 19, since the air mass will begin to be injected through the north of the peninsula, causing a widespread thermal collapse and that arrives just after the passage of the storm Claudia. Different evolution. According to AEMET itself, In a special warning that has been issued, the first ‘affected’ will be those who live in the Cantabrian Sea who will see moderate rainfall this Wednesday and that will become snow from about “900 – 1200 m, exceeding thicknesses of 5 cm in points of the Cantabrian mountain range and the northern face of the Pyrenees.” On Thursday this mass will continue to enter our country, and this will translate into snow levels that will drop to 600 meters generally in the northern third of the peninsula. Although the AEMET itself emphasizes above all the snow that expected in parts of the Basque CountryNavarra or the north of the northern plateau because they can be very copious. The worst day. Between the last hours of Thursday and Friday morning is when the snow level will be between 300 and 400 meters in the eastern Cantabrian and upper Ebro, up to 5 cm of snow may accumulate, which will affect the main transport routes in the region. Although in general, we are going to expect snow in Vitoria, Pamplona, ​​Burgos, León, Soria and potentially in Segovia. The figures of the cold. But in addition to these, up to almost 20 cm of snow in 24 hours in the Cantabrian Mountains, we must highlight the drop in temperatures that has already been experienced since this Wednesday in a large part of the country. What is expected is that maximum temperatures will be 10 degrees below zero in much of the peninsular territory, with the exception of the southwest and the coasts where they will be a little higher. It will be on Saturday when a somewhat warmer air mass enters the territory that will cause temperatures to begin to rise and the snow level will also be limited to the highest mountains. Is this snowfall normal? In the month of November it does not seem normal that we have a heavy snowfall at the doors of the country as if it were December or January, as happened with the Filomena storm in 2021. But if we look further back, we can remember one of the historic snowfalls in November that took place in Madrid between November 27 and 30, 1904considered the heaviest snowfall in more than 150 years, with snow accumulations of between 70 and 150 cm that completely paralyzed the city. A problem on the roads. The problem that this storm just arrives on the weekend is a great inconvenience for roads of our country that are subjected to a greater amount of traffic. The models are quite clear in showing that many of the main roads in the north of the peninsula will be exposed to these adverse weather conditions and that is why extreme precautions must be taken on the days of greatest risk in the country. And in the past we have seen how some cars were trapped in the middle of a highway, such as the AP-6, due to these intense snowfalls. And in the end we don’t know where the most aggressive moment of this storm could surprise us. Are we prepared? In Spain, the truth is that experience tells us that we do not give too much importance to these alerts, as we have seen with the different storms or even the DANA. Faced with this new situation, we must keep in mind that we are faced with a heavy snowfall that affects more than half of the country at really low levels. And although there is logistical preparation in the most affected territories to guarantee mobility and security, past experience, such as historic snowfall from Storm Filomena in 2021shows that the country faces great challenges in managing heavy snowfall, with significant difficulties for transportation and basic services in some cases. Some communities, such as Castilla y León, already prepares with 900 snow plows and salting plants and some 4,400 agents for everything that may be to come. Images | AEMET Marco De Gregorio In Xataka | What is a dry storm: when the sky throws lightning, but the rain never reaches the ground

Ukraine has returned from Europe with 250 fighter jets under its arm. The problem is that only Spain has told him the truth

by

The new European trip of the president of Ukraine, Volodymyr Zelensky, has finished in Spain and has crystallized into a military agenda that aims to reconfigure the Ukrainian air force over the next decade, based on political agreements of enormous symbolic scope. If nothing goes wrong, the Ukrainian nation has nothing less than 250 European fighters under his arm along with a huge aid package and arsenal. The problem is that the financing is very uncertain and its execution is very distant. Aerial reconstruction as a continental ambition. In Paris, the Ukrainian president signed a letter of intent to acquire up to one hundred Rafale fightersdevices that France presents as the heart of the future defense of Ukraine, complemented by Samp/T systemsnew generation drones, guided munitions and incipient industrial cooperation to manufacture interceptors on Ukrainian territory. The French bet aims to elevate Ukraine to European technological standardintegrating it into a long-term security architecture and relying on a financing framework yet to be defined, where the European Union and frozen Russian assets appear as the great promise, although deeply controversial. The political gesture, celebrated as historic in parisresponds to the French ambition to lead the regeneration of Ukrainian air power and to reinforce the role of its defense industry in a continent that is rapidly rearming. Doubts about the bet. Diplomatic enthusiasm contrasts with operational uncertainties. They remembered TWZ analysts either The Wall Street Journal that Ukraine does not have of the financial margin to pay for neither the acquisition nor the maintenance of a hundred Rafale, and France is going through a period of budget fragility which makes sustained long-term commitments difficult. The idea that Europe could finance the purchase through new joint debt mechanisms or from income generated by frozen Russian assets divides the states members and poses enormous legal risks, especially for Belgium, which holds most of those funds. Added to this is the industrial reality: the Dassault production chain is saturatedwith deliveries committed for years, and the manufacturing of 100 additional devices would require extraordinary efforts. The perspective of a parallel program, with 150 Swedish Gripen also agreed in the preliminary phase, increases doubts about whether Ukraine could sustain, train and maintain such a vast fleet of 4/5th generation aircraft. For many, the initiative reflects more a political movement to keep France at the center of the Ukrainian equation and to boost European industry in the face of a United States more distantthan a realistic military acquisition plan in the short or medium term. A Gripen fighter The military horizon. Zelensky’s trip has also highlighted the arrival of a winter that anticipates a new Russian campaign focused in energy infrastructure and strategic cities. France insists that Samp/T systems are demonstrating remarkable effectiveness against Russian missiles with a complex trajectory, even higher, some French commanders claim, than the performance of the Patriot in certain scenarios. In parallel, Paris reinforces its role as a provider of interim air capabilities, including Mirage fighters and precision ammunition, while promoting a future coalition of countries Europeans willing to guarantee the security of Ukraine after an eventual ceasefire, a project still impossible as long as Moscow rejects any negotiation. This strategy, which attempts to combine immediate support with an architecture of long term securityreveals both French determination and the continent’s real limitations in simultaneously sustaining the current war and future rearmament. Among others, Spanish military aid to Ukraine will consist of 40 IRIS-T missiles Spain and the contrast with the promises. The final stop of the trip, in Madrid, has revealed a very marked contrast between the declarative exuberance of some allies and the measured (and often austere) approach of the Spanish Government. Spain announced a package of 817 million of euros, which includes 300 million in nationally produced weapons, 215 million channeled through European programs and additional 100 million to acquire US missiles through PURL initiative of NATO. It is a significant effort in political and logistical terms, but modest in comparison with the great European powers and especially small in the face of the air ambitions presented in France or Sweden. In practice, it is a calibrated support for immediate needs from the Ukrainian winter: anti-aircraft missiles to repel drones and protect critical infrastructures, plus a commitment to accelerate joint industrial capabilities in areas where Spanish companies (with Indra at the head) can offer practical solutions such as deployable radars or anti-drone systems. Spain and realism. If you also want, the Spanish case reflects a much more realistic line than that of other countries visited by Zelensky. Since the beginning of the war, Spain has contributed with useful materialsbut in many cases coming from surplus (Leopard 2A4 retired, M113 obsolete, Hawk batteries aging) and has prioritized its participation in European programs where the direct cost to its budget is lower. In comparative terms, and especially measured as a percentage of GDP, Spain is far behind of the hard core of military support for Ukraine. However, what it offers now is probably more sincere and sustainable: an acceptable package, focused on urgent and realistic needs, that does not promise fighter fleets, perhaps impossible to finance, or industrial projects that exceed national capacity. Spanish extra ball. Furthermore, Spain stands out where other countries they can’t: in the reception of refugees, in the medical rehabilitation of Ukrainian soldiers and in light but reliable industrial cooperation. So, on that journey that began with spectacular advertisements in Paris and Stockholm, the Spanish stop has served to balance in a way the expectations. In that sense, Spain appears as one of the few allies that gauges its support by looking ahead. the budget figuresavoiding promising what it will be difficult to fulfill and remaining firm in what it can offer: a modest but operational contribution. Image | Ronnie MacdonaldTuomo Salonen, Air and Space Army Ministry of Defense Spain In Xataka | Europe already knows the arsenal it needs for rearmament. Now the most difficult thing remains: how to make it arrive in time if Russia attacks … Read more

Xiaomi, Samsung, iPhone phones and more with discounts of up to 50% at the Amazon Second Hand outlet

by

If you are thinking of changing your mobile phone, you should know that Amazon has a section Second Hand in which you can get refurbished smartphones and with a good discount. Below, we offer you a selection of the best models on offer. Xiaomi Poco C85 by 106.59 euros: 6.9 inches and with a 50 MP camera. Motorola Moto G85 by 138.58 euros: 6.67 inches and with Dolby Atmos. CMF Phone 2 Pro by 189.05 euros: with triple camera and 6.77 inches. Apple iPhone 17 by 930.23 euros: 6.3 inches and Front Center Stage camera. Samsung Galaxy S25 by 872.03 euros: 6.2 inches and triple camera. Xiaomi Poco C85 He Little C85 It is a cheap Xiaomi mobile, which stands out for its large capacity battery (6,000 mAh) and that supports fast charging at 33 W. You can get it now, in acceptable condition, for 106.59 euros. This cheap Xiaomi mobile has a 6.9 inch screen with Full HD+ resolution. Its processor is the MediaTek Helio G81 Ultra, with a RAM 6 GB and 128 GB of memory. Its main camera is 50 MP and it is a mobile phone with a fingerprint sensor on the side and facial unlocking with AI. Xiaomi POCO C85 – 6+128GB Smartphone The price could vary. We earn commission from these links Motorola Moto G85 With almost 200 euros off (compared to its usual RRP of 349 euros), you can now get this one on Amazon. Motorola Moto G85 by 138.58 eurosin acceptable condition. One of its most striking features is its impressive infinity edge design. Its screen is 6.67 inches and its speakers offer sound Dolby Atmos. Its camera is signed by Sony and is 50 MP and has optical image stabilization. It has a latest generation Snapdragon processor, 12 GB of RAM (expandable up to 24 GB) and 256 GB of storage. Moto g85 5G 24GB (12GB+12GB RAM Boost)/256GB The price could vary. We earn commission from these links If you are looking for a mobile with a different design from the restbut at an affordable price, this CMF Phone 2 Pro It is one of the ones we recommend in the Second Hand section of Amazon. You can buy it, in very good condition, for 189.05 euros. If there is something that characterizes this mobile phone, it is its space with AI, called Essential Key/Space. In this space you can save screenshots and text and voice notes. Its AMOLED screen is 6.77 inches and has a 50+50+8 MP triple rear camera. The price could vary. We earn commission from these links Apple iPhone 17 If you want a high-end mobile phone, you can save a few euros at the Amazon outlet for second-hand mobile phones by buying this iPhone 17. You can get it, specifically, by 930.23 euros in his version of 256GB and in very good condition. This is the base mobile new generation of iPhone. Its screen is 6.3-inch Liquid Retina. Mount the A19 chip and a Center Stage Front Camera and it is a mobile phone that has greater resistance to scratches. It also features an advanced 48MP Dual Fusion camera system. Apple iPhone 17 256GB The price could vary. We earn commission from these links Samsung Galaxy S25 Another high-end mobile that you can buy with a slight discount at the Amazon Second Hand outlet is this one. Samsung Galaxy S25. It is available, in very good condition and with 512GBby 872.03 euros. This is one of Samsung’s top phones and it is a compact phone, with a 6.2 inch screen. It has a Qualcomm Snapdragon 8 Elite processor and its battery supports fast charging at 25 W. As for its photographic system, it is made up of a 50+12+10 MP triple camera. The price could vary. We earn commission from these links Some of the links in this article are affiliated and may provide a benefit to Xataka. In case of non-availability, offers may vary. Images | Webedia, Xiaomi, Apple, Motorola, CMF by Nothing and Samsung In Xataka | The best mobile phones, we have tested them and here are their analyzes In Xataka | The best quality-price mobiles. Their analyzes and videos are here

The largest nuclear power plant in Europe has been connected to diesel generators for a month. It’s as encouraging as it sounds.

by

Europe is once again walking a nuclear tightrope. After more than three years of war, the largest atomic plant on the continent —the Ukrainian Zaporizhia plant— has gone from being an industrial symbol to becoming at a point of friction capable of triggering an emergency of continental reach. In parallel, other plants in the country operate at reduced power after attacks on the electrical grid. The situation is so unstable that the director of the International Atomic Energy Agency (IAEA), Rafael Grossi, recently traveled to Kaliningrad, Russia, for emergency talks with the head of Rosatom, Alexey Likhachev, according to the Anadolu agency. It is a gesture that reflects the extent to which the risk is real. An attack that left two centers at minimum. According to a statement from the IAEAa military attack during the night of November 7 damaged an electrical substation critical to nuclear security. This incident left the Khmelnitsky and Rivne plants disconnected from one of their two 750 kilovolt lines and forced the electricity operator to order a power reduction in several of its reactors. Ten days later, one of the lines was still out of service and three reactors continued to operate at limited power. The agency emphasizes that these substations are essential nodes of the network: they allow the voltage levels that feed the security and cooling systems to be transformed and maintained. Without them, plants cannot guarantee safe operation. One month depending on diesel generators. The situation in Zaporizhzhia is even more critical. According to an opinion column by Najmedin Meshkati, professor of engineering and international relations published in the Financial Timesthe plant spent a full month without outside power after its two main lines were cut. During that time it survived solely on diesel generators, a resource that the industry considers strictly temporary: they are designed to run for around 24 hours, not for weeks. Technicians were only able to repair the lines under the protection of localized ceasefires negotiated by the IAEA, according to NucNet. Even so, one of the two restored lines was disconnected again on November 14 due to the activation of a protection system. Grossi summed it up like this: “The electrical situation at the plant remains extremely fragile.” The condition for a shut down reactor to remain safe. Although Zaporizhzhia’s six reactors have been on cold shutdown for more than three years, the plant requires a constant three to four megawatts to maintain cooling pumps and other essential systems, according to Meshkati. The professor emphasizes that even huge emergency batteries require external electricity to stay charged. It is a vicious circle: without the electrical grid, batteries are used, but without external electricity, these batteries cannot be recharged and, without both, the cooling systems fail. And without cooling the risk of nuclear fuel melting or overheating increases. The University of Southern California professor warns that this scenario reproduces the conditions that transformed Fukushima into a global disaster: “What turned an earthquake into a catastrophe was the total failure of the electrical system.” And he adds that, unlike 2011 in Japan, this time the risk comes from deliberate human action. A network reduced to its minimum expression. Before the war, according to the Kyiv Independentthe Zaporizhia plant was connected through ten power lines. Today it only has one or two operations and has lost all connection ten times since the beginning of the invasion. The IAEA itself has described the situation power plant as “extremely precarious” and “clearly not sustainable” when it depends for long periods on diesel generators. Short and medium term risks. The notices in the last report on Ukraine by the IAEA point in the same direction: the main danger is not a Chernobyl-type explosion, but a prolonged cooling failure. This scenario could cause overheating of the reactors in cold shutdown, damage to the spent fuel pools and a possible localized or regional radioactive release, with the consequent need to create an exclusion zone in the heart of agricultural Europe. For its part, according to Meshkatiadds two other relevant elements. On the one hand, it points out that a serious accident will exceed the economic impact of Fukushima, estimated at about $500 billion. An incident of that magnitude would affect agriculture, transport, supply chains and the European insurance market. On the other hand, he maintains that if Russia manages to consolidate the precedent that an occupying army can take control of a nuclear power plant and connect it to its own network, the global nuclear security architecture would be seriously compromised. It would be a precedent without equivalent since the creation of international standards that regulate the civil use of atomic energy. Is there a meeting point? The IAEA has acted as an intermediary between Moscow and kyiv on multiple occasions. According to the Anadolu agencyGrossi traveled to Kaliningrad to meet with Likhachev, director of Rosatom, in order to directly discuss the situation in Zaporizhzhia and the minimum conditions to guarantee nuclear safety. At the same time, the agency is trying to technically shore up the Ukrainian electrical system. According to their own statementshas so far coordinated 174 deliveries of essential equipment – ​​switches, electrical cabinets, radiation monitoring stations, vehicles and computer equipment – ​​worth more than 20.5 million euros, intended to sustain nuclear security in Ukraine during the war. Nuclear security supported by fragile cables Europe breathes thanks to a handful of cables repaired under fire and diesel generators that have already proven to be well beyond their limits. As the Financial Times explainsthe continent’s security depends on electricity continuing to arrive and on the parties respecting the fragile ceasefires needed to repair lines when they go down. Grossi summed it up with a mix of relief and alarm after the restoration of one of the lines: “It is a good day for nuclear security, although the situation remains highly precarious.” And the precarious thing, in this case, is that a new attack, a mechanical failure or a downed line is enough to bring … Read more

When they sold us generative “Artificial Intelligence” we did not know that it was going to be artificial and generative but not “intelligent”

by

A few months ago, a group of Spanish researchers thought of putting an AI chatbot to the test with a curious test. They uploaded an image of an analog clock to the chatbot and asked the AI ​​a simple “What time is it on that clock?” The AI ​​failed disturbingly. Machine, can you tell me the time? Researchers from the Polytechnic University of Madrid, the University of Valladolid and the Politecnico de Milano signed a month ago a study in which they wanted to evaluate how intelligent the artificial intelligence of those models was. To do this, they built a large set of synthetic images of analog clocks—available in Hugging Face— in which 43,000 different hours were shown. Before fine-tuning their behavior, the AI ​​models consistently failed when trying to tell the time. After the adjustment the behavior was much better, but still imperfect. That should not happen with such a “simple” issue for humans. disastrous result. From there they asked four generative AI models what time those images of those analog clocks showed. None of them managed to tell the time accurately. That group of models was made up of GPT-4o, Gemma3-12B, LlaMa3.2-11B and QwenVL-2.5-7B, and all of them had serious problems “reading” the time and differentiating, for example, the hands or the angle and direction of those hands in relation to the numbers marked on the watch. Fine tuning to improve. After these first tests, the group of researchers managed to significantly improve the behavior of these models after performing fine tuning: they trained them with 5,000 additional images from that data set and then re-evaluated the behavior of the models. However, the models again failed consistently when tested with a different set of images of analog clocks. The conclusion was clear. They don’t know how to generalize. What they discovered with this test was confirmation of what we have been observing from the beginning with AI models: they are good at recognizing data that they are familiar with (memorized), but they often fail in scenarios that they have never faced and that are not part of their training sets. Or what is the same: they were incapable of generalizing. Dalí enters the scene. To try to find out the causes of these failures, the researchers created new sets of images in which, for example, they used the Dalí’s famous distorted clocksor those that included arrows at the end of the hands. Humans are able to tell time on analog clocks even if they are distorted, but for AI models that was a huge problem. If they do this with watches, imagine with medical analysis. The danger of these conclusions is that they reignite the debate about whether generative AI models are indeed artificial and generative, but not very intelligent. If they have these difficulties in identifying the hands or their orientations, things are dangerous if what the models have to analyze are medical images or, for example, real-time images of an autonomous car driving through a city. AIs are stupid. Although it is true that generative AI models are fantastic as aids in various scenarios such as programming, the reality is that what they do is “regurgitate” responses that are already part of their training data. As Thomas Wolf, Chief Science Officer of Hugging Face, explained, a generative AI “will never ask questions that no one had thought of or that no one had dared to ask.” Although thanks to their enormous memory and training they can recover a multitude of data and present it in useful ways, finding solutions to problems for which they have not been trained is very complicated. For experts like Yann LeCun, the reality is clear: generative AI it’s very stupid and, furthermore, a dead end. Source: clocks.brianmoore.com AI doesn’t draw watches very well either. Added to the experiment of these researchers is another small test that once again calls into question the capacity of generative AI. It involves asking different models to create the code that allows an analog clock to be displayed with the current time. A designer named Brian Moore wanted to share the result of several AI models and the truth is that the result obtained in most of them is terrible, although others like Kimi K2 achieve a good result. We have tested with the recent Grok 4.1 and GPT-5.1. After a little insistence, Grok 4.1 has drawn the perfect clock and it works. With GPT-5.1 there has been no way, at least in our tests. A worrying reality. This inability to solve tasks that seem simple certainly means that these models are not in a good place. It is true that a good prompt can help resolve some of these limitations, but what is becoming increasingly evident is that AI models continue to make mistakes despite the passage of time. The theoretical revolution of this technology precisely needs to eradicate them, and it does not seem that we are on the way to achieving it. The models improve, yes, but not enough for us to trust them 100%. Image | Yaniv Knobel In Xataka | As if there weren’t enough AI companies, Jeff Bezos has just returned from the shadows to build another one, according to the NYT

The emir of Qatar travels in a private jet so big it helped Sardinia airport upgrade

by

In 2021, the airport Olbia Costa Smeralda In Sardinia, it undertook work to expand its runway to be able to receive long-distance flights, thus opening the door for international airlines to bring a greater volume of tourists to the island. However, the inauguration of this work was somewhat special. As and how did he count Luxury Launchesthe ceremony inauguration of the new track It starred the Emir of Qatar, Tamim bin Hamad Al Thani, but he did not do so by unveiling any commemorative plaque or cutting any ribbon. He did it bravely: landing his huge private jeteither. Who said fear? A private jet so big that it changes the category of the airport In the summer of 2021, the works on the Sardinian airport had just been completed. In an attempt to escape the scorching heat of Doha, the emir wanted to spend a few days of relax in the Mediterranean. Neither quick nor lazy, the president gave the order to embark to his crowded entourage who usually accompanies him on his private plane, and they headed to Sardinia. The Boeing 747-8, in addition to being one of the largest airplanes in the worldis the plane that Qatar Amiri Flight, the airline owned by the Qatari emirate, has assigned as a private plane for the top leader of the country. The emir’s plane, valued at around 370 million euros, has impressive dimensions, being 76 meters long, more than 68 meters wide and weighing close to 450 tons at takeoff. Qatar Boeing 747-8 Amiri Flight. The “private jet” of the emir of Qatar Olbia airport was already a key point due to its capacity to move almost 1.8 million passengers in 2008, operating mainly with domestic flights and some destinations in Europe. The infrastructure had just been expanded, lengthening the main runway by about 300 meters to a length of 2,740 meters, the safety zones were expanded and the taxiway was improved, which speeds up the approach to and departure from the runway. In principle, there would be no problem for the huge private jet to land. There was only one small detail: the track had not been tested previously and, in fact, It wasn’t even approved so that planes the size of the emir’s 747-8 could land there. Unimportant details. Olbia Costa Smeralda airport in Sardinia after its expansion As the airlines had not yet scheduled any long-haul international routes from that airport, the airport authorities took advantage of the visit of your important tourist to officially certify the ability to operate this type of flights that use aircraft such as the Boeing 747, Boeing 777, the Airbus A330, the Airbus A340. If the emir could land with your private jet loaded with his entourage, international tourists could too. The operation was carried out without incident, confirming that both the length and the paving of the runway were adequate to support the operations of these air giants. According what was published through the local environment The New Sardegnathanks to the inaugural maneuver of the private jet of the Emir of Qatar, in November of that same year the first flights connecting Sardinia with Los Angeles, China and Singapore with direct flights were inaugurated. The emir of Qatar: main interested party Even if all precautions had been taken during the landing operation, being the first aircraft of its kind to use the runway always entails some risks. However, the emir of Qatar was especially interested in international planes being able to land on that runway. full of tourists. The reason is easy to guess. The most prestigious hotels, marinas and resorts on the Emerald Coast belong to Emerald Holdingwholly controlled by the Qatar Investment Authority. Hotel Cala Di Volpe in Sardinia. One of the five-star hotels of the Emir of Qatar We are talking about a series of five-star hotels that offer luxury stays on the shores of the Mediterranean for clients as select as the Emir of Qatar. Therefore, it is not strange that the highest representative of this hospitality empire opens the way for millionaires from all over the world to use the new runway to land with their private jets or arrive accommodated in the seats business of international airlines. In Xataka | A single millionaire spent the equivalent of 10,000 tourists on his luxury vacation in Mallorca: the Emir of Qatar Image | Wikimedia Commons (Khamenei.ir, Mehmet Mustafa Celik, John Murphy), Marriott

SpaceX changed the space economy. Now he wants to do the same with the cost of satellites

by

The cost of launching cargo into space was, for years, one of the great limits of the aerospace industry. LaNASA documents in several works, including the analyzes of Harry W. Jonesthat during the last decades of the 20th century many pitchers moved in a typical range of between 10,000 and more than 20,000 dollars per kilowith an average cost of around $18,500/kg in low orbit, with the space shuttle far above due to its complexity and operating expense. It was not just the price of the launch systems, but of a model based on disposable components, manual processes and highly specialized operations. The situation remained stable for decades, until SpaceX decided to rethink how the economics of orbital launch should work. Instead of assuming these costs as inevitable, the company opted to reuse stages, optimize processes and manufacture its own engines and systems from scratch. This combination allowed the price per kilo to be reduced to unprecedented levels, although the change did not occur immediately. What is relevant is that, for the first time, a private actor demonstrated that launches could be much cheaper and that price did not have to be a structural barrier for the industry. When launch is no longer the limit, attention shifts to satellites The resulting prices began to change behavior in the sector. With Falcon 9 and Falcon Heavy, the cost per kilo became in the range of 3,000 to 1,500 dollars, according to NASA calculations based on catalog prices. These figures not only mark a reduction, but a turning point: for the first time, companies, institutions and even governments could rethink the design of missions knowing that launch was no longer the main economic barrier. From there a question arose that until then had no answer: if the trip had been made cheaper, what would happen to what was sent into space? The traditional satellite model was built on the idea of ​​optimizing each unit. It was not important to produce many, but to produce one that could operate for years, with high capacity and low probability of failure. Manufacturers and operators were investing in complex systems, with long development cycles, exhaustive testing and specialized structures to fulfill specific and prolonged missions. This strategy responded to an environment in which launch was so costly and infrequent that it was more profitable to prioritize reliability and durability than to think about scalability or rapid replenishment. One of the first companies to help change this approach was OneWeb, that introduced a manufacturing model designed for scale. Instead of ordering each satellite as an individual piece, the company designed a common architecture and partnered with Airbus to produce repeatable unitswith standardized processes and shorter manufacturing times. The plant installed in Florida in 2019 was presented as the first factory of satellite serial production on a large scale, with two lines capable of removing up to two units a day. It was not about building a better satellite, but about building many. SpaceX took the satellite constellation idea and turned it into its own industrial system. With Starlink, it not only replicated the use of mass-produced satellites, but also linked that production to its launch capacity with Falcon 9, operated by the company itself. This integration allowed the deployment to be accelerated without depending on external release windows or commercial suppliers. The constellation began to grow at an unprecedented rate and, in a few years, it vastly surpassed any other similar project in number and pace. The difference was not only in manufacturing satellites, but in being able to launch them at will. Although OneWeb was one of the first players to apply industrial logic to satellite manufacturing, its constellation has grown at a very different pace than Starlink. At the end of 2025, OneWeb has around 648 satellites in orbit, while SpaceX exceeds 8,000 operational satellitesaccording to the most recent data published by orbital monitoring firms. The difference is not only due to the number of launches, but also to the mode of production. According to an economic analysis published in 2025the estimated manufacturing cost of OneWeb satellites is around $14,000 per kilo, compared to approximately $2,500 per kilo for Starlink satellites. These figures reflect a gap that has more to do with the integration model than with the technology itself. The estimated manufacturing cost of OneWeb satellites is around $14,000 per kilo, compared to approximately $2,500 per kilo for Starlink satellites. The reaction of the sector did not take long to arrive. With the advancement of Starlink, both companies and public institutions Similar projects began to be considered based on constellations with a high number of satellites and sustained deployments. Amazon launched KuiperEutelsat and OneWeb reinforced their alliance to maintain presence in the market and the European Union approved the IRIS2 program with institutional support.China is also working on its own large systems. It is not just about competing in numbers, but about accepting that scale and replacement capacity are part of the new spatial model. When the satellite becomes a replicable product, the way of planning its presence in orbit also changes. It is no longer about launching a mission and hoping it works for as long as possible, but rather about building a structure that can grow, modernize and replace units regularly. The satellite becomes a component of a network, not the center of the mission. This logic favors models based on scalability and continuous replacement, similar to those of other technological infrastructures. Space stops being a destination and becomes a platform. SpaceX demonstrated that the cost of the launch was not a technical limit, but rather a model one. Now it is trying to apply that same logic to satellites, with an approach based on scale, continuous manufacturing and integration with its own launch systems. The result is not only a larger constellation, but a different way of understanding what it means. operate in orbit. The question is no longer how much it costs to get to space, but who can … Read more

In a gesture of incalculable Frenchness, France has named the first rocket launched from its borders “Baguette One”

by

The Spanish Miura 1 rocket took off from southern Spain. He french rocket Baguette One will do the same next year from the south of France. It’s not a joke. It is the real name of the next bet of the European New Space. And it is very serious: the French company HyPrSpace has just closed an agreement to launch an experiment on board, confirming that the launch will take place from mainland France: something unprecedented in the civil sector. Traditionally, France launches its missions from the Kourou Spaceport in French Guiana. However, the Baguette One will take off from Europe. The suborbital rocket, about 10 meters high (slightly lower than the Miura 1), will take off from the Biscarrosse missile testing center, in the Landes department, thanks to an agreement with the French Directorate General of Armaments. You already have a client. The little rocket will not go empty. HyPrSpace has signed a memorandum of understanding with ATMOS Space Cargo to launch a demonstration mission. The German space logistics company will take advantage of the suborbital flight to test its Phoenix-2 reentry capsule. The French startup HyPrSpace, based in Bordeaux, is developing Baguette One as a preliminary step to validate the technologies of its future commercial rocket Orbital Baguette One. The project has just closed a financing round of 21 million euros from private funds. They are added to the 35 million that HyPrSpace had secured from the France 2030 public plan. Orbital Baguette One. The OB-1 will follow the Baguette One with a first launch scheduled for the end of 2027. This microlauncher promises to put between 200 and 250 kg into orbit with low prices as its main attraction. Instead of using pure liquid or solid fuel engines, HyPrSpace (short for Hybrid Propulsion for Space) will use a mixture: solid fuel made from recycled plastic and liquid oxygen as an oxidizer. The advantage of this architecture is that it eliminates turbopumps, one of the most expensive and complex pieces of aerospace engineering, which reduces the cost of the launcher by 40%. The disadvantage is that they are less versatile engines and without the possibility of reuse, something that PLD Space does plan for future versions of the Miura 5. Image | HyPrSpace In Xataka | The only photo you need to understand the scale of what Blue Origin, Jeff Bezos’ company, has just done

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.