Dangerous

Mythos will be the most dangerous AI model, but companies are already taking note of its security tips

by

Top AI companies are in the race to create the best artificial intelligence model. That race has been won by Anthropic with Mythos. At least, That’s what they claim (of course)with phrases like it is so powerful that they cannot make it public. There is reasons to take Anthropic’s words with a grain of salt, but what is evident is that Mythos is already working. Although the company has not released it, has already given access to certain technology partners. The decision is based on the company’s fear that the model will be used maliciously. They themselves have described as a threat to cybersecurity based on the number of zero-day vulnerabilities that Mythos would have found in both the main operating systems on the market and in browsers. And, just when the model is arousing opinions from some and others, Mozilla arrives to affirm that the latest version of Firefox 150 It has security fixes for 271 vulnerabilities that have been discovered thanks to this preliminary version of Claude Mythos. For its part, OpenAI does not believe anything at all. “Just as capable as a human” Mozilla it details in one of the latest posts on his blog. The company had been collaborating with Anthropic for some time and using the Claude Opus 4.6 model to find errors. In January, it found 22 vulnerabilities in a couple of weeks, 14 of them rated very serious. Of those 22 found by Opos 4.6, which is already a powerful model, we move on to the 271 discovered by Mythos. It is a huge leap and Mozilla wanted to continue investigating to see to what extent the new model surpasses Opus. Analyzing Firefox 147, Mythos generated 181 functional exploits. Opus 4.6? Just two. 90 times less. Those results have led Mozilla to write that Mythos Preview is “just as capable as the best human cybersecurity researchers”adding that they have not found any categories that humans can detect that Mythos cannot. This has another reading since, as the company itself states, seeing that the model is capable of finding so many errors in such a short time makes them wonder if it is possible to stay up to date in cybersecurity work when alternatives to Mythos are developed that do fall into hands not controlled by those responsible. There is always the fact that Mythos has not found any errors that Mozilla’s human ‘watchmen’ have not detected and that a tool like this will help to have a more secure system. All of this, in the end, pushing that narrative that Mythos is practically a technological miracle. a nuclear bomb The other side of the coin is that Sam Altman, head of OpenAI, doesn’t believe anything. Taking advantage of his recent participation in a podcast, he has qualified The entire Anthropic movement as a fear-based marketing ploy. He accuses Dario Amodei’s company (Altman’s public enemy) of wanting to restrict AI to a small number of people in a strategy that he has compared to having an atomic bomb, threatening to release it and making a living by selling bunkers to protect themselves from that same bomb. “It is evident that this is an extraordinarily powerful marketing strategy. We have created a bomb and we are going to drop it. You can buy a bunker from us for 100 million dollars” It is one more point in that historical rivalry in which both companies (and managers) have been involved for some time, but it comes just when Anthropic is having a greater role and OpenAI is being forced to release ballast in the form of services like Sora. Altman is not the only one who thinks that Anthropic is repeatedly using this discourse of “We have something so powerful that we cannot make it public” because it is a good strategy to obtain financing. There are already voices that they point that Mythos is not that big of a deal and, in fact, other models have proven to be able to do the same, finding the same errors and problems detected by Anthropic. But, above all, we must remember that, in 2019, someone already said that a model was too dangerous for public release. Who? OpenAI itself with GPT-2. Obviously, it wasn’t that dangerous. In Xataka | OpenAI and Anthropic have proposed the impossible: lose $85 billion in one year and survive

from the anti-aging miracle with scientific backing to the dangerous world of injectable ‘looksmaxxing’

by

We live in a time where people do not stop complement your nutrition with magnesium, collagenvitamins and more. But in recent months you have surely come across the famous ‘peptides’, a compound that has gained quite a bit of momentum in the world of cosmetics under the trend called looksmaxxing and also in bodybuilding. But… Do they have any scientific endorsement? What are peptides? In biological terms, peptides are short chains of amino acids that act as the building blocks of proteins, such as collagen, elastin and keratin. They are naturally in our body from the protein that we administer in our diet, and that the body uses as bricks to build the elements of the skin, muscle, immune system and many more functions. But this is something that has happened from natural biology to the field of cosmetics and nutricosmeticssince it has been seen that when these peptides are applied to the skin, they act as “messengers” that trick the body into believing that it has lost collagen, stimulating its production. In this way an anti-wrinkle effect is promised. What does science say? Unlike many “miracle supplements” that flood the internet, topical and oral peptides do have robust scientific backing, although, as always, keep in mind that you should not expect an amazing miracle when applying them. Among one of the most representative studies we have that of the Spanish Society of Aesthetic Medicine (SEME) which analyzed the effect of biomimetic peptides in patients between 40 and 70 years old. Here it was seen that, after four sessions, the biopsies confirmed real changes in the skin as there was a greater proliferation of collagen and elastin. There is more. Clinical trials with active ingredients such as Matrixyl in 93 people showed also a noticeable reduction in fine lines after 12 weeks of use, and also collagen peptide supplements such as Peptan have been shown in clinical trials to be able to reduce wrinkles around the eyes by 13% and pores by 57%. The dark side. When this becomes an obsession, that is when the problems begin, and again social networks have been a trigger. Here TikTok or Instagram has caused a trend known as looksmaxxing to break out., which in Spanish could be translated as ‘maximize appearance’. And it is nothing more than a subculture, predominant in younger men, who wants to optimize their physical attractiveness to the maximum. While the softmaxxing includes gym routines, haircuts and intense skincare, the most extreme aspect has popularized the use of injectable peptides. Suddenly, it’s not uncommon to see videos of content creators showing off refrigerators full of vials that promise ultra-luminous skin, instant muscle recovery, extreme fat loss, or hyper-defined jawlines. And it is a danger. The big problem with these injectables is that they are often not regulated, and people resort to the Internet to buy them thanks to the legal loopholes created by products under investigation or not suitable for human consumption. Here different organizations have launched alerts warning that using products that have not been authorized by the responsible agencies can cause serious risks such as infections, abscesses or even tissue necrosis. A gym shortcut. Beyond wanting to maximize beauty and reduce the number of wrinkles, in the world of bodybuilding, peptides have burst onto the scene, presenting themselves as a modern alternative to classic steroids. In this way, it is not uncommon to hear about BPC-157, TB-500 or CJC-1295, which are peptides that promise great aesthetic results. His promises. BPC-157 or TB-500 have been dubbed ‘healing peptides’ due to the great fame they have gained by promising regeneration of tendons, ligaments or muscle tears in record time. Another popular group are secretogogues such as CJC-1295 They stimulate the pituitary gland to produce growth hormone ‘naturally’, which enhances muscle growth and fat burning much more quickly than by directly injecting the hormone. Your problems. The problem in the field of bodybuilding is the same as in the aesthetic field, but multiplying the doses. Here science points out that although it is true that the BPC-157 can ‘heal’ tissues, at the moment it has not been approved by the EMA or the FDA for this use because there are a lack of studies to support its safety and the recommended dosage. Furthermore, playing with hormonal levels does not come for free, and without going any further, enhancing the release of growth hormone to have more muscle increases IGF-1 levels. And this in the long term can cause insulin resistance and, therefore, open the door to type 2 diabetes in several years. That is why the recommendation is always to avoid their consumption without medical supervision and logically if they have not been approved by the agencies responsible for drug control. Images | Norbert Buduczki In Xataka | Magnesium, creatine, collagen: we are taking supplements above what science believes is useful

Booking has been hacked. If you thought phishing was dangerous, wait until you see the follow-up phishing attacks

by

Basic-Fit’s hack yesterday It has not been the only relevant event in terms of cybersecurity in recent days. Last weekend several Booking users received emails with less than reassuring content. In these messages, the company reported that a cyber attacker could have had access to the information on its reservations. On Monday Booking confirmed that the security flaw existedbut has not given too many details about the problem. Your name and reservations were leaked, your card details were not. The information accessed by the attacker(s) includes names, email addresses, phone numbers and booking details. However, Booking has highlighted that the users’ financial data have not been part of this unauthorized access and they have not accessed the users’ home addresses either. To try to mitigate possible problems, the company forced reset of backup PINs of all affected reserves, both active and past. Too many unknowns. Although it has confirmed the incident, Booking has not provided clarification on it and it is not clear if its systems were hacked directly or the problem occurred through other means. There are also no details on the number of users affected nor is it a problem of real scope or limited to certain countries or regions. Booking has indicated that it will inform affected users individually without giving figures. According to its own website, Booking manages hundreds of millions of reservations a year and it is estimated which have about 135 million users of their mobile app. Phishing attacks have already started. These types of data thefts are exploited for massive phishing attacks, and it appears that such attacks have already begun. At least one user indicated on Reddit that he had received a suspicious message on WhatsApp with details of his reservation and personal information. That seems to confirm that the attackers were already using the stolen data to deceive customers before the public announcement occurred. But beware of “tracking” phishing. But in this case the risk is somewhat greater because this is the type of platform from which we are not so surprised to see messages that inform us of the follow-up of the reservation (with the style “There is one week left for your trip!”). Precisely these types of smishing messages can now be generated by attackers fraudulently leveraging the reservation data they have extracted to appear legitimate. If you are a Booking customer and have a pending reservation, be especially careful if you receive one of these follow-up messages. It’s not the first time. In 2021, Dutch regulators fined Booking.com with 475,000 euros after a hack exposed the data of more than 4,000 customers, including credit card information in some cases. On that occasion, Booking notified the Dutch authorities of the cyberattack 22 days late, well above the 72-hour limit required by the GDPR, which caused the company to be fined. In June 2024, the platform itself warned that phishing attacks against its clients had increased by 900% thanks to the use of AI. The company has reported the security breach to Dutch authorities, but it remains to be seen if it again took too long and could face further fines. What to do if you are a Booking user. Theoretically nothing if you have not received an email from Booking.com notifying you of the problem. If you receive it, it is important that you distrust any message, call or WhatsApp that mentions details of your reservation even if they seem legitimate. Attackers may have data about your reservations and may be using it to deceive you. You should not provide your financial data through any channel other than the platform’s official website or app. This data can be used for phishing attacks from other services that use your name or email, since this information is usually sold to be reused by other groups that carry out massive phishing attacks. In Xataka | A family paid 1,800 euros for a tourist house in Galicia. Upon arrival there was no house and no response on Booking

The most dangerous time of the drought is now. Just when we have the reservoirs full

by

Spain has just officially emerged from the drought that it has been dragging on since 2021. And no wonder: the reservoirs are at 83.5%; That is, the highest level recorded in the month of March in the entire historical series. That also explains why we are not talking about it: restrictions have been lifted, administrations have been relaxed and, beyond some very specific places, no one talks about the drought anymore. It is right at this moment that the next water crisis is being prepared. The paradox of abundance. At least, that’s what explains Jorge Rodríguez-Chueca in The Conversation. This professor from the Polytechnic University of Madrid is convinced that now is the time to think about what would happen if it doesn’t rain more all year. Because it is precisely when water begins to run out that the system is most in danger. After all, just one dry year (without changes in consumption) would be enough for the drought to return. The wettest January on record may be, for all we know right now, a mirage. What really is a drought? And it is that, according to the researcherdrought does not begin when there is a lack of water; It begins when consumption is unable to adjust to the variability of inputs. That is why we must stop reactive management and start thinking ahead. But let’s not rush: there are still scars from the drought. And, no matter how happy we are about the current situation, it would be reckless to forget that the effects of the previous drought are still with us. According to many researchers, the It started in 2021 and was the worst drought in 200 years. and, in early 2024, reservoirs reached historic lows. It is true that the situation began to recover shortly after, but it has taken more than two years (and a historic event) for the drought to end. A historical pattern. The most interesting thing about Rodríguez-Chueca’s work is the idea that, in periods of abundance, demand increases (not only is more water consumed, but more is irrigated, more permits are granted, and more facilities and parks are created). When drought hits, consumption is higher and that accelerates the crisis — and the margin for action is smaller. We have seen it many times. As explained in Datadista“since the deep drought of the 1990s, each dry period has served to implement emergency measures (…) or allow practices that were not eliminated when the rains returned, they were used to expand irrigation, increasing the problem of overexploitation and contamination of aquifers and the wetlands they feed.” Will we fall into the same mistake again? That is the big question and there are no signs to be optimistic. Above all, because the problem goes beyond what Rodríguez-Chueca points out: we are talking about a structural problem. Irrigation modernization is a poisoned gift: it reduces water needs per hectare, yes; But that has pushed many dryland farms to be converted into irrigation. Ultimately, each innovation and improvement creates a more efficient system, but more dependent on missing water. This is what has led us to this situation. Image | Anthony Da Costa In Xataka | In the middle of one of the most extreme droughts in living memory, Catalonia has had an idea: start cutting down trees

Moving ‘Guernica’ requires a complex and dangerous operation for the painting. Now the Basque Government wants to do it

by

‘Guernica’ is an unusual painting in many aspects. Its history is. It is he tour that took him to several continents during his first decades. And so is its size, much (very) larger than the vast majority of paintings that hang in museums. This sum of factors explains why it is now at the center of a bitter controversy. The Basque Country wants to temporarily take it from Madrid to Bilbao to celebrate the 90th anniversary of the bombing which inspired Picasso, but its current custodian, the Reina Sofía, believes it is a bad idea. The debate is served. What has happened? That the Basque Government wants ‘Guernica’, probably Pablo Picasso’s most famous work, finally exposed in Euskadi. A few days ago, during a meeting with the Minister of Culture, the vice lehendakari Ibone Bengoetxea requested the Government to temporarily transfer the painting to the Guggenheim in Bilbao. She wasn’t the only one. The same request Lehendakari Imanol Pradales has transferred it to the President of the Government. The idea is that ‘Guernica’ ends up in Basque lands nine monthsfrom October 2026 to June 2027. After that period, he would return to what has been his home since the beginning of the 1990s, the Reino Sofía Museum in Madrid, where he acts as the main attraction, capturing tens of thousands of visitors. Click on the image to go to the tweet. Why is it important? Because of its symbolic load. ‘Guernica’ is not just any painting. Picasso painted it between May and June 1937 in his workshop on Rue des Grands-Augustins, Paris, commissioned by the Government of the Republic. The work is also inspired by one of the most disastrous episodes of the Civil War: the bombing of the town of Guernica (Vizcaya) at the end of April 1937 by the Condor Legion and the Italian Legionary Aviation. Although during its first decades it was the protagonist of an intense journey that took it through a good part of Europe, North America and South America, the work did not land in Spain until September 1981. Some historians like The Barroquistahave interpreted his arrival as “the symbolic return of the last exile.” And why is it news? That Euskadi wants it to be exhibited in Bilbao right now, between October 2026 and June 2027, is no coincidence. It would coincide with the 90th anniversary of the constitution of the first regional Executive and the bombing of Guernica. Hence Bengoetxea has insisted in the “deep historical, symbolic and emotional meaning” that the transfer would have for the Basque people. Will it be possible? Of course it won’t be easy. Just one day after the meeting between Bengoetxea and the Minister of Culture, the Reina Sofía Museum published a report of 16 pages in which he “strongly advises against” the transfer of the painting from Madrid to the Basque Country. The reason: the process could damage it. “The work is kept in stable conditions thanks to rigorous control of the environmental conditions. However, in view of a possible transfer, its format, nature of the elements that compose it and state of conservation, together with the numerous damages suffered over time, make it especially sensitive to all types of vibrations that are inevitable in transporting works of art.” Does it say anything else? Yes. In case there are any doubts, underlines: “Such vibrations could generate new cracks, lifting and loss of the pictorial layer, as well as tears in the support.” The opinion of the Reina Sofía of course has not pleased the Basque Government, dissatisfied with both the substance and the form. “It would be serious for a formal request from a government to be responded to without a serious and in-depth analysis. The order must be an analysis of the needs so that the painting can be in Euskadi temporarily,” claims Bengoetxea. The regional Executive emphasizes that this is not a simple technical issue. In the background, they insist, there are much deeper readings that affect “memory” and “repair.” The vice lehendakari first complaint and that at the moment it has not received “any official response” from Moncloa. Is it that surprising? Yes. And no. Everything that revolves around ‘Guernica’ arouses expectation, something understandable if one takes into account that the artistic value of the work is added to its historical and symbolic relevance. However, Reina Sofía herself has been responsible for highlighting that his position is not new. In fact, it has been closing the door to organizations that request a loan for the work for several decades. In 1997 he already said ‘no’ to a request for the painting to be included in the inauguration from the Guggenheim in Bilbao, and that it arrived backed by a report in which “the technical conditions” of the transfer were detailed. Click on the image to go to the tweet. Have there been more cases? In 2000 ddenied a request of MoMA, in 2006 he did the same with the Royal Ontario Museum and in 2007 he rejected another request from the Basque Government. Two years later he again said ‘no’ to the Fuji Group, interested in including the piece in the “50th Anniversary Fuji TV” exhibition, held in Tokyo, and in 2012 he also rejected the request presented by a Korean museum. The painting’s last trips date back a few decades: in 1981 it was packed up at the MoMA for transfer to Spain, where it was first exhibited at the Casón del Buen Retiro and later (from 1992) at the Reina Sofía. There alone the exhibition “Piety and Terror in Picasso”, organized during the 80th anniversary of the work, attracted more than 625,000 visitors. And that in less than half a year. Is it so problematic to move it? The report published by the Reina Sofía Museum not only advises against the transfer of ‘Guernica’. Before reaching that conclusion, he offers a detailed analysis of the current state of the painting, in which he notes “alterations such as cracks, cracks … Read more

Three findings about astronauts’ blood have set off all the alarms. Going to Mars will be more dangerous than expected

by

We do not want to recognize it, we are not willing to accept it, we refuse to see it; but no, we are not made for space. And our persistence, in the context of large, long-duration manned missions, can cost us dearly. The last reminder has been the blood. The blood? Indeed. Three recent findings (accelerated destruction of red blood cells, platelet dysfunction in microgravity and somatic mutations of hematopoietic stem cells) make it clear that we still have a long way to go before we can enter the depths of outer space without putting our lives at risk. A giant elephant shaped like hematological syndrome. Because this is important, it is not a small health problem. None of that: we are talking about a whole hematological syndrome that affects us on numerous physiological fronts. And it makes sense: the blood leaves a lot to be desired. Is too prone to clots and too slow to clot when it is needed. Plus, he’s not very good at putting up with things either. in space more red blood cells are destroyed than are produced and that generates persistent anemia that can take up to a year to recover. This year it took place the first medical evacuation from the ISS and everything suggests that it will not be the last. A very real problem. That’s what the evacuation of Colonel Mike Finckethat space medicine is not a theoretical question. Even more so, taking into account that every time there will be more people up thereorbital health has become a key issue. What’s new? There is no big news, really: what is new is that an overall vision is now beginning to emerge. And that is giving us a clear idea of ​​the problems we face. For example, space increases the risk of thrombosis and bleeding simultaneously: they are two completely opposite things that have no clear pharmacological approach. And then? Simply be cautious. The new era of space exploration is going to expose us to the evils of space like never before. If we are not prepared, the ‘Gelsinger effect‘ may end up setting everything back a couple of decades. Image | Bradley Dunn In Xataka | NASA astronaut remains hospitalized after returning from space on a SpaceX Crew Dragon spacecraft

to open Hormuz the US is no longer going to bomb, but rather something more dangerous

by

In the Persian Gulf there is an enclave of just a few square kilometers that, despite its size, became bombed hundreds of times during the war between Iran and Iraq in the 1980s while continuing to function as one of the main crude oil outlets in the world. Their history shows that sometimes the smallest places are also the hardest to replace. The war is changing the verb. Over the weekend, the arrival of a second amphibious group US launch into the Gulf, with thousands of Marines on board, is not just another tactical move but rather a sign that the war is possibly coming to a head. a new phase: to open the Strait of Hormuz, Washington is no longer thinking only of bombing, but of doing something much more dangerous, taking the key territory. How have we been countingKharg, the small island off the Iranian coast, concentrated near the 90% of exports of the country’s oil and has become the true center of gravity of the conflict, not because it is large or defensible, but because whoever controls it control the flow economy that sustains the regime. After weeks of remote attacks, the accelerated dispatch of amphibious forces indicates that the United States is preparing the option that involves boots on the ground, a qualitative leap that transforms an air campaign into a potential occupation operation. The plan is not new, it is from 40 years ago. I remembered the financial times this morning that what today seems like an improvised escalation actually has much deeper roots, because the idea of ​​taking Kharg is not new, but is part of a script that Trump had already outlined in the eightieswhen he openly argued that the United States should directly hit Iranian oil assets to force concessions. So talked about “go and take the island” as a response to any challenge in the Gulf, and four decades later that same scheme (ultimatum, economic pressure and decisive use of force) reappears almost no changes. The difference is that now it is not campaign rhetoric, but a very real option on the table, turning an old strategic intuition into an operational plan with global implications. The economic switch of war. The logic behind this move seems quite obvious: Iran has managed resist bombing and, at the same time, maintain its crude oil exports while blocking those of its rivals, turning the closure of Hormuz into an economic weapon that puts pressure on the rest of the world. From that perspective, for the United States, taking Kharg would break that dynamic by cutting off Tehran’s main source of income and striking back in the same area, the economic one, where Iran is trying to win the war. In other words, it is not so much about destroying as to control and taketo use the island as a negotiating lever to force the reopening of the strait and, ultimately, force the regime to accept imposed conditions from outside. The impossible operation. On paper, the capture of the island could be relatively fastsupported by previous attacks and the deployment of amphibious units capable of assaulting key points such as the airport and port facilities. However, the difficulty is not in conquering Kharg, but rather in holding it: its proximity to the Iranian coast makes it an exposed target to missiles, drones and constant attacks, while American supply lines would be vulnerable in an environment saturated with asymmetric threats. That is to say, the scenario looks less like the traditional blitzkrieg campaigns of the Americans and more like a war of attritionwhere holding a small island can become a large-scale strategic problem. The risk of escalation without return. Most analysts agree on the same diagnosis: the real danger is not only military, but political and economic. Such an assault operation would imply a direct escalation against the economic heart of Iran, with unforeseeable consequences: from regional attacks to energy infrastructures (Iran, in fact, has already warned with this) to a prolonged rise in oil prices and increasing pressure on the United States to exit the conflict. Furthermore, it must be taken into account that there is no guarantee that taking the island will force Tehran to give in. In fact, it could, on the contrary, further harden its stance and widen the conflict. In this unstable balance, Kharg Island has ceased to be just a military objective and has become a strategic bet high risk for Washington: a move that could change the course of the war… or trap it in an even more dangerous phase. Image | USN In Xataka | We wonder if it is safe to fly now that there are more drones than Ryanair planes: the answer is an Ockham’s razor In Xataka | The weapon to liberate Hormuz has fled 6,000 km from the war. And that just means the US is preparing for what comes next.

the dangerous TikTok trend of chewing food with plastic that camouflages an eating disorder

by

Eating something that we love very much, but without adding a single calorie to the diet, seems like something that resembles a true miracle, but the reality is that in China social networks are being flooded with a method that promises this. And we are not dealing with something revolutionary to trick the brain, but rather eating food wrapped in plastic. Something that has been baptized like ‘plastic eating’ as El País has reported. How it started. This trend has been with us for a short time, and the origin is in Douyin, the Chinese version of TikTok. Here the videos showed young people placing a piece of plastic wrap over your mouth or wrapping food in plastic and then chewing and spitting it out. The goal here is to taste high-calorie foods without swallowing them so as not to gain weight. Extended. The algorithm worked its usual magic, and the trend has quickly spread to other countries, even reaching TikTok, where this new challenge has been replicated. And when you start with this trigger in a new population, logically you have to take into account the risks of replicating it and turning it into something viral. Especially focused on adolescents, who may be more vulnerable in these situations. Its consequence. Beyond how bizarre it is to put plastic in your mouth to enjoy the flavor, but not have the effect of the calories, the more immediate physical damage must be considered. One of the most striking, related to repeatedly chewing a packaging that has not been designed for human consumption, carries a very high risk of suffocation and also dental damage. But we must not forget that we are chewing plastic here, so there is a risk of ingesting toxins. Different medical and scientific sources warn that these practices can expose us to the consumption of microplastics, which we have already been able to talk a lot about, as they are present in some important organs such as the placenta or testicles. Something that little by little is being related to hormonal disruption. Psychological risks. Without a doubt, it is another of the most important risks that we must take into account here, since what the networks sell as a trick to reduce the cravings we have throughout the day, is actually a classic symptom of eating disorders or eating disorders. In the clinical setting, it is known as ‘chewing and spitting’, which is a very common compensatory behavior in the diagnosis of anorexia and bulimia. It is not a new idea, since the iconic designer Karl Lagerfeld popularized and defended publicly this technique years ago after losing between 30 and 40 kilos. However, science denies that it has real benefits, since different studies suggest that when we chew food, the body prepares for digestion and increases the levels of ghrelin, which is the hunger hormone. But in reality, by not receiving food, hunger and anxiety are triggered, causing a severe loss of control, metabolic alterations and malnutrition. Social networks. The proliferation of these types of trends puts the role of social networks in the mental health of young people back on the table. Scientific data provided by recent studies indicate, for example, that exposure to content that promotes anorexia on TikTok significantly decreases body satisfaction in a matter of minutes, increasing the internalization of “thin ideals.” It has also been proven that 73% of young users with moderate or high risk of suffering from an ED show symptoms directly related to their interaction on TikTok. Images | Clown World In Xataka | We believed that extreme thinness was a fashion that had happily been overcome. What is happening on networks contradicts us

Iran’s drones have aimed at the same target as the US. And now that they have pulverized it, they are going to unleash their most dangerous weapon

by

In the Middle East there are radars capable of tracking objects thousands of kilometers and distinguish between dozens of targets in mid-flight. They are machines the size of a building, cost hundreds of millions of dollars and are part of the system that detects attacks before they even cross the atmosphere. However, in the current war they are discovering something uncomfortable: the greatest danger to these technological gems may come from weapons that cost a fraction of its price. The eyes of the shield. Since the beginning of the war, Iran has directed a very specific part of his attacks against an objective that rarely appears in the headlines but that underpins the entire defensive architecture of the United States in the Middle East: the radars that allow detecting and tracking missiles in flight. These sensors (like the AN/TPY-2 associated with the THAAD system or the gigantic AN/FPS-132 deployed in Qatar) act as the “eyes” of the regional anti-missile shield, feeding data to Patriot interceptors, THAAD or Aegis destroyers to destroy threats before they reach their objectives. However, several of these systems have been hit in the last days by Iranian attacks, some confirmed through satellite images. Among them is the strategic radar of the Al-Udeid base in Qatar, valued at nearly a billion dollars, and an AN/TPY-2 radar in Jordan directly linked to THAAD batteries. Other locations in Kuwait, Saudi Arabia or Bahrain as well have suffered impacts in facilities related to radar or communications, partially weakening the surveillance capacity of the regional defensive system. The shaheds against the most expensive system. The paradox of these attacks is that many of them have been carried out with unidirectional attack drones relatively cheap, like the Shahed, whose cost is only a fraction of the missiles and sensors they try to neutralize. While US systems were designed to intercept much more expensive and sophisticated ballistic or cruise missiles, Iran has bet for saturating or damaging them with much simpler platforms. These drones fly low and slow, which can make it difficult to detect for defenses designed for faster threats. Furthermore, the country has proven to have the capacity to produce them in large quantitiessomething that is already left patent in Ukraine with its export to Russia. In this war, that industrial advantage translates into a pretty clear strategy: launch constant waves of drones against sensors, command centers and communication systems, gradually eroding the network that allows us to detect threats in the air. An Army and Navy transportable surveillance radar (AN/TPY-2) positioned on Kwajalein Atoll during FTI-01 flight testing Blind the shield. The pattern that emerges suggests that these attacks are not simply scattered retaliation, but rather part of a much more calculated approach. Radars not only detect threats, they are the element that makes it possible to intercept them. Without them, even the most advanced anti-missile systems remain partially blind or rely on incomplete information. Hitting these sensors, therefore, has a multiplier effect– Each radar out of service increases the likelihood that future waves of attacks will penetrate defenses. In that sense, the Shahed seem to have aimed at the same target since the beginning of the conflict: the eyes of the American anti-missile shield. And the more that network is degraded, the greater the scope for other, more dangerous weapons (stored in underground silos and fortified bases) can come into play with greater chances of success. A satellite image taken on March 2, 2026 shows debris around a blackened THAAD radar at Muwaffaq Salti Air Base in Jordan The problem of radars. The episode also highlights a structural weakness that analysts have long pointed out. Large early warning radars are extremely sophisticated, but also huge, expensive and largely static. Each one costs hundreds of millions or even billions of dollars and there are very few in the world, which means that replacing them can take years. At the same time, their size and fixed nature make them on relatively easy targets to locate through intelligence or commercial satellite images. Even seemingly minor damage can cause a “mission kill”that is, leaving the radar inoperative for long periods, even if the structure is still standing. In other words, a cheap drone can temporarily disable a central piece of the strategic defense of an entire region. The new logic of air war. Plus: what is happening reflects a deeper change in the way defensive systems are attacked. For decades it was assumed that destroying strategic radars required sophisticated missiles or large-scale complex attacks. The proliferation of drones has altered that equation. Today even actors with limited resources can employ cheap platforms to degrade sensors that cost hundreds of millions. This logic has already been seen in other conflictsfrom Ukrainian attacks against Russian radars to Israeli operations against Iranian air defenses. In all cases the principle is the same: “shoot the archer” before facing his arrows. If the system that detects threats disappears or is degraded, the entire shield loses effectiveness. A warning for the future. Beyond the immediate damage, these attacks have opened a broader strategic debate about resilience of American missile defense. The current architecture relies heavily on a small number of extremely valuable ground sensors. If those sensors are destroyed or neutralized, even temporarily, the defensive balance can quickly shift. That is why more and more experts advocate complementing or replacing part of these capabilities. with space sensors capable of tracking missiles from orbit, creating redundancy against ground attacks. However, these technologies, if they arrive, will take years to be fully deployed. Meanwhile, the current war has left an uncomfortable lesson: a system designed to stop the world’s most sophisticated weapons can be weakened. by swarms of drones cheap. And when the radars stop seeingthe next move on the board can be much more dangerous. Image | Google Earth, X, Missile Defense Agency, Airbus In Xataka | You’ve probably never heard of urea. The missiles in Iran are destroying their production, and that will affect your food In Xataka … Read more

will no longer pause dangerous models if the competition releases them first

by

Anthropic is in the middle of an important issue with the Pentagon in the United States that may end up shaping the future of the company. Founded with security as its reason for being, it has just rewritten the rules that defined it. And his “Responsible Scaling Policy“, the document that established when to stop the development of a model that is too dangerous, has evolved into a mere roadmap with flexible objectives. And this change is much more important than it seems. Not only for Anthropic, but for the rest of the industry. Let’s get to it. What exactly has changed. Until now, Anthropic policy stated that the company would pause training or delay the launch of a model if its capabilities exceeded the speed at which sufficient safeguards could be developed. That is to say: if the model was too powerful to be controlled safely, it was stopped. This is over. And it is that the new policy removes that automatic braking mechanism and replaces it with a series of public commitments, along with regular third-party audited risk reports. The change was confirmed by the company itself in an official statement. Why have they done it? The company gives two main reasons. The first is the competitive environment: OpenAI, Google and xAI advance without those types of restrictions. “We didn’t feel it made sense to make unilateral commitments if competitors are moving full speed ahead,” counted Jared Kaplan, chief scientific officer at Anthropic, told Time. The second, as it could not be otherwise, is political: Washington has turned its back on AI regulationand Anthropic acknowledges on its blog that the current anti-regulatory climate makes its own safeguards asymmetrical with respect to the rest of the sector. Paradox. From Anthropic’s point of view, it is not a renunciation of security, but a decision made based on it. Their reasoning: if the actors who are more responsible (they fall into this bag, logically) stop while the less careful ones move forward, the net result is “a less safe world.” The logic has a certain coherence, but it also means accepting that security depends on what the competition does. And that is a very dangerous game. Context. Anthropic was founded by former OpenAI executives, including Dario Amodei, who left that company precisely because they believed that it did not pay enough attention to the risks of AI. The new policy comes at a time when several security researchers have left the company. Just like shared Wall Street Journal, one of them, Mrinank Sharma, wrote a letter to his colleagues this month saying that “the world is in danger” because of AI, before announcing his departure. In fact, according to sources close to the media, his departure would be partly related to this decision. What’s happening with the Pentagon?. The announcement comes in full tension with the Pentagon. US Secretary of Defense Pete Hegseth gave Anthropic an ultimatum the same Tuesday that the policy change was made public: modifying its red lines on the use of Claude or risk losing a $200 million contract with the Department of Defense. Anthropic has made it clear that both issues are independent, but the temporal coincidence has not gone unnoticed. What remains of the security policy. It is not a total abandonment. Anthropic remains committed to delaying the development or deployment of “highly capable” models in specific circumstances, and is committed to publishing detailed, externally verified risk reports every three to six months. The company also now separates its own internal guidelines from its recommendations for the rest of the sector, implicitly acknowledging that the commitment to a “race to the top”, which other companies are adopting, has not worked as expected. Cover image | Wikimedia Commons and Anthropic In Xataka | The US has a message for AI companies: if necessary, that AI belongs to the State

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.