reason

We are increasingly looking for human answers on Reddit. That is the reason why the Google search engine is now a Reddit in disguise

by

Google has updated its search platform for the umpteenth time, but it has done so with an especially significant change. The user experience in its AI search engines (both AI Overviews and AI Mode) attempts to become more “human”. And to do this, in these searches Google will add more context to the links, such as extracts from internet forums and blogs. And if there is a beneficiary (or harmed) of that movement, it is Reddit. Google was already a gateway to Reddit. There is a behavior that Google has been seeing in its data for years and that for a long time it preferred not to publicly acknowledge: when someone wants a real answer to a real question, add “Reddit” to the end of the search. Not because Reddit is necessarily a reliable source, but because Reddit brings together real people who have experienced this issue, tried to solve it, and written about it without anyone paying them to do so. Google, with all its infrastructure and all its algorithms, had not managed to replicate that. So instead of trying it’s going to incorporate those answers directly. What exactly has changed. The search engine update will make in AI Overviews Fragments from forums, social networks and other “first-person sources” appear. When someone searches for something for which there is no single objective answer, Google’s AI will include perspectives and opinions found in all kinds of (supposedly) human sources online. Doing so will add the name of the creator of that content (or their avatar) and the origin from which said perspective comes. Google also promises to add more context about the origin of its AI-generated answers, similar to how ChatGPT or Claude include links supporting their answers. Tired of so much SEO. The reason is obvious: Google’s organic results for practical and subjective questions—”what vacuum cleaner should I buy”, “how do I cure my dog’s ear”, “what is the best neighborhood to live in in Valencia”— They are dominated by SEO and those techniques optimized to appear on Google. It is important to position, not answer the question well. That is precisely where Reddit, like other forums or personal blogs, has something that this content usually does not have: the real experience of someone who was in the same situation. Google sums it up in its own statement bluntly: “For many searches, people are increasingly looking to other people for advice.” The contradiction that Google has not resolved. There is a potential problem in this new way of conceiving these searches with AI. AI Overviews were designed to answer questions directly and thus save the user the work of clicking, reading and researching. Now they will include diverse and even contradictory perspectives from forums and social networks. So, will AI Overviews answer the question, or will it make us go back to the sources to find the answer? If it is the latter, it will not be very different from what I already did the traditional Google search engine. There is an interesting imbalance here between “we give you the answer” and “we give you context so you can find the answer.” In a sense, Google’s decision complicates searches. AI models are becoming less prone to failure. The famous cases of add glue to pizza are much less common now, and new models often boast a significant reduction in “hallucination” rates that they have. GPT-5.5 Instant, released this week, “produced 52.5% fewer hallucinations than GPT-5.3 Instant,” OpenAI indicated in its official announcement. The problem is that these hallucinations are increasingly difficult to detect because these chatbots hide these mistakes very well. That the system now includes unverified or validated content from networks like Reddit can be problematic: community votes do not always measure how truthful or useful a certain thread is. Using Reddit has its drawbacks. This platform has value precisely because it is not optimized for Google algorithms: It is chaotic and contradictory.. Sometimes there are brilliant responses from people, but other times there are completely wrong comments. When a user adds “Reddit” to their search and reads the results, they are automatically weighing which comments are useful and which are not. But that step disappears if Google extracts fragments of those discussions to include in an AI Overview. Eliminate that human filtering step and presents those answers with an authority that perhaps they should not have. Google will have much more difficulty than a human in distinguishing the comment of someone who has been working in plumbing for twenty years from that of someone who tinkers as a hobby. The shadow contract. This is not just an editorial or technological decision. In 2024 Google signed a deal worth $60 million a year with Reddit to access their data and train their models. You are not incorporating content from this social network as a public service: what you are doing is monetizing a commercial contract. Your message that you are highlighting those “original voices” is really saying that you have paid for that privileged access to Reddit content and now you are going to take advantage of that access and make it profitable. That revenue is interesting for Reddit, no doubt, but there is a problem: clicks. The Stack Overflow Precedent. There is no need to speculate much about what may happen because it has already happened. Stack Overflow is the largest technical Q&A community on the internet, but has lost most of its traffic in two years because AI companies They started collecting all those answers. to train your models and then serve them to your users directly. That caused users to stop visiting Stack Overflow and experts to stop answering questions. The quality of the new content on this network was clearly affected, and it became clear that if the AI ​​already gave you the answer without having to enter Stack Overflow, why enter? The danger for Reddit is exactly the same. Google didn’t have many alternatives. ChatGPT, Claude and Perplexity They have been capturing … Read more

Jeff Bezos’ superyacht is one of the largest and most expensive in the world. Now it is for sale for a curious reason: parking

by

At more than 127 meters in length, Jeff Bezos’ superyacht is one of the largest in the world. In fact, it is so big that even caused some problems to its Dutch builder when he was trying to take it out to sea from the shipyards. The ship was so large that it did not pass under a bridge, over which it was even considered disassembling it. It was just the first of the problems that Jeff Bezos was going to have with the size of his ship. According to advanced Page Sixnow the millionaire would be considering putting the Koru up for sale. The reason has nothing to do with the price or its maintenance. The problem is that the boat is so big that it doesn’t fit almost anywhere, and wherever it manages to get in, everyone instantly recognizes it. A huge boat with a price to match. The Koru is a three-masted schooner more than 70 meters high. built by the Dutch shipyard Oceanco and delivered to Bezos in April 2023. At 127 meters in length, it was for a time the second largest sailboat in the world and is currently among the largest in its category. In fact, it is so big that when it approaches Miami, Jeff Bezos’ usual place of residence since his move in 2025, the luxury sailboat must moor. along with large cargo ships and oil tankers because it doesn’t fit in the nearby marinas. Its construction cost around 500 million dollars and is accompanied by a support ship called Abeona, valued at another 75 million dollars. According to calculations of Robb Reportkeeping both vessels in operation costs about $30 million a year. Practically pocket change for someone who could spend a million dollars a day and still it would take more than 548 years in ruining. The problem: parking. According to a source close to the millionaire consulted by Page SixBezos considers that the yacht has become “too big to manage.” But it’s not just about the size: the Koru has become so popular thanks to its owner, that it is impossible to maintain privacy where it anchors. Hide a sailboat the size of a ten-story building off the coast it is not a simple task. One of the drawbacks of the Koru’s size is that, for example, the millionaire could not even get close to it. the marina of Monte Carlo during the last Monaco Grand Prix, a sporting event in which millionaires watch the cars pass by without even getting off their yacht. The Koru, on the other hand, had to settle for remaining anchored far from the moorings and use a small boat to get to land due to its enormous proportions. Something similar happened during the celebrations prior to the Jeff Bezos’s wedding and Lauren Sánchez in Venice, where the Koru had to remain anchored in the middle of the Venetian lagoon because it didn’t fit at the moorings near Venice. A sale without an official price and many unknowns. At the moment the sale has not been confirmed by any intermediary or by the founder of Amazon himself, and it is also not clear if the Abeona support ship will be part of the sale agreement. What does seem certain is that Jeff Bezos could be tired of all the inconveniences involved in operating a boat of that size, and would be considering buying a somewhat more discreet and manageable superyachtwhich does not cause so many “parking” problems. In Xataka | We already knew that superyachts were floating mansions: Roman Abramovich’s is a fortress with an anti-missile shield Image | Oceanco, Smithsonian

Barbacid’s promising cancer study has been withdrawn. The reason is not science, it is a “hidden” spin-off

by

Last December, the team led by the prestigious researcher Mariano Barbacid filled the headlines of the main media with great news: had found a triple therapy to eliminate pancreatic tumors in animal models. Very relevant news because of how deadly pancreatic cancer is and how it affects our society, but now this euphoria has hit a wall after the decision of the US National Academy of Sciences to remove the item from PNAS magazine. The context. The original article, published on December 2 of last year, was not just another publication, but described the results of administering three drugs in 45 mice who had pancreatic cancer. And although it was a preclinical study that had not been tested in humans and was the expected next step, it generated great expectation. The promise of a cure, even if it was in the animal phase, propelled intense fundraising campaigns to be able to start a clinical trial with humans as soon as possible. In this way, foundations such as CRIS against cancer achieved raise 3.7 million euros in the heat of these advances and thanks to the media showcase that was given to them. And now they withdraw it. The first thing to keep in mind when faced with so many alarmist headlines is that it is not removed from the PNAS magazine because the results have been invented or exaggerated, but rather the reason lies in the omission of important information regarding to conflict of interest. In this case Mariano Barbacid, taking advantage of his status as a member of the National Academy of Sciences of the United States, used a “fast track” of publication that is reserved for academics of this institution. The problem is that this privilege requires scrupulous and impeccable transparency. Data omission. As detailed by El Paísthe alarms went off in February 2026, when the academy received notices about possible conflicts of interest that have now led to the sudden retraction of the article. The problem is that Mariano Barbacid, along with researchers Carmen Guerra and Vasiliki Liaki, are co-owners of Vega Oncotargetsa spin-off which was born in the ecosystem of the National Cancer Research Center (CNIO) with the aim of developing and marketing therapies against pancreatic cancer like this one. This is why informing the journal that the authors had a direct economic and business interest in the success of the study is a violation of the most basic transparency regulations in scientific publication. It always happens. When a researcher wants to publish the results of his or her research, a lot of data must always be provided, both about the method that has been followed and everything behind it, such as the source of financing or the conflicts behind. For example, if a researcher owns shares of a large pharmaceutical company and studies one of its drugs, logically good results will benefit him because the value of the company will increase. And this is something that should always be reported so that anyone reading the research knows if the researcher may have been influenced by an economic component. And in this particular case, the fact that there is already a company that will commercialize the future therapy that is being investigated is logically something that must always be specified, because if the study goes well, it logically benefits the company enormously. There are already answers. As we say, PNAS sanctions bad practice when it comes to being transparent, but in no case does it indicate that the research is poorly done. Along these lines, Carmen Guerra has already admitted the error, as El País points out, and has confirmed that the team has resubmitted the article with this correction, detailing that they do have participation in Vega Oncatargerts. The problem is that now they are going to have to go through the entire standard review process and the republishing will not be fast. Images | UPV brgfx on Freepik In Xataka | Mice today, hope tomorrow: researchers have managed to attack pancreatic cancer before it forms

TSMC is not going to use its High-NA machines at the moment and has a compelling reason not to do so

by

On April 23, TSMC made official a strategic decision very important: has postponed the adoption of ASML’s extreme ultraviolet (EUV) and high aperture lithography machines until 2029. These are the equipment of manufacturing of more advanced integrated circuits that this company from the Netherlands currently has in its portfolio, and TSMC’s announcement caused ipso facto a drop of 3.3% of the value of its shares. It is not in vain that this Taiwanese chip producer is ASML’s largest client. In 2025 23.9% of total sales of this Dutch company came from TSMC. The main reason why this last company has decided not to use UVE High-NA machines of ASML in the short term is strictly economic. Each of them has a price of around 350 million euros, and, in addition, a single cutting-edge semiconductor plant requires the installation of several dozen of this equipment. TSMC considers that they are currently too expensive to make the manufacturing of advanced chips profitable. And, interestingly, Intel, Samsung and SK Hynix they are already adopting High-NA technology. This decision by TSMC brings great technical challenges The step taken by TSMC has not been improvised, as might be expected. In fact, over the past two years several managers at this company have publicly expressed doubts about the short-term adoption of ASML’s High-NA equipment. In January 2024 CC Wei, the current president and CEO of TSMC, surprised us with this statement: “We are studying it carefully, evaluating the maturity of the tool and examining its costs. We always make the right decision at the right time in order to offer the best service to our clients,” Wei assured. during a meeting. A few weeks earlier Szeho Ng, an analyst at China Renaissance, predicted that TSMC would not use ASML’s high-aperture UVE equipment until it introduced its 1nm integration technology. “We always make the right decision at the right time with the purpose of offering the best service to our clients” Last week it was Kevin Zhang, TSMC’s deputy chief operating officer, who clarified something very important: “I am amazed by our R&D team. They continue to find ways to drive technological development without using ASML’s High-NA UVE equipment. Someday we may have to use them, but right now we can continue to reap the benefits of current EUV technology without moving to High-NA which, as we all know, is extremely expensive.” In 2029, TSMC intends to have the A12 and A13 integration technologies ready for large-scale production, which are nothing more than derivatives of its A14 photolithography. From a commercial point of view these will be the first 1.2 and 1.3 nm technologies of this company. They will use GAA transistors (Gate-All-Around) and NanoFlex Pro technology. This latest innovation will allow IC designers to use fast cells for the critical parts of the GPU that need speed, and dense or efficient cells for the rest, thus optimizing the chip area down to the last millimeter. What we still do not know is what technical solutions TSMC engineers are going to implement to make it possible to manufacture 1.2 and 1.3 nm integrated circuits using ASML’s UVE equipment. It’s just a guess, but it seems unlikely that they will resort to the multiple patterning because this procedure compromises the performance per wafer and the cost of the semiconductors. TSMC would lose competitiveness. One last note: the multiple patterning Broadly speaking, it consists of transferring the pattern to the wafer in several passes with the purpose of increasing the resolution of the lithographic process. Image | ASML More information | Innovation Origins In Xataka | Bill Gates has X-rayed Intel. And his diagnosis is overwhelmingly accurate.

Your t-shirt from 15 years ago is better preserved than the one from last summer for a very simple reason: the yarn

by

Among the things that are no longer what they were cotton t-shirts take center stage. That’s because we have physical evidence: almost all of us have some cotton garment from years ago with better look than a newly purchased one. Why is this happening? For technological development. Like any industry, the textile industry seeks to be more efficient, to achieve more with less. What is sacrificed in the process is quality. However, all is not lost and those great t-shirts from before can still be found, if you know what to look for (and when you finish reading this article you will). In the case of cotton fabric, quality is determined by two variables: The raw material. The spinning process. Let’s start with the raw material. The best varieties of cotton are Sea IslandEgyptian, Egyptian American and pima. Its fibers exceed 32 millimeters in length, and are fine and resistant. Despite this, its production volume is ridiculously low in proportion to international volumes: only 3%. The problem is that it grows slowly and its harvests are small. 90% of world cotton production belongs to the type American Upland. Within this variety there are different levels of quality, depending on the thickness and length of the fibers, but these do not usually exceed the 25 millimeters. The problem is the yarn Depending on the quality of the fiber, one spinning process or another is used, causing the fabrics to improve or worsen greatly. The vast majority of advertising t-shirts or brand t-shirts fast fashion They use a yarn called “open end”. What does it consist of? It is a technology developed in the ’60s. in Czechoslovakiacreated specifically to take advantage of low-quality, short-fiber cotton. It is very similar to the manufacture of cotton candy: the fibers spin in a giant drum and if you start pulling on a rope, they wind around it, creating the thread. (Keagan Henman/Unsplash) A T-shirt made with “open end” cotton has a less soft touch, breaks faster and pills after a few washes. The explanation? If we greatly enlarge the photo of such a thread, we see that it looks like a tangled skein, with many ends sticking out. It is precisely these ends that form the balls. The problem is not only the use of short-staple cotton. In addition, in the process it is damaged even more: with the rotation of the drum, any imperfection on its surface functions as a bladechopping up the fibers. To improve the quality of this yarn, surgical steel drums have begun to be used. Its surface is very smooth, which reduces breakage. This is how, for example, t-shirts are made. Fruit of the Loom marked with the Belcoro seal. Improving the “open end“ The next level of quality is the threads ring spun. In its manufacture, medium-length fibers and an older technology are used, which imitates manual spinning: the cotton is carded to eliminate impurities and to align the fibers that are then twisted and stretched. The result is a smooth, uniform thread, with few protruding ends. A fabric made with these threads is much more resistant, hardly pills and has a pleasant touch. It is used by brands such as Gildan, Russel or Anvil. But the maximum level of quality in t-shirt fabrics is ring spun combed cotton. For this type of fabric, the highest quality fibers, the longest, are reserved. In addition to following ancient spinning technology, this cotton includes an additional step after carding, combing. In this process all the shorter fibers are removed, discarding up to 15% of the cotton. A fabric of this type is resistant, has a smoother and more uniform surface and is softer to the touch. They are easy to identify, because the inside label usually says “combed cotton.” However, a loss of 15% of cotton cannot do anything other than make the material more expensive, so they are also the most expensive. (Jason Leung/Unsplash) The brands that use this type of fabric are American ApparelSol’s, Bella & Canvas. Even so, it is advisable not to blindly trust brands. Not all t-shirts from a producer are made with the same cotton. What is not very relevant to the quality of a t-shirt is the information that many of us use: the weight of the fabric. A thick, heavyweight, knit t-shirt open endit will be of poorer quality and will last less than a thin combed cotton t-shirt. Comparing the weight of the fabric makes sense only within the same quality. So if you are looking for a cotton t-shirt like the ones before, the ideal is to ask about the type of cotton it uses. If they say “combed cotton”, it is the best you can find. If they say “ring spun”, you will get a much higher quality than what you are used to. If they tell you: “What cotton? 100% cotton”, you will be able to provide all this information and differentiate between poor quality t-shirts and those that will last a lifetime. Image | https://unsplash.com/es/fotos/variada-ropa-hung-en-perchero-3JAOcgZ_ZXU In Xataka | Converse sneakers were once the symbol of the millennial generation: now they have been in free fall for years In Xataka | France had maintained the monopoly on luxury perfumes for centuries: Arab countries are taking it away

Behind oil, the US had a much more mundane reason for attacking Iran: pistachios

by

Since the United States and Israel struck Iran on February 28, unleashing a war that has lasted more than a month and now hangs on a fragile truce, the world has been attentive to the ups and downs in the price of oil and the traffic of goods such as urea either helium. Logical Your flow has been greatly damaged by the closure of the Strait of Hormuz and sectors as important as transportation, agriculture or the technology industry depend on them. There is, however, another commodity that has grabbed much fewer headlines and is equally affected (perhaps even more so) by the war: the pistachio. green gold. No market remains immune to the passage of time, but few have changed as much over the last half century as that of pistachio. If we go back to the 60s, even the 70s, talking about the world pistachio market was talk basically about Iran. The country dominated global trade, placing itself far above from rivals such as the United States or Türkiye. Today the photo is different. Has it changed that much? It comes with looking at the graph above. According to the United States Department of Agriculture (USDA), during the 2025/2026 season the US will strengthen its global leadership with 712,700 tons metric, 65% of total production. Iran takes 18% of the pie, followed not so far by Türkiye (11%). These are not current figures, but the new reality. Although the pistachio industry is a business marked by cyclical patterns of its production, its global photo has hardly changed in the last decade: the USA dominates, followed by Iran and Türkiye, which have sometimes reached exchange the second and third position. At the distancefollowed by Syria and the EU, Spain included. It’s the market… and politics. That Iran has lost its global leadership in favor of the United States is hardly a coincidence. Nor is it explained only by reasons of production or pure economics. As I remembered recently analyst Justin Fox in Bloomberg, in reality the US authorities did not begin to bet on pistachio production in California until the middle of the last century. The plantations as such did not arrive until the late 1960s and the first commercial harvest with a certain scope was harvested in 1976. However, the future of the world pistachio market has been influenced by both the geostrategic decisions made in Washington and the work of pistachio farmers in the San Joaquin Valleyin the state of California. Reviewing history. At the end of the 70s, after the overthrow of the Shah and the takeover of the embassy American in Iran, Washington imposed a trade embargo on the country that cleared the way for Californian farmers eager to dominate the national market. The trade penalty was lifted in 1981, but just a few years later the US gave another boost to its industry by applying a tariff of 241% to raw Iranian pistachios in shell. Since then the scenario has become more complicated, but its result is evident: California has become a heavyweight in global production. And with it the US, which surpassed Iran for the first time in the 2004 campaign and has been more than doubling its annual harvest since 2020. “What’s behind that takeoff?” That’s the question Justin Fox asks himself in your analysisin which he slips several ideas: this boom is partly explained by changes in water policies that led American farmers to bet on almonds and pistachios, the advantages of their production during droughts and the boost of Stewart and Lynda Resnickowners of Wonderful Company, a firm that brings together between 15 and 20% of California pistachios and found the key to popularizing the product. And for proof, a button: since the middle of the last decade, per capita consumption in the country has tripled. Beyond the geostrategic value of Iran, its weight in the oil industry or the turbulent relationship with Israel, there are those who have seen the pistachio market as one of the factors that have conditioned the relationship between Washington and Tehran over recent decades. “Hostile relations with Iran seem to have benefited California producers,” says Fox, who recalls that there is even a documentary, ‘Pistachio Wars’which “even hints that pistachio interests are partly responsible for that hostility.” Is it that important? It is estimated that the ‘vede gold’ was the 17th export in terms of value of the US agricultural industry during fiscal year 2025. And it is not unreasonable to think that this position will improve. Both for the growing popularity of pistachio, driven in recent years by the fever of ‘Dubai chocolate’as well as the commitment of US farmers. The New York Times esteem that pistachio orchards have exploded in surface area in the last quarter of a century: from around 100,000 acres in California in 2001, they have grown to more than 600,000. And the war came. At this point the question is obvious… How is the war in Iran affecting the world pistachio market? There are those who believe that the American industry will be one of the best stops. “This war will limit what Iran can make and export to customers in Europe and China,” explains to TNYT Adam Orandi, responsible for a pistachio tree extension in San Joaquín. It is not only about a possible loss of strength of the Islamic Republic in the market, but about the behavior of prices. Orandi is not the only one who has pointed in that direction. In recent weeks other voices have speculated about the benefits that California companies could obtain, especially considering the good estimates of harvest that they handle in the US. Click on the image to go to the tweet. Has the war affected that much? Yes. A few weeks ago Times of India slid and to some of the threats that the war represents for the Iranian pistachio trade: logistical paralysis (conditioned by disturbances in maritime routes), the increase in premiums charged by insurance companies, power … Read more

This planet is too big for its star. When we tried to find out the reason, we found something even more disconcerting.

by

The universe is so immense that it should not surprise us that it is full of exceptions. But even so, there is still such disconcerting findings that obsess astronomers. This is, for example, the case of TOI-5205 b, an exoplanet that attracts attention due to its size, too large for its star. That alone would be truly exceptional, but a new study has found that, if that were not enough, it also has a very unusual atmosphere. Too big for a red dwarf. TOI-5205 b is a gas giant, slightly larger than Jupiter. But only a little. While Jupiter orbits the Sun, this exoplanet orbits a red dwarf. That is, a relatively cold and very small star, with a mass ranging from 7.5% to 50% of the mass of our Sun. Typically, stars are MUCH larger than the planets that orbit them. However, the radius of this red dwarf is only four times that of TOI-5205 b. To continue with the comparisons, our Sun has a radius approximately 10 times larger than that of Jupiter. And it’s not just a radio issue. The mass of this exoplanet is also striking, as it is equivalent to 0.3% of the mass of the red dwarf. Jupiter’s mass is approximately 0.095% of the Sun’s mass. All this tells us that TOI-5205 b is too big for its star. An even more disconcerting clue. Recently, a team of scientists from NASA and the Carnegie Science Institute decided to study the composition of the atmosphere of TOI-5205 blooking for clues about its origin that explain why it is so big. However, what they discovered was even more disconcerting. They carried out the analysis of the atmosphere studying the transit of the planet. That is, analyzing the changes in the light of its star when the planet passes in front of it. When light interacts with the planet’s atmosphere, it interacts with the molecules found in it. Each element reflects light in different bands of the electromagnetic spectrum, so, with the help of a spectrograph, you can know which elements the light has interacted with and, consequently, what the composition of the atmosphere is. Astronomical metals. For astronomers, any element other than hydrogen or helium is considered a metal. Just for them, chemists don’t like this. The point is that this decision the concept of metallicity arises. It refers to the proportion of metals that a planet or star has in its atmosphere. When a star forms, it is assumed to take most of the hydrogen and helium present in the stellar nursery with it. Therefore, when a planet later forms around it, it is normal for its atmosphere to have a higher proportion of metals. For this reason, it is said that the metallicity of the planets is higher than that of their stars. But with TOI-5205 b that does not happen. According to analyzes of its transit, its metallicity is lower than that of the red dwarf. hidden metals. To verify what this phenomenon is due to, the authors of the study that was recently published carried out a series of mathematical models. With them, they wanted to check how the atmosphere of this exoplanet could have evolved under different scenarios. This allowed them to verify that the current situation is consistent with their metals having been buried inside the planet. It is true that when it was formed absorbed a greater amount of metals, since the star had taken more helium and hydrogen. However, these metals did not remain in the atmosphere, but were they saved inside TOI-5205 b. In the atmosphere, however, there is some helium and hydrogen, but also other compounds, such as methane and hydrogen sulfide. What this exception teaches us. As explained in a statement One of the authors of the study, Anjali Piette, “these findings have implications for our understanding of the process of giant planet formation that occurs early in the life of a star.” Sometimes, it is the exception to the rule the one that can provide us with the most data. There’s nothing like thinking outside the box. Image | Katherine Cain (Carnegie Science) In Xataka | Since we were children we have been told that Jupiter is enormous, colossal, exaggeratedly large. It is 8 km smaller and that changes everything

the reason is the laws of physics

by

Surely you already know (online advertising is reminding you day in and day out) that with a simple prompt you can generate a video game. The AI ​​does it for you, but what it can’t do is play it. The reason is not that games are difficult in the abstract: it is that the real world obeys the same physical laws everywhere, and video games do not. Do, not play. The paradox is striking: with tools like Cursor either Claudea prompt generates a clone of a functional classic game. ‘Asteroids’, for example. However, that same system would not even surpass the first level of its own creation. Julian Togelius, director of the Game Innovation Lab at New York University and co-founder of the testing company Modl.ai, has been investigating why for months, and has broken it down in an interview. Programming is not a game. Togelius defines programming from a structural point of view: a very well designed game. Each line of code comes with a clear statement, a verifiable success criterion and feedback on possible failures, and the program indicates exactly where and why it failed. LLMs (language models) have been trained with massive amounts of code and fine-tuned using reinforcement learning to solve exactly those types of problems. Programming is, in terms of task structure, an exceptionally “well-behaved” game, as Togelius defines it. That’s why so many people find programming fun. However, video games are another story: the action space is governed by more arbitrary rules, feedback can be immediate or take hours to arrive, spatial reasoning is essential and the margin of error is much smaller. When an AI model is asked to play something, the result documented in the paper that Togelius made is unequivocal: “absolute failure.” With a guide, please. Gemini 2.5 Pro completed ‘Pokémon Blue’ in May 2025, but it took considerably longer than any human player, made repetitive mistakes, and relied on auxiliary software to achieve it. The TIME magazine analyzed Why the best AI systems still struggle with ‘Pokémon’. And that is one of the few titles that manage to finish. They achieve this because these systems have specific APIs to consult strategic guides. That ‘Pokémon’ or ‘Minecraft’ (another title that AIs can navigate) are two of the most documented franchises in the history of video games, with millions of hours of walkthroughs available on the internet, is the key to making it easier for them. The key is in physics. But… why can a language model write an essay on quantum physics and at the same time fail in both ‘Halo’ and ‘Space Invaders’? Togelius’s response is that “those two games are more different from each other, in a sense, than two different academic essays.” Looked at another way: video games are very heterogeneous. Each one invents their own rules, their own space logic, their own reward system. The mechanics of a platform game are absolutely different from those of a ‘Tetris’. Spatial reasoning (where objects are, how they move, how they relate) does not appear in the pre-training data of the language models because it cannot be understood from one game to the next. However, let’s look at a task seemingly more difficult than playing ‘Super Mario’: driving a self-driving car. And AIs do that well. The difference with games is that the real world obeys the same physical laws anywhere on the planet. The asphalt behaves the same in San Francisco as in Shanghai, the traffic lights follow the same principles, the vehicle always responds the same. As Togelius points out, “driving is much more homogeneous than video games as a whole.” Learn to drive and you can do it anywhere on the planet. Learn how to play ‘Doom’ and you will have no idea how to play ‘Age of Empires’. The definitive criterion. That is why Togelius proposes video games as a criterion to determine the success of an AI: it is necessary to gauge whether an agent capable of learning can complete any game in the top 100 on Steam in approximately the same time as a skilled human player, without access to prior documentation or specific integration. To that scale (which does not require winning on the first try, but rather learning at a human pace) there is no system today that comes close. Header | Photo of Erik Mclean in Unsplash In Xataka | AI entered video games as an experiment. Today more than 80% of developers no longer know how to produce without it

for whatever reason, it resists nuclear explosions

by

China has just landed a project that has been on the table for a decade: that of the first floating island of deep sea research of the world. It looks like an oil well, but it is actually a megastructure that is prepared for everything. And, when we say “everything,” we include the end of the world. Because it is a research center, but also a command center and a nuclear bunker. And it has already raised concerns about possible dual use by China. In short. A few months ago we echoed the China’s plan to formalize this project that has been in the oven for years, but that has not materialized. It was now when, according to the state channel CCTV Newsthe country will begin the final phase of design and construction of the platform. The name by which it has been known until now is the “Deep-Sea All-Wather Resident Floating Research Facility”, and it will basically be a facility that can do everything. The installation. Jio Tong University in Shanghai is in charge of the bulk of the project, a platform that will have a double twin hull of 78,000 tons and scandalous dimensions. We are talking about 138 meters long, 85 meters wide and a main deck located 45 meters from the waterline. He Fujianthe brand new new generation aircraft carrier from China, has a displacement of 80,000 tons. To contextualize. It is semi-submersible and, from the University, they have detailed that it will come to fill a gap that they have detected in the country’s arsenal: that of a research facility that can navigate quickly and remain in an area of ​​operations for prolonged periods. “And, if it reminds us of an oil well, it is because they have been inspired by those facilities, they have combined the design with that of the research vessels and what has come out is… well, what we see in the conceptual image. Investigation. The idea is that the facility can house almost 240 people for months thanks to the backup energy systems and the main objective set out by CCTV News and Jiao Tong University is to explore deep waters and serve as a research field. mining system testingoil and gas prospecting, as well as research into that unexplored ocean floor. Bombproof. But there is a twist. The structure is designed to be a fortress capable of withstanding nine-meter-high waves and category 17 typhoons, the highest for this type of cyclones. It is normal since it will be in areas where the hull can suffer, but what is no longer so normal is that the armor has been designed to resist nuclear explosions. Instead of conventional steel armor, the walls of the complex will be like a sandwich with several layers that will dissipate the shock wave from a nuclear explosion. For construction, it is proposed to use a metamaterial that, under pressure, compresses to create a denser structure than thicker steel panels. The simulations indicated that these walls will resist more pressure than that of a submarine. And that, together with the fact that it will have a command center, has raised some doubts about the possible dual use of the facility. Mapping the terrain. Because the ocean floor has become the new battlefield. We don’t even talk about space, since the United States claims that There has been a war with China and Russia for months for control of space, but the launch of this platform project comes shortly after the publication of some information that reveal how China has deployed dozens of research vessels to map the ocean floor. According to reports published in Reutersdozens of ships have been studying the terrain for years, mapping it and deploying sensors in a strategy to be able to monitor in real time data such as water temperature, salinity, the best prospecting areas and… also everything that moves in that territory. From the United States it has already been raised the voice pointing out that these civilian research vessels “can” collect military intelligence, which represents “a strategic concern.” This systematic mapping, for some military analysts, has a single objective and it is not to find oil wells: it is to erode the advantage that the United States had in the oceanic battlefield.” And a facility like the one they are now preparing with their sights set on 2030 can be a true marine fortress. Image | SJTU In Xataka | Japan has dozens of “forgotten” islands off the coast of China: it is now preparing for the worst scenario

There’s a reason Warren Buffett is known as the “Oracle of Omaha” and this chart is the best proof

by

There are numbers that don’t need much explanation. Just take a quick look at them to understand why Warren Buffett has been considered the best investor in history. The difference between the cumulative performance that have generated the investments that Buffett has made throughout his career as an investor through Berkshire Hathaway, and what it would have meant to invest it in any security in the S&P 500 index during the same time is not a small advantage: it is an abyss. In its traditional letter to its shareholders The company collects on a single page the history of returns obtained since 1965. That page is, possibly, the most eloquent summary of what it means to invest well over a lifetime. The data portal VisualCapitalist.com has prepared a very illuminating graph with this data and a simple glance is enough to realize the enormous difference between the returns obtained by Buffett and those of the S&P 500. So much difference, that they even had to zoom in on the image so that the evolution of the S&P500 can appear. Six decades of financial returns He annual report of Berkshire Hathaway 2025 reflects the evolution of the company’s value per share compared to the performance of the S&P 500 since 1965. The accumulated result during that period is no less than 6,099,294% for Berkshire, compared to 46,061% for the most followed stock market index in the world. Translated into money, those figures indicate that if you had invested $100 in Berkshire Hathaway in 1965, Buffett would have turned it into more than $6 million today, while that same amount put into the S&P 500 would have grown to just over $46,000. The compound annual gain that Berkshire has delivered between 1965 and 2025 was 19.7%, almost double the 10.5% achieved by the S&P 500 in that same period. It is not a difference of a few percentage points that accumulate; when we talk about decades, that gap becomes an abyss. He compound interest It has that multiplier effect, and Buffett understood it better than anyone. Good years and bad years (which are not so bad) Despite the good historical data for the entire period, not all years ended with green numbers. The table provided by the company each year also reveals that Berkshire Hathaway did not outperform the S&P 500 every year. In 2025, for example, Berkshire’s value per share rose 10.9% while the S&P 500 it did so by 17.9%. There were other exercises with similar results: in 1967, the S&P 500 index rose 30.9% compared to 13.3% for Berkshire; In 1999, Buffett’s company lost 19.9%, while the S&P gained 21.0%. In 2019, the S&P 500 soared to 31.5%, while Berkshire Hathaway only managed to post a not inconsiderable 11.0% return. Buffett wasn’t playing to win every battle, he was playing to win the war. But where the veteran investor’s strategy especially shined was when the bad times came. The S&P 500 closed negative 13 times between 1965 and 2025. However, Berkshire weathered the storm better than the S&P 500 in 11 of those 13 years. That is, it lost less during the worst years and even ended positively when everyone lost. The only two exceptions are found in the results of 1974, in which Berkshire closed the year with a fall of 48.7% compared to 26.4% for the S&P 500, and in 1990, when Buffett’s company fell 23.1% compared to 3.1% for the stock index. The secret of compound interest The key to this success in profitability is not in the individual years but in the consistency of the results over time. When an investment grows at a compound annual rate of 19.7% for 60 years, the cumulative effect is exponential: Each year’s profits are added to the previous capital and generate new profits on that growing total. It’s the difference between adding and multiplying, and Buffett made that principle the centerpiece of his entire investment philosophy. He new CEO of BerkshireGregory Abel, described in his last letter to investors This way of operating compares Buffett with the legendary baseball player Ted Williams, who divided the hitting zone into 77 segments and tried only in a much smaller area, achieving a historical average. “A similar discipline, patience and judgment define Warren’s approach to investing: determine preferred pitches, wait for them, and then strike decisively,” Abel said in the letter to shareholders included in the 2025 annual report. The Oracle’s investing nose Abel outlined in the 2025 shareholder letter the philosophy that has guided Berkshire’s investments over the decades, based on moat theory in which capital is concentrated in a small number of companies that understand their business, and defend lasting advantages. Along these lines, the report highlights Buffett’s historical investments in Apple, American Express or Coca-Cola as the best example of that approach. Businesses that, in Abel’s own words, Berkshire considers to be “companies that we understand well, have great respect for their leaders and have waited for them to grow for decades.” That investing nose, built with patience and without getting carried away by market fashionsis precisely what has made Buffett a investment legend. Coca-Cola has been in Berkshire’s portfolio since 1988 and American Express since 1991, investments that over time have generated returns that more than multiply the original purchase price. The Oracle of Omaha did not predict the future: he chose well, waited decades and let time do the rest. The same time that has proven him right. In Xataka | Seven of the ten largest fortunes in the world in 2026 are due to AI: this illustrative graph makes it very clear Image | VisualCapitalist

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.