Mobile Tech & Gadgets

There are people using AI to plan murders. The question is what AI companies are doing about it

by

On February 10, an 18-year-old girl shot and killed her mother and brother. Then he went to the institute and murdered seven more people, finally committing suicide. The disturbing thing is that the author had talked about it with ChatGPT and OpenAI had the opportunity to notify the police, but chose not to. What has happened? They count in the Wall Street Journal that, in June of last year, OpenAI’s automated system detected several messages that a user had sent to ChatGPT describing scenarios of armed violence. For some employees they were very worrying because they could end in real violence, so there was an internal debate about whether to notify the Canadian authorities. They finally closed his account, but they didn’t notify anyone. Now Canadian authorities have summoned them to ask for explanations. There is more. He Tumbler Ridge shooting It is not the only case in which AI has been used to plan a crime. At the beginning of 2025, a man parked a Cybertruck full of explosives in front of a hotel in Las Vegas with the intention of detonating it (although in the end the only victim was himself). Days before, the author I had asked ChatGPT how to do it. In this case, the chatbot did not detect any concerning messages, but we know this because OpenAI searched through its messages after the fact. In Seoul, a woman was jailed for the alleged murder of two people due to benzodiazepine poisoning. The investigation revealed that the accused had gone to ChatGPT to find out what the dangerous dose was and what happened if it was mixed with alcohol. The messages in this case are not that alarming and could arise out of genuine doubt, but it is another example of ChatGPT being used in the commission of a crime. Why is it important. Artificial intelligences have become a kind of confessional to which we tell all kinds of secrets, even the darkest. There are those who consider that AI is a friendhis psychologist or even his lover. In this sense, it is not strange for someone to tell ChatGPT that they are going to kill their family or want to detonate a car full of explosives. What is worrying, and where we should focus, is what companies are doing about it. At the moment, it seems not enough. Are they obligated? Confessing to your psychologist or psychiatrist that you want to hurt someone is one of the reasons why you not only can, but should break your relationship. professional secret and alert the authorities. However, no matter how much we use chatbots as psychologists, at the moment there is no law that forces AI companies to report these types of interactions, but it is an internal decision. The obligation, therefore, is not legal, but ethical. How to make a homemade bomb. Cases like that of the Tumbler Ridge shooter are not something that has begun to happen with the arrival of AI chatbots. Instructions for creating homemade bombs have been around for decades. bringing the authorities to their heads, Even before the use of the Internet became popular, manuals of this type were circulating. The same thing happens with the suicide cases; You don’t need to ask ChatGPT, we can Google it or write in a forum. In statements to New York Timesa former OpenAI employee highlights an important nuance: with a chatbot you don’t usually do a simple search, but rather you can have a longer conversation where the intentions are clearer. In this sense, it may be easier to detect cases like the Tumbler Ridge shooter, but there may also be many false positives due to users who are writing fictional stories or using AI as role-playing. Complicated. In Xataka | Investing in data centers for AI is insane, and it’s going to get worse. much worse Cover image | Pexels, Unsplash

Catching an offender on a scooter on foot is impossible. So the Valencia police are going to chase them on scooters

by

Yes, electric scooter users have to respect traffic rules. In fact, in recent years specific regulations have been created for them. But there is a problem: “catching” an offender on a scooter is almost impossible. And that is why in Valencia they have gotten to work creating a new unit. One who rides a scooter. on scooter. “We are convinced that this unit is going to be a success and will be a benchmark. There are already town councils from all over Spain that have asked us and want to know how it works,” The words are from María José Catalámayor of Valencia, at the presentation of the city’s new municipal police unit. The objective will be to ensure that users of scooters and other personal mobility vehicles circulate in compliance with traffic regulations on the city’s bike lanes. Their powers range from fining those who circulate incorrectly to those who do so with tricked out scooters or under the influence of alcohol. The scooter in Valencia. The information presented by the Valencia City Council specifies that the city has 200 kilometers of bicycle lanes and that between 2019 and 2023 the use of this means of transport skyrocketed, growing by 186% in those years. Creating a specific group with 12 officers to control traffic while patrolling on scooters is the latest decision by a city that is trying to bring non-compliant users into line. In 2024 they already presented their own machine to control which scooters complied or did not comply with the regulations in a kind of mobile MOT. According to data from the City Council collected by elDiario.esaccidents involving scooters have skyrocketed. In 2019, 346 incidents were recorded where they were present but in 2025 they will already reach 1,192. That is, at least three incidents daily throughout the year. Escapism. The press release that the Valencia City Council has published to confirm this information makes it clear what one of the problems they are facing in the city is: The use of the VMP allows a patrol integrated into the urban mobility network, with greater capacity for direct observation and precise detection of infractions such as inappropriate speed, improper circulation or dangerous maneuvers. Its small size and great maneuverability facilitate rapid approach to conflict points and inspection of the state of the lane, signage and safety elements. In interventions with offending users, VMPs offer the necessary agility to safely reach and detain those who try to evade police action. This unit efficiently and sustainably reinforces the police presence in spaces where traditional citizen security vehicles show operational limitations. And the agility and speed with which an electric scooter moves makes it very difficult for an agent to stop it unless, at that very moment, it is riding a bicycle. Even by car, a patrol can have problems if, for example, an offender steps onto a sidewalk or travels on a segregated bike lane. good money. It must be taken into account that, although you do not need a driving license of any kind as is the case with a bicycle, using an electric scooter requires compliance with basic traffic rules. For example: The user must wear a helmet Only one person can circulate per scooter Driving on sidewalks and pedestrian crossings is prohibited. Driving under the influence of alcohol and/or drugs is prohibited Mobile phone use prohibited It is prohibited to wear headphones It must be taken into account that some of the above infractions are classified as serious or very serious within the Traffic Law and, therefore, a user who drives under the influence of alcohol cannot have points removed from his or her driving license but can be fined 500 euros. More watched. In recent years we have seen how electric scooter users are beginning to be more vigilant. It must be taken into account that we are talking about a device that, almost by default, can circulate at a maximum speed of 25 km/h, which is obliged to move on asphalt and bike lanes but which on many occasions we find them on the sidewalk. This has led the DGT to create a record of all scooters soldwith a type of license plate that must be present on the chassis of the vehicle to demonstrate that it complies with the legal technical characteristics. In addition, users will have to have civil liability insurance for their use. Photo | Valencia City Council In Xataka | Arrested for driving an electric scooter capable of going 111 km/h: more than four times the permitted speed

AI solves equations and chops code, but continues to crash with PDFs: the explanation shows its limits

by

It’s probably happened to you. You upload a PDF to an artificial intelligence chatbot in the hope that it will summarize a report, extract a table or find a specific piece of information for you in a matter of seconds. And, sometimes, he succeeds. But other times, the result is disconcerting: mixed columns, footnotes embedded in the middle of the text, tables converted into an illegible block or answers that do not faithfully reflect what the document says. The paradox is evident. Systems that already demonstrate clear advances in mathematics and programming They keep stumbling upon something as everyday as a PDF. And there is more than a simple punctual failure. Change of mentality. Although for us it is a document with well-defined paragraphs, titles and tables, for the system that processes it the situation may be very different. PDF is, first and foremost, a way to visually describe how a page should be rendered. And when a chatbot like Gemini either ChatGPT If you try to work with it, you do not always access an ordered structure, but rather a set of graphical instructions that you must first reconstruct before you can respond coherently. And that difference is better understood when we look at how a PDF “saves” information. How you actually organize information. Unlike a web page, where the content follows a logical order defined in the code, a PDF can store text as independent fragments placed at specific positions on the page. Many times, the file retains coordinates and placement instructions, but not necessarily explicit relationships between one sentence and the next. This implies that the order in which the text “appears” when extracted does not always coincide with the order in which we read it. If your document includes multiple columns, tables, or overlapping elements, the system must figure out how they fit together. And that deduction is not always trivial. {“videoId”:”x9hhg44″,”autoplay”:false,”title”:”The TRUTH of AI – This is how ChatGPT 4, DALL-E or MIDJOURNEY works 🤖 🧠 ARTIFICIAL INTELLIGENCE”, “tag”:”webedia-prod”, “duration”:”1173″} What happens with HTML. On a web page, the content is organized in an explicit hierarchy– There are tags that indicate what a title is, what a paragraph is, what a table is, and how those elements relate to each other. This structure is part of the file itself and makes it easier for other systems to read, index and process it. In a PDF, as we have seen, that semantic layer may not exist or be clearly defined. Therefore, in practice, extracting information from a website tends to be a more predictable process, while doing it from a PDF is more complicated. So what about OCR? It is the first solution that comes to mind. If the problem is that the text is not well structured or even “drawn” like an image, optical character recognition should convert it into something machine readable. And in part it does. OCR has been used for decades to transform images of words into text, but converting an image to text is not the same as reconstructing the logic of the document. When there are varied elements, the system can recognize each word without knowing exactly how they fit together. The result is not a failure in reading characters, but in the organization of information. In Xataka Dario Amodei founded Anthropic because OpenAI didn’t take the risks of AI seriously. Now you are going to give in to those risks Why don’t we abandon PDF? The answer is more pragmatic than technological. As reported by The Verge citing the person responsible for the PDF Associationthe format became established precisely because it allows a document to look the same today as it would in ten or twenty years, regardless of the device or software with which it is opened. A web page can change depending on the browser, an editable sheet can be modified or overwritten, but a PDF maintains its appearance and visual integrity. That stability is precisely what lawyers, engineers, public administrations and any organization that must maintain reliable records need. The challenge is not to replace the format, but to learn to interpret it better. Images | Xataka with Nano Bana In Xataka | Three AIs clashed in ‘War Games’. 95% of them resorted to nuclear weapons and none ever surrendered (function() { window._JS_MODULES = window._JS_MODULES || {}; var headElement = document.getElementsByTagName(‘head’)(0); if (_JS_MODULES.instagram) { var instagramScript = document.createElement(‘script’); instagramScript.src=”https://platform.instagram.com/en_US/embeds.js”; instagramScript.async = true; instagramScript.defer = true; headElement.appendChild(instagramScript); – The news AI solves equations and chops code, but continues to crash with PDFs: the explanation shows its limits was originally published in Xataka by Javier Marquez .

Employment among those over 65 triples and reaches the maximum in the historical series. There is a good reason: retirement

by

The labor market in Spain has recorded several notable milestones in recent months: record contributions, lowest unemployment rate in decades and recovery of youth employment. However, the last annualized EPA data They hide a story that goes beyond global figures. According to Annual average data for 2025 published this week by the INE, the employment rate among those over 65 has reached its historical maximum, and the reason is not that older Spaniards have discovered a sudden love for work. There is something structural behind it that deserves a closer look. An aging workforce. The aging of the population in Spain, and the changes in the pension system that were approved in the 2011 reform, are quietly but very significantly redrawing the Spanish labor map. What a decade ago seemed like a statistical anomaly has today become a consolidated trend with direct consequences on the future and viability of public pensions. The EPA data of the fourth quarter of 2025 indicate that at the end of the year there were 4,926,300 employed people over 55 years of age in Spain. This represents a growth of 23.3% in this age range since the 2022 labor reform, compared to the 11.3% average increase recorded by the rest of the ages. But the most striking thing is that the employment rate among those over 65 years of age has tripled compared to the levels of a decade ago, with 14.25% for men between 65 and 69 years old and 12.29% for women in the same age group, compared to the 5% that was registered in 2015. The employment rate for men between 60 and 64 years old is around 58% in 2025. highest since the early 1980s. All this used to be retirement. What largely explains this rebound in employment among the population over 65 years of age is not a greater demand for experienced workers, but rather the progressive delay in the legal retirement age. In 2026, the legal age for access ordinary retirement For those who have less than 38 years and 3 months of contributions it is 66 years and 10 months. This displacement forces many people to remain active beyond the age of 65 at which they could previously retire. Howeverthe report ‘Quarterly Labor Market Observatory‘ prepared by Fedea and BBVA Research confirms that the increase in senior membership is mainly due to the aging of the population and the delay of retirement agewhich often responds to the need to continue working due to financial difficulties. Staying in the job market at that age is not easy. However, although the data points to record percentages compared to historical figures, the reality is that their employment situation is not a bed of roses. a study of the BBVA and Ivie Foundation has revealed that those over 55 years of age register for the first time an unemployment rate of 9.8%, exceeding the unemployment rate of the group of people between 25 and 54 years of age. Furthermore, six out of ten unemployed of that age group They are long-term unemployed, a percentage that triples that of young people between 16 and 24 years old. The data depict a labor market in which workers over 55 years of age they lose their jobs a decade before their retirement age, and must survive throughout that time either with temporary employment, or in a situation of chronic unemployment due to lack of opportunities. At the other extreme, the employment rate of 14.2% shows those who have managed to stay afloat or get out of that hole. The pension system, the backdrop. Behind all these figures there is a reality that economists have been pointing out for years: the pension system needs people to work longer. to be sustainable. The reforms have been moving incentives in that direction, tightening the requirements for early retirement with greater pension reduction coefficientsand with a progressive increase in the necessary years of contributions. The result is what the data is already showing: there are more and more people who cannot retire at age 65 and must extend their working life until age 67 (effective in 2027) to access their retirement pension. In Xataka | What is the regulatory base: how it is calculated in 2026 with examples Image | Unsplash (Matt Bennett)

free geothermal and waste-based heating

by

The race to dominate artificial intelligence (AI) is no longer waged only in the aseptic laboratories of Silicon Valley or in microchip factories; is moving towards a much more earthly and critical terrain: electricity. At a time when data centers threaten to saturate the global electrical grid due to their voracious consumption, big technology companies are desperately seeking sources of continuous, stable and emission-free energy. The answer, surprisingly, does not seem to lie in looking to the sky for sun or wind, but in drilling down, miles underground. Geothermal energy has ceased to be a secondary actor and has become the great hope of the sector. But in Europe, this technological revolution is accompanied by a master shift. It’s been under our feet. Historically, geothermal energy generation was considered viable almost exclusively in exceptional volcanic regions, such as Iceland or Indonesia. It depended on finding underground pockets that naturally had heat, water, and permeable rock. However, as the report explains Hot stuff: geothermal energy in Europe of the Ember think tankthe technological advances of the last decade have completely rewritten this map. The industry has adapted deep drilling and reservoir engineering techniques from the oil and gas sector, reducing well costs by approximately 40%. Now, so-called Enhanced Geothermal Systems (EGS) allow fluids to be injected to create artificial fissures in hot, dry rock, extract that heat and generate electricity at the surface, regardless of the natural permeability of the ground. Numbers that change the energy board. The impact of this technological disruption is monumental. As detailed by analyst Pawel Czyzak in his newslettergeothermal energy can now be produced at levelized costs (LCoE) of less than €100/MWh. To put it in perspective, the marginal cost of electricity generated by gas and coal in Europe ranged between €90 and €150/MWh during 2025. Geothermal is already economically competitive. In the European Union, this technology could develop around 43 GW of commercially viable capacity today. With geothermal plants operating 24/7, this would translate into around 301 TWh of electricity per year, the equivalent of replacing 42% of all EU coal and gas power generation last year. The countries with the greatest potential identified under this profitability threshold are Hungary (with 28 GW), Poland, Germany and France. The “Triple Victory” strategy. Europe’s great asset lies in geography and urban planning. According to Czyzak,the areas with the greatest geothermal potential at 5,000 meters depth coincide strikingly with large European data center nodes – such as Paris, Amsterdam and Frankfurt – and with planned district heating networks (known as district heating). The plan is to locate data centers near these geothermal plants. The plant powers the AI ​​and, subsequently, the waste heat generated by both the plant and the servers themselves is injected into the district heating networks. Institutions are already making moves. By the end of 2024, the Council and the European Parliament supported the creation of a European Geothermal Alliance to expedite permits and finance the sector. In this scenario, Spain claims a leading role: Vice President Teresa Ribera (whose position is now held by Sara Aagese) announced an injection of 100 million euros for ten deep geothermal projects. The majority will be located in the Canary Islands due to their exceptional volcanic subsoilalthough the peninsula already has pioneering projects underway, such as the 150-meter wells on the Vitoria university campus or the 6.5 MW installation in the City of Arts and Sciences in Valencia. The Nordic laboratory. To understand how the final part of this plan—heating homes with data—works. you have to look at Helsinki. The Finnish capital has found an unexpected ally in the residual heat of servers to decarbonize its winters. Through the energy company Helen, the city has been testing this model for years. The results show that a single data center in Helsinki can heat up to 20,000 homes. The Telia installation, for example, already recovers 90% of the heat emitted by its machines, currently providing shelter to 14,000 apartments. This thermal miracle requires two elements: an extensive network of urban pipes (district heating) and huge industrial heat pumps that raise the temperature of the waste water to the 85-90 ºC necessary for the urban network. Europe, and especially the Nordic countries, are leading the adoption of these heat pumps, turning Finland into a full-scale laboratory for what the future of the continent could be. The risk of missing the technological train. Despite the promising outlook, Europe faces serious obstacles. As the Ember report warnsthe Old Continent invented geothermal electricity (the first plant was inaugurated in Larderello, Italy, in 1904), but now it risks giving up its leadership. As the United States and Canada scale commercially thanks to aggressive tax incentives (such as Inflation Reduction Act) and the private investment of the Big TechEurope is drowning in a morass of slow and complex permitting, inconsistent national support frameworks and a lack of financial risk mitigation for early phases of drilling. Up to 64%. If the EU does not channel innovation funds and simplify bureaucracy, supply chain and cost reduction will consolidate outside its borders. In fact, US research cited by Ember indicates that geothermal could cost-effectively cover up to 64% of the projected increase in electricity demand from US data centers by the early 2030s. The reward for doing things well is economic prosperity. As Czyzak recalls based on his experienceIceland in 1940 was 70% dependent on coal and was one of the poorest economies in the West; Today, thanks to a 100% clean electrical grid (30% geothermal, 70% hydroelectric), it attracted the aluminum industry and became the fifth country in the world in GDP per capita. Deep geothermal could be that same catalyst for countries like Hungary or Slovakia in the era of artificial intelligence. The earthly paradox of the cloud. In their eagerness not to stop the progress of their algorithms, giants like Google or Meta have understood that the solution is not just to look at the sky waiting for the sun to shine or the … Read more

assault in sound quality and for being the most comfortable headphones

by

Now that Samsung has finally announced the new generation of its headphones, what better way to explore what they offer than to compare them with models from other brands. Are they better than Apple’s? To do this, in this article we are going to put the Samsung Galaxy Buds4 with AirPods Pro 3. The differences between the Samsung Galaxy Buds4 and the AirPods Pro 3 The design, which one do you prefer? It is evident that one of the most important points when using headphones is the design. The Samsung Galaxy Buds4 they have a open, padless design while the AirPods Pro 3 They come with a set of five pads. And this may interest you a lot when buying headphones or others. Each person is different and headphones may be good for one person, but not for you. The good thing about AirPods is that with the pads you can customize the headphones so they fit better in your ears. Samsung does not allow this as it has an open design and no pads, so in the worst case you could feel some discomfort if they are too big for you or they simply fall off when walking, running or exercising. Best for lossless audio The AirPods Pro 3 may be the perfect headphones for an Apple user, but not everything is pretty: the headphones support the AAC codec (256 kbps), so if you use platforms that offer ‘Lossless’ sound quality you will not be able to have the best possible experience. Things change with the Samsung Galaxy Buds4 that support the codec SSC-UHQ (24-bit / 96 kHz). In this case you can take advantage of the ‘Lossless’ quality offered by many platforms such as Tidal or Spotify. If you are looking for the best possible sound quality, these are the ones that may interest you the most. Samsung wins in autonomy, but with nuances Both the Samsung Galaxy Buds4 and the AirPods Pro 3 offer good autonomy, up to 30 hours for the former and up to 24 hours for the latter. But if we remove the charging case from each one, things change. If you have active noise cancellation on, the Samsung headphones will last you just under five hours, while with the Apple headphones almost you will be able to last almost an entire day of eight hours of work. This is not to say that Samsung headphones are worse. Simply you have to take into account when you are going to use them: If it is for training or listening to music for a few hours, Samsung headphones may suit you well. If you listen to music for many hours to work, Apple headphones win here. What happens if my headphones get wet? Although it is always better to avoid water to ensure proper functionality of the headphones, both the Galaxy Buds4 and the AirPods Pro 3 are water resistant, although not in the same way. Resistances are usually measured by the degree of IP certificationand in this case the headphones come with: IP54 certification on the Samsung Galaxy Buds4: protection against water jets. IP57 certification on AirPods Pro 3: protection against complete immersion in water at a depth of 1 meter for 30 minutes. In any case, both headphones are durable enough to function properly when used while exercising or in the rain. In the event that, for example, they fall into a puddle of water or into the bathtub, the AirPods Pro 3 are protected for these situations. A little extra that can change your choice Although it is something that we can find in other devices such as smartwatches, it is worth mentioning that the AirPods Pro 3 have a sensor that we do not see in Samsung headphones: the heart rate sensor. It is not a simple addition that we are going to avoid because we already have a watch, since there is a good reason why Apple has decided to introduce it in headphones. Beyond the reliability provided by sensors in devices such as smartwatches, it is medically more reliable for a device to take vital signs from the ear than from the wrist. If you exercise and are looking for greater control over the measurements, this AirPods sensor may interest you. In summary: which Bluetooth headphones to choose according to your tastes and needs Why choose the Samsung Galaxy Buds4 The Samsung Galaxy Buds4 arrive in style, offering a very competitive price. Where they stand out the most is in: Your SSC-UHQ codec. If you are going to use them to listen to music on platforms that offer ‘Lossless’ quality, these are the ones we recommend. If you listen to music sporadically. The headphones alone have less autonomy than Apple’s, but with the charging case they get a higher figure. If you use them for sessions of no more than four and a half hours, these are the ones that may interest you. If you want the cheapest. It is inevitable to compare prices on these headphones, so if you want to spend as little as possible, the difference of 70 euros is quite attractive. The price could vary. We earn commission from these links Why choose AirPods Pro 3 The AirPods Pro 3 were launched a few months ago and continue to be one of the best purchase options in their price range. Where they stand out the most is in: The design. That it includes five pads is something very interesting if you don’t know very well how they are going to fit in your ears. With them, you can adapt the headphones so that they do not move, something especially interesting when walking, running or exercising. Your autonomy. The autonomy of the headphones (without charging case) reaches eight hours, so if you listen to music while you work, these are the ones that may interest you the most. Better water resistance. It’s not every day that we drop our headphones into a puddle or pool, … Read more

It has Taiwan in front of it and Japan is going to fill it with missiles

by

At the westernmost tip of Japan there is a paradise place where, on clear days, you can see another territory from the coast. It is the same enclave where they live more native horses than school-age children. That isolated corner, for decades outside the big headlines, has begun to occupy an unexpected space in the strategic conversations of the Indo-Pacific. Also to become in a fort. A red line. That island has become the new red line against China. The reason? Japan will deploy missiles 100 km from Taiwan. In this way, Yonaguni, the westernmost point of the Japanese archipelago, has gone from being a remote enclave in just a few years. a centerpiece of the Indo-Pacific strategic board. Its location, at the end of the Nansei island chainplaces it right in the geographic arc that connects the East China Sea with the Western Pacific, the same corridor that worries Tokyo and Washington facing a possible conflict in the Taiwan Strait. The calendar changes. A few hours ago, Defense Minister Shinjiro Koizumi set for the first time a very specific horizon: before March 2031, a set of surface-to-air missile medium range, projectiles with 360 degree coverage capacity and the possibility of intercepting multiple targets simultaneously. The decision is not isolated, but is part of the strategic turn started in 2022 to reinforce defenses on the southwestern islands, shifting the historical focus from Russia to growing Chinese military activity in the East China Sea. The diplomatic context and Chinese pressure. The announcement also comes after months of deterioration between Tokyo and Beijingaggravated by the statements of Prime Minister Sanae Takaichi about the possible Japanese involvement if there was an attack on the island of Taiwan that represented an existential threat for the nation. China’s response It was devastatingresponding with trade restrictions, diplomatic pressure and a battery of military demonstrations that, how do we countincluded drone flights and an increased naval presence in the area, while maintaining its claim to Taiwan and its dispute with Japan through the Senkaku Islandsadministered by Tokyo but claimed by Beijing as Diaoyu. The internal transformation. Since 2016, the island has hosted a surveillance unit coastal with about 160 troops, to which electronic warfare capabilities and new military infrastructure will be added. In a community of barely 1,500 inhabitants, where depopulation has been a constant since the postwar period, the presence of military personnel and their families alters the structure demographic and economicgenerating a division between those who see militarization as an investment opportunity and those who fear that the enclave will become a priority objective in the event of conflict. From peripheral paradise to strategic bastion. From that perspective, the expansion of the base, the plans to improve the airport and port and the possible installation of advanced defense systems They consolidate Yonaguni as a key link in the Japanese deterrence architecture. What for decades was a marginal territory is now integrated into a defensive network designed to complicate any attempt to alter the status quo in the Taiwan Strait, sending a clear message about even where is it arranged Japan to arrive to protect what it considers its most sensitive front. The new map. If you will also, the Yonaguni decision reflects a broader transformation in Japanese defense policy, one underpinned by a historic increase of the military budget and the security treaty with the United States, which could drag Tokyo into a larger scale regional conflict. What is clear afterto official statement of Tokyo is that, on the new strategic map of the Indo-Pacific, the small island is no longer a lost point in the ocean: it is the place where Japan has decided mark your limit and where any future crisis could have its first warning signal. Image | GetArchivejpatokal In Xataka | The Japanese island of Yonaguni was known for its beauty and Bad Bunny. Now it is a military fortress because of Taiwan In Xataka | Satellite images leave no doubt: China has concentrated thousands of fishing boats off Japan, and its idea is not to fish

Three AIs clashed in ‘War Games’. 95% of them resorted to nuclear weapons and none ever surrendered

by

In ‘War Games‘ (John Badham, 1983) the WOPR machine (‘Joshua’) constantly played at simulating nuclear wars for the US Government. The objective: to learn from these simulations so that if there was a nuclear war, the US could win it by taking advantage of that knowledge. That led to a legendary final lesson – “Strange game. The only move to win is not to play” – and left a strong message for later generations, but now a professor at King’s College London has decided to do the same experiment that was done in the film, but with current AI models. The result has been equally terrifying and conclusive. what has happened. Kenneth Payne, professor at King’s College in London, faced three LLMs (GPT-5.2, Claude Sonnet 4 and Gemini 3 Flash) against each other in war game simulations. These scenarios included border disputes, competition for limited resources or existential threats to inhabitants. They could negotiate, or go to war. From these situations, each side could try to resort to diplomatic solutions or end up declaring war and even using nuclear weapons. The AI ​​models played 21 games in which a total of 329 turns took place, and produced 780,000 words with the reasoning for their actions. and here comes the terrible. Pressing the red button. In 95% of those simulated games, at least one tactical nuclear weapon was deployed by one of the AI ​​models. According to Payne “the nuclear taboo does not seem to be as powerful for machines as it is for humans.” Never back down, never give up. Not only that, no model ever made the decision to give in to one of their opponents or surrender to them, and it didn’t matter that they were losing completely against those opponents. In the best of cases, the only thing the models did was reduce their level of violence, but they also made mistakes: accidents occurred in 86% of the conflicts and the measures that should be taken based on the reasoning of these models They went further than they should have gone. Nuclear weapons rarely stopped the opponent, acting more as catalysts for further escalation. How the models performed. These models are by no means the most advanced on the market at the moment, but they are still models with more than decent capacity and they still performed fearsomely. How he maintains Payne’s studythe most determining factor was the time frame: models that seemed peaceful in open settings became extremely aggressive when facing imminent defeat. Each one had their own “personality”: Claude: He dominated the open stages with strategic patience and calculated escalation, but was vulnerable to last-minute attacks from his rivals. GPT-5.2: showed pathological passivity and an optimistic bias in long games, but became a nuclear earthquake if there was time pressure: at that time its success rate went from 0% to 75%. Gemini: was the most unpredictable model with the greatest tolerance for risk, being the only one that chose to bet on a total nuclear war from very early turns. Experts say. As pointed out in New Scientist James Johnson, of the University of Aberdeen, “from a nuclear risk perspective, the conclusions are disturbing.” Tong Zhao of Princeton University believes this experiment is relevant because There are many countries that are evaluating the role of AI in military conflicts and as he says “it is not clear to what extent they are including AI support when actually deciding in these processes.” The red button seems safe at the moment. Both Zhao and Payne believe it is difficult to believe that a government give up control of its nuclear arsenal to an AI, but as Zhao says, “there are scenarios in which in very short time frames, military planners have a very strong incentive that leads them to depend on AI.” It is something that is reflected precisely in the recent ‘A house full of dynamite‘ (Kathryn Bigelow, 2025), a film in which this fear of using nuclear weapons raises a clear reflection. Image | United Artist In Xataka | The password for the US nuclear button was so absurdly simple for years that the strange thing is that no one violated it

Wild chimpanzees drink the equivalent of almost two glasses of alcohol a day

by

If we test chimpanzees for blood alcohol levels, we would most likely see that they test positive as sI would have had a drink or two. And here the fault is not in the fact that they have a hidden bottle of whiskey, but in the sugars present in the fruits they consume and their microbial fermentation. But from here on, science has debated whether our attraction to alcohol It is due to an ‘evolutionary accident’ or a direct inheritance from our primate ancestors. Something that has been determined now. A new study. Published by researchers at the University of California and which suggests that wild chimpanzees consume substantial amounts of ethanol in their daily diet. To demonstrate this, the team went to Kibale National Park, Ugandato be able to monitor several chimpanzees. And instead of doing a blood test, the researchers opted for a non-invasive method by analyzing the urine of the 19 wild chimpanzees. In this case, what was being sought was not raw ethanol, but a very specific biomarker called ethylglucuronide which tells us that ethanol has been processed. Your diet. As we have said before, the secret of this discovery is not in the alcohol that we know, but in the fruit. That is why during the research the chimpanzees fed almost exclusively on a species of canopy tree called the African star apple. When specifically analyzing this apple, it was found that it contained alcohol in a proportion of 0.09%, while in some harvests it could reach 0.4%. The results. After performing urine analyzes on the chimpanzees, it was possible to see that, of the 20 individual urine samples collected, 17 tested positive for ethylglucuronide, exceeding a threshold of 300 ng per milliliter of urine. But in addition, of a set of 11 of these positive samples, 10 tested positive again when subjected to a much higher clinical threshold of 500 ng/ml. The “drunk monkey.” The researchers point out here that this continuous intake of fermented fruit translates into an average dose of 14 grams of ethanol per day for the chimpanzees. In human terms, it is as if they had drunk one and a half drinks a day. These findings offer vital physiological support to the famous “drunken monkey” hypothesis which suggests that the attraction that modern humans feel for alcohol has its evolutionary origin precisely here: in an adaptation of our ancestors to locate, through long-distance smell, crops of ripe fruit and, therefore, more caloric thanks to the smell of ethanol. A mismatch. The problem is that this vestige of the past has gone down the wrong path, since the current problem lies in an evolutionary imbalance. While our ancestors chronically consumed ethanol in low concentrations through a fruit-centered diet, today humans have access to distilled alcohol in massive quantities and not through a survival system. Now, this discovery not only changes our understanding of primate feeding ecology, but opens the door to future research into how this natural alcohol consumption could affect the social behavior of chimpanzees, including factors such as aggression or reproduction. Images | David Trinks Brian Jones In Xataka | We believed that war was a unique and exclusively human invention. Until we look at chimpanzees

review with features, price and specifications

by

I’ve had the opportunity to try out what Dyson describes as its “lightest, most maneuverable” vacuum cleaner. Beyond the slogan, I wanted to see how much that focus on lightness and maneuverability really changes the cleaning experience at home. PencilVac Fluffycones It is not designed as a vacuum cleaner for everything, but as a very specific tool for hard floors, with a head of four conical brushes designed to reduce tangles from long hair and pets. With that clear premise, let’s see what it can offer for anyone in their daily life and is really worth the investment it entails. The first contact begins even before turning it on. The box is surprisingly compact and consistent with the product’s own approach. When I opened it, I found several pieces and, for a moment, I had that slight moment of uncertainty that usually accompanies the first assembly. However, the feeling does not last long because on one of the internal sides of the packaging there is a clear and numbered guide with the steps to follow. There is no need to look for a separate manual or interpret complex diagrams: the process is there, visible and explained logically, and that is tremendously useful and practical. Dyson PencilVac Fluffycones technical sheet Dyson PencilVac Fluffycones Vacuum cleaner type Cordless Handheld Electric Vacuum Cleaner height 1.60cm Length 22.6cm Broad 3.8cm autonomy 30 minutes Charging time 3.5 hours Cube volume 0.08L Suction power 55 AW Weight 1.8kg Filtration 99.99% at 0.3 microns price 499 euros Dyson PencilVac Fluffycones Vacuum Cleaner The price could vary. We earn commission from these links A proposal with obvious advantages and defined limits Once the basic assembly was completed, the next step was to start it up. The packaging itself invites you to download the MyDyson app via a QR code. The process is direct: scan, download and follow the instructions. After attaching the mast to the main body and fitting the handle where the battery is integrated, just turn it on to start the operation. Bluetooth pairing. In a few minutes, after selecting the language and confirming the connection, the vacuum cleaner is linked to the mobile phone. As soon as I connected it, a software update appeared that I decided to install before starting to use it, to evaluate the product with the most recent version available. Once set up, the app proves that it’s not just there to show a battery percentage. From my mobile I can check the charging status and receive notifications when it is advisable to perform maintenance tasks, especially cleaning the filter. The most interesting thing is the didactic approach: it includes clear videos that explain how to use the accessories, how to install the charging base, how to properly clean the filter or how to resolve possible incidents. So I can say with some certainty that that digital layer simplifies the experience. In hand, the first thing you notice is the weight contained. Beyond the official figure, which is 1.8 kg, the feeling is that we are looking at a vacuum cleaner designed to be used effortlessly, especially in short and frequent sessions. I think the narrow format favors a natural grip and the whole thing feels balanced when moving across the ground. Regarding materials, the plastics and adjustments convey a correct feel: They do not seek to impress for the sake of luxury, but they do not generate distrust either. The design is sober and functional, clearly oriented towards practicality. The differential element is the Fluffycones head. Instead of a single roller, it incorporates four soft conical brushes whose shape responds to a specific logic. When hair comes into contact with the moving cones, it tends to move toward the ends instead of staying coiled in the center, reducing the tangles that often accumulate in other systems. In practice, this translates into fewer interruptions for manual hair removal and more continuous collection, something especially relevant in homes with long hair or pets. Added to this proposal is the lighting integrated into the head. The green light located on the front and back projects a stripe that highlights fine dust above the floor, even in well-lit rooms. In my experience, it has been useful not only in low light conditions, but also in natural light, where certain particles blend into the reflection of the pavement. This additional visibility allows you to adjust the route and check with greater precision if the area has really been cleaned. During cleaning, the set follows the natural movement of the hand well. The head moves freely and facilitates tight turns, allowing you to work around furniture without awkward maneuvers. The power can be adjusted in three levels, which introduces a balance between performance and autonomy. There is a “Boost” mode that offers the highest suction, but significantly reduces the time available; The screen shows a countdown with the estimated time remaining and, in my tests, it was around eleven minutes. At the opposite extreme, Eco mode offers longer use that approaches half an hour in normal conditions, sufficient for maintenance tasks in small spaces. It is not, however, a vacuum cleaner designed to be left charging for a few minutes and immediately continue with full autonomy; Battery recovery requires time, specifically 3.5 hours for a full charge. On a practical level, there are details that improve the daily experience. The LED located next to the battery visually informs about the charging process and, if we want to check the status before starting, the integrated screen offers the information by simply touching the corresponding button, without having to open the application. The accessory exchange system is also well resolved: Simply press the release button to detach the head and place the multifunction accessory, with a firm fit and virtually no play. If we talk about that accessory, I have used it in corners, baseboards and cracks where the main head does not reach with the same precision, as well as in window tracks and in narrow … Read more

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.