give

give our iris to Sam Altman

by

Sam Altman, OpenAi CEO, He warned the world In an intervention from the US Federal Reserve: “I am very nervous because an imminent and serious fraud crisis is coming.” The leader expressed it in the context of a speech where he reviewed how malicious actors are defeating authentication systems through artificial intelligence (AI). A problem that is already real. Sam Altman’s notices are framed in a reality that the authorities have already warned. The FBI last year of the growing use of AI by cybercriminals. The methods described to deceive federal government officials focused on voice cloning and Vishing. The Scams with cloned voices With ia They are increasingly frequentalthough there are seemingly simple solutions such as “family passwords.” However, there are many situations where something like this cannot be established in advance, as would be the call of a politician. Nor of a bishop, as has happened in a Sevillian brotherhoods, who have warned that the voice of Monsignor Teodoro Muñoz is being cloned asking for Bizums. Incibe also takes time detailing Cases of voice impersonation in which the scammers request money posing as relatives. Besides, The .es domains are fired in phishing casesand the agents will not help, but quite the opposite. Altman has interests in the way of solving it. The problem of Altman alerts coincides with real cases and alerts of the authorities. However, the OpenAi CEO is warning about a present and future situation of which it is interested. In addition to the position for which he is more famous today, Altman is co -founder and It is still involved of the controversial verification company Worldcoin (now World). The director for Europe of Worldcoin He told us So a couple of years ago: “It is increasingly difficult distinguish if you are talking to a human or if you are looking at something created by an AI. How do humanity prove in the era of artificial intelligence? This is where Altman next to Alex BlaniaGerman physicist who was in Caltech, they started working on this project “ New solutions after stopping. Worldcoin defended that Iris scan is the most infallible method to verify that we are human, although after the problems with the AEPD they had to accept that Asking for iris scan was too muchand have a new alternative with World ID credentialsa system that identifies us based on something universal: passports. Not anyone, yes, but those who have NFC. They also have World ID Deep Face, a tool that serves to confirm in video calls that the people involved are human. OpenAI is part of the problem. In 2019, more than two years before Chatgpt’s media explosion, OpenAi announced that I was not going to spread to the public The full version of GPT-2its brand new language model. The reason? The fear that a “misuse” such as that Sam Altman alert would wreak havoc. As he did Infibe with Deepseek. However, with the much more powerful GPT-3 (which was based chatgpt in launch) and GPT-4 There was no caution. In fact, according to the Wall Street Journal, what he carried in part to the Said Sam Altman himself It was that he said that three new functionalities had been approved by a security committee created in conjunction with Microsoft … and then discover that only one of those three had really approved. Ilya Sutskever wants to fix it. To this was added permission to carry out security evidence in India without having asked the Board or the Security Committee. The story ended with Ilya Sutskever leaving OpenAi after having co -founded. Way of Create Safe Superintelligence Inc (SSI), a company that pursued precisely a superintelligence with “nuclear” security. Without having “nothing” It is worth more than 30,000 million. Images | OpenAI, Worldcoin In Xataka | 017 has been attending cybersecurity for five years. The question is who calls and for what

In Pinto and Sanxenxo they could not sign doctors, so they opted for a radical solution: “give them” a house

by

The challenge brings them. In Pinto, a municipality located south of the Community of Madrid has long faced the complicated task of ensuring that their 56,000 inhabitants will meet a doctor when they go to their health center. The reason? Its outpatients are cataloged as “Difficult coverage”a handicap to which the shortage of specialists in family and community medicine nationwide and the cost of housing in the city (11.7 euros/m2), that (although it remains significantly below the prices of the capital) has seen how rents were more expensive 17.5% in a matter of one year. Given that panorama, the City Council of Pinto has opted for a radical measure: ‘give away’ floors to doctors who move to the municipality. It is nothing new. Before He already did For similar reasons another Galician town: Sanxenxo. What happened? That Pinto has opted for a radical measure to capture new doctors and guarantee health care in their health centers and PAC. So that the city squares are more attractive and the house does not involve a problem, the City Council will offer accommodation without cost to the doctors. Literally. The decision was announced by the mayor himself, Solomon Aguado (PP), At the end of Juneduring the debate on the state of the city. And how will it do it? Your idea is to offer up to six municipal ownership homes, apartments from one or two bedrooms in which doctors can settle for free. The agreement will be formalized through the bailment, a contract by which the City Council yields the enjoyment of the floor without cost for a certain time. Once that term is fulfilled the Consistory recovers the house. Except for last minute or unforeseen changes, the objective is that the measure is launched immediately, throughout The second half of the year. What result has you given? For now the offer seems to have aroused interest. Only a few hours after the announcement, in the City Council mailbox There were already applications of doctors from different areas of the country. And since then the drip of curricula has remained. Today The confidential account That Pinto has more than 40 candidates from the Balearic Islands, Galicia, Andalucía, Castilla-La Mancha, Canary Islands … and even countries such as Cuba, Venezuela, Ecuador or Mexico. Their curricula will be transferred to the Community of Madrid to include them in the bag to which, periodically, vacancies available in the town are offered. How serious is the problem? The City Council of Pinto recalls that its health centers are considered “difficult to cover” and, although in recent years the shortage of doctors has softened, the situation in the city’s consultations remains without being the ideal. Hence you want to capture more doctors. The country remember how at its worst moment the town stayed No doctors in the emergency room and the template of its two health centers was reduced to minimum levels. Free housing is not the only claim to attract doctors to Pinto. In 2024 the regional government announced an incentive of 500 euros per month For professionals who work in “difficult coverage” centers, a complement that, as the Executive himself remembered, adds to other pluses already approved. Ayuso recently He influenced again In that line when announcing a salary increase of 280 euros for nurses who work in centers with the same consideration. Is it something new? No. In recent years Other municipalities They have launched to offer Free housing to cover certain services or even stop the depopulation. However, there is a specific town that already resorted to a formula similar to that of Pinto with the same purpose: Sanxenxo, a municipality of Las Rías Baixas of 18,000 neighbors What sees how its population multiplies exponentially During the summer months. Precisely for that reason, a few years ago his City Council He had an idea To ensure that your health care is reinforced from June to September: offer free housing to doctors. The measure was supported by an agreement signed by the Consistory and the Sergas (the Galician service of Saúde) and contemplated that the municipality took charge of the rental of four homes during the summer months. The idea was the same: free house for doctors willing to reinforce attention in the centers of the area, although in its case the offer was temporary, of June 1 and September 30. The initiative worked A timebut It was abandoned This summer. The reason? Sergas can already cover places with their own doctors. In Sanxenxo the M2 for rent reached in June the € 22.8almost 33% more than a year ago. Images | Pinto City Council, Luis Meléndez (UNSPLASH) and Wikipedia (Zarateman) In Xataka | There is a Spain in which housing is built faster than homes are formed. Nor is she getting rid of the price increase

The US wants to give up bringing the most valuable samples collected on Mars. Lockheed promises to do it for less than half

by

Since February 2021, The Rover Perseverance patiently travels the Jezero crateran old river basin on the surface of Mars. Over there, where millions of years ago the water flowedNASA Robot It has been collecting fragments of rock and Martian dust With a very specific objective: Find signs of past life. It is not any mission. Is, According to NASA itselfthe first step of an ambitious plan to bring intact samples from another planet. For more than three years, Perseverance has done his job in silence and the samples that Now rest inside of small sealed tubes, carefully deposited on the Martian surface or stored aboard the Rover itself. From space, bread crumbs would look like a trace drawn through a desert planet, hoping to be collected. A truncated promise The plan, known as Mars Sample ReturnI had to send another ship to Mars, launch from there a rocket with the samples and return them to the Earth for analysis in laboratories. But the project began to crumble. An independent review raised the cost estimated until 11,000 million dollars and delayed the possible return of the samples until 2040. In May, the new US administration presented its first budget draft: proposes to cut 24 % of NASA’s financing and cancel Mars Sample Return for considering it an exorbitant cost program. The plan must still go through Congress, but marks a clear turn: the menions manned to deep space are prioritized, such as Artemisand the projects with great budget and scarce immediate return are frozen. With the current budget cut and without guarantees of continuity, NASA decided to reexamine its options. As explained by the former administrator Bill Nelsontwo more viable alternative routes were being evaluated: one that takes advantage of the “Sky Crane” type landing system used successfully in the Curiosity and Perseverance Rovers, and another that opens the door to new proposals from the private sector. Lockheed Martin’s letter Amid the budget uncertainty, one of the great space contractors in the United States has decided to move file. Lockheed Martin, with half a century of experience in missions to Mars, has presented NASA a proposal to execute the Mars Return mission with a radically different approach: for less than 3,000 million dollars and under a fixed price contract. The change is not less. Faced with the traditional model, full of budgetary risks and with multiple public actors involved, Lockheed promises a simpler architecture. Its proposal includes a more compact landing module, based on the ship Insight that already touched Martiano soil in 2018a lighter and lighter ascent vehicle – designed to be the first to take off from another planet – and a system of re -entry to the land derived from missions such as Genesis, Stardust and Osiris-Rex. It is a commitment goes beyond engineering. Being a “Firm-Fixed Price” contract, Lockheed Martin is responsible for absorb any possible extra cost. That is, if something is complicated, the invoice does not rise. According to the company itself, that model has already proven effective in other scientific missions of deep space, where they even managed to return part of the NASA not used budget. The message is clear: if NASA wants to save its most ambitious mission without spending, Lockheed Martin is ready to lead it. Bringing back about thirty small tubes could help us answer one of humanity’s great questions. Was Mars ever inhabited? Scientists do not seek fossils or complex structures. They look for subtle indications that can only be analyzed with the level of precision allowed by land laboratories. And for that, the samples that Perseverance has collected are not any rock. They have been selected one by one depending on their location, their age, their composition and their geological context. Are, In Nasa’s own wordsthe most likely material to contain a Martian “biofirm.” But the value of these samples goes beyond the biological. They can reveal how the wet marte of 3.5 billion years ago was, how its climate evolved, why it became an arid and inhospitable planet, and how the geological, atmospheric and chemical processes interacted for millennia. They will also tell us what resources could take advantage of future manned missions: where it is safer to land, what materials are usable, what areas have risks. Images | Lockheed Martin In Xataka | NASA locked four volunteers one year simulating their life on Mars. What did not miss: an entire PS4 Games Library

The heat of the summer destroys the battery of your mobile. But there is a trick to give it a second life

by

It’s summer, I get home with a 20% of battery and the Telephone quite hot. I stop to see some videos in Tiktok and … I appreciate how the battery jumps directly from that 20% to 16%. What noses have just passed? Bulk. Your mobile has an internal microcontroller that uses estimation algorithms based on different variables. Voltage Temperature Current Load cycles As the battery is degraded, The system is losing precision: Errors in the calculations (something especially simple if we only make partial load cycles) accumulate, the loss of battery capacity is not exactly estimated, and this ends up translating into reading errors that directly affect the most visible point: the percentage indicated by the phone. How to know that if my battery is not well calibrated. It is quite easy to know if your mobile has or not the barefoot battery, since its behavior will be quite erratic. It goes out before reaching 0%, in upper percentages. There are notable battery jumps, more than 1%, both loading and downloading. The device never marks 100% or 1%. That a mobile has the bare battery It does not translate into a malfunction in the same: The error focuses on the visible reading that the user perceives, not on the performance of the battery itself. Why summer affects bruise. Telephone batteries have a great enemy: heat. There is no time of the year in which they suffer greater wear and are more likely to barefoot, precisely because of the acceleration of this damage they suffer. If we do not remedy it, it will be practically inevitable that we suffer sudden turns off for ignoring the exact battery percentage, and that the mobile stops to load efficiently when reaching certain percentages (fast from 20%, slow from 80%). How to calibrate a battery. Calibrate the battery It is nothing more than the process to help software to get precise readings about the real capacity of the battery. We seek to learn new vacuum points (0%) and full (100%), so the process is something drastic. Load the phone to the fullest and leave it, at least, a couple more hours loading. Download the phone completely and leave it a few hours without lighting. By battery protection, there is some reserve capacity after that 0% that is not lost until a few hours. Load the phone up to 100%, and leave it an additional time until the phone reaches the maximum load point. Avoid using applications that promise this purpose: none works. The only way to force calibration is to perform this process manually. Is it advisable to calibrate the battery? Evidence about calibrating the battery is still limited. There are those who aim to do them every three monthsbut if we do not observe anomalous behavior, stress the battery at this level may not be recommended. Calibration is necessary when readings are wrong and end up translating into a real problem to estimate how much battery we have left. The manufacturers are in full war to achieve batteries that there are many more cycles to go down to 80%below which the industry considers the replacement of it. Image | Xataka In Xataka | Traveling by plane with a monstrous battery sounds great. Until airport safety thinks

Goal has just remembered why we should never give all our photos to a light app

by

I recognize that I am the type of person who, between allowing access to the entire photo reel or only a few, I usually choose the first, especially in apps that I use a lot as Instagram. Being able to access the entire reel and not having to take photo photo to photo is more comfortable. Total, if I don’t upload the photos, nobody can see them, right? It is not so clear. Meta wants the photos that we do not publish The news comes from Techcrunchwhere a few days ago they told how Facebook had begun to ask users for “cloud processing to obtain creative ideas made for you from your photo reel.” To create those ideas, They choose photos of the reel and upload them to their cloud. In addition, by accepting, he warns us that we also accept the terms and conditions of goal AI, which include the analysis of photos and videos, including the facial characteristics of everyone who appears in them. From The Vergethey contacted a goal to ask the question that you will surely be doing, and it is if they were using those unpublished photos To train your AI. Goal ensures that not currently, but They do not deny that they can do it in the future. Although for now this novelty only affects users of the United States, it could later reach more countries, so better to be prevented. Access yes, but better on accusagotas There are many apps that ask for access to our photos and videos and it is in our hand to decide whether we give them access or not. Both iOS and Android allow Choose what images we want to share with each app. In the case of iOS, it even reminds us periodically if we want to continue sharing our entire reel with certain apps. I am one of those who ignore warnings. Seeing cases like the goal I am clear to me that it is a bad idea. The option to limit access to photos in iOS (left) and Android (right). I have begun to limit access in apps that I use daily as Instagram and the truth is that It is not as tedious as I thought At first. When I want to upload something, I just have to give that something and that’s it. In the end it is to add another step, yes, but it is done quickly. If you want to do it you just have to go into adjustments and go to the app you want, for example Instagram and, within the gallery permission, choose ‘limited access’. The importance of the small print in the AI ​​era We take care with the photos and videos we share, but we do not think about what happens to everything we do not upload. It is normal, the logical thing is to think that, if you do not click on ‘Share’, no one except you can access that photo. This case shows the importance of reading the small print of apps, something that we are honest, very few people do. Maybe before it was not so worrying, but in the current context, with so many models, the thing changes. To train those models It takes a lot of content, very much. There has been Many controversies Around this: the use of Content protected with copyright, LinkedIn training its AI with user data almost without warning, AI models that draw platforms like Reddit And even others who train With photos of minors. If it was already convenient to be cautious, now much more. Image | Own in mockup of Freepik In Xataka | Spain wants to implement the “pajporte” for access to part of the Internet. China has a much more ambitious plan under

If the question is whether Spain can deny its bases to give air support to Israel, the answer is not so simple

by

And suddenly, Spain. Actually, and as we will see, the country’s geographical situation makes it a kind of “technical stop” for the different military conflicts in which the United States has been. The war that is taking place In the East It is another chapter, but with the exception that, quite possibly, Spain will not enter. What raises a logical question: could it refuse to use its air bases? Reinforcements from Spain. First the news. Within the framework of the growing military escalation between Israel and Iran, the United States has discreetly intensified its deployment in the Middle East with the help of key infrastructure in Europe, including Bases in Spanish territory. It is official, since the Minister of Defense, Margarita Robles, confirmed that Washington has begun to use the Rota bases (Cádiz) and Morón de la Frontera (Seville) To park cistern airplanes, a measure framed in the bilateral agreements in force between the two countries. Robles said that the presence of these aerial means is carried out within the limits established by the joint defense treaties. Specifically, the agreement allows the deployment of up to 15 replenishment aircraft in Morón, although international media Like BBC They have reported the arrival of “something else”: at least 30 KC-135 aircraft In recent days, distributed between Spain, England and Scotland. Logistic support for fighters. We have come counting these days. The role of these cistern aircraft It is strategic: They allow to expand the operational scope of combat aircraft such as the F-16, F-22 and F-35 that the Pentagon has mobilized towards the Persian Gulf region. According to Reutersthis aerial reinforcement would also include the deployment of a USE USERindicating a projective combat capacity in several dimensions. Although the Secretary of State for Defense, Pete Hegseth, has insisted that it is a deployment for purposes andStrictly defensiveUS military sources have confirmed that these platforms have already been used to Interception operationsdemolishing drones and missiles launched by Iran in response to the Israeli attacks initiated the previous Friday. Reactions in Spain. No doubt, the use of Spanish military facilities by the United States It has generated restlessness within the Spanish parliamentary arch itself. Podemos has presented a Question battery In Congress to demand explanations to the Executive on the arrival, last Friday, of the cistern aircraft to the base of Morón. In their brief, they express suspicions that these facilities are being used as a logistics scale in support of military operations in favor of Israel. The party led by Ione Belarra has also questioned the government about itself I was aware Of these movements and if it supervises its purpose, putting on the table the debate on operational sovereignty and indirect involvement of Spain in an international conflict of high intensity. A Eurofighter Typhoon from Ala 11 in Morón in 2015 Legal and preceding basis. The current use of the military bases of Rota and Morón by the United States is part of a strategic relationship Started in 1953when Spain, even under the Franco dictatorship, signed the calls Madrid agreements. In exchange for financial and military aid, the installation of US bases in Spanish territory was authorized, in what was a shy international opening step. This initial network It included the bases from Zaragoza, Torrejón de Ardoz, Morón de la Frontera and broken, and constituted one of the First gestures from Spain to aspire to NATO entersomething that would not be completed until 1982 with the firm, and until 1999 with full accession to the integrated military structure of the alliance. The agreed of 88. The current legal basis that regulates the shared use of Rota and Morón It was established With the Defense Cooperation Agreement Signed on December 1, 1988 Among the governments of Felipe González and Ronald Reagan, he was finally the final of the Cold War. This text has been amended Three occasions (In 2002, 2012 and 2015) to adapt to geostrategic and operational changes. According to the Ministry of Defense, the second amendment protocol set a period of validity eight years old since its entry into force (May 21, 2013), which made it expired on May 22, 2021. However, article 69 of the agreement provides for a Annual automatic extension If none of the parties expresses its opposite will six months in advance, which has been happening until today, with Some exceptions. Rattan Operational limitations. Although US jurisdiction governs certain aspects within the perimeter of the bases, Spain retains sovereignty and political control over its strategic use. In fact, the United States cannot use the facilities unilaterally, but requires express permission from the Spanish government, as stipulated The agreement. This principle has been maintained, although in practice it has not meant obstacles during, For examplethe wars of Iraq and Afghanistan, when the governments of José María Aznar and José Luis Rodríguez Zapatero authorized their use No restrictions. Similarly, in 2021 Spain allowed both bases to temporarily welcome Afghan collaborators of the United States during his escape from Kabul. Military presence. According to the agreementUnited States can keep up to 2,200 military, 36 aircraft and 500 civilians in Morón, although the current contingent is around 600 troops. In rota, the allowed limit amounts to 4,250 military and 1,000 US civilians. These figures reflect a gradual reduction compared to previous decades, and in 2023 The transfer of the rapid response force for Africa was confirmed from Morón to A base in Italyevidencing an operational replication that directly affects the employment and economy of the nearby areas, which receive about two million euros per year in direct income of the State. Real veto capacity. Therefore, to the big question, could Spain deny the use of its air bases to the United States with respect to the Israel-Iran conflict? The short answer is that yes, in fact, There are examplesas with the return of Torrejón in 1991 and Zaragoza in 1992. Plus: In 1986, the government of Felipe González advertisement that would not automatically renew the pact, also demanding The withdrawal … Read more

Sam Altman states that Chatgpt’s water and energy consumption is tiny. The problem is that it does not give evidence of it

by

An email of 100 words generated by GPT-4 Consume 519 milliliters of water. That was the conclusion to which researchers at the University of California arrived a few months ago after analyzing this OpenAi model. Sam Altman, CEO of the company, has just yielded its own estimate on the consumption of water and energy of each consultation of Chatgpt. And it is very different. 1,000 times less than what was said. According to Altman, an average consultation in Chatgpt consumes much less than what had been indicated in previous studies. Your data are strikingand to understand them makes interesting analogies: “As production automated in data centers automates, the cost of intelligence should approach electricity. (People are usually curious to know how much energy consumes a chatgpt consultation; the average consult (0.32 ml); A previous study of Epoch ai corroborates the data that Sam Altman has now wielded. Source: Epoch AI. And the tests? Those figures mentioned by the OpenAi CEO have a problem: they have no visible support. He throws them without citing sources or explaining where he has taken them out, something that makes it difficult to believe. A Meta executive answered the question of How much consumes the inference AI A year and a half ago, responding that “only two nuclear reactors would be needed to cover it.” But previous studies coincide with Altman. Although he does not mention any evidence, in February, Epoch AI researchers precisely They published a study trying to estimate the energy consumption of chatgpt. In their conclusions they indicated that on average a chatgpt consult Previous report of the researcher Alex de Vries. Since then, of course, many things have happened. Too pessimistic. And as they commented on the study of Epoch AI, the difference comes from the fact that the models are today much more efficient than in 2023, when VRies conducted their study. So is the hardware in which these models are executed, and that estimate was also used a “especially pessimistic” approach. In Openai’s study they also threw an especially pessimistic estimate and pointed out that “most of the requests (A chatgpt) are much cheaper (energetically).” More studies. Another independent study published by Andy Masey in January 2025 reached a similar conclusion and claimed that “using Chatgpt is not bad for the environment.” It was based on EPRI data May 2024 that also estimated a high consumption of 2.9 Wh by chatgpt consultation. Estimated water consumption In data centersfrom A SUNBIRD studyit was also very modest compared to other online activities. Water consumption in data centers for various online activities. Source: Andy Masley. Fifte. Precisely the data of water consumption was another striking in that estimate of Sam Altman. According to him, a chatgpt consultation barely consumed 0.32 ml of water, “a quinceava part of a teaspoon.” The figure suggests that the water needed to refrigerate data centers that process these requests is much less than what was thought only one year ago. And training, what? These estimates focus on the AI ​​inference section, that is, our use of chatgpt that receives a consultation and processes it inferring (generating) a text result. Although Altman does not clarify it, he does not seem to include here the energy and water cost of training AI models, which is very high and makes thousands of Gpus They work at full power For months, with the consequent water expense in data centers to refrigerate all those components that dissipate high heat amounts. As I pointed out The researcher Ethan Mollick, GPT-4 probably used more than 50 GW to be trained, enough to give energy to 5,500 homes in a year. We continue without definitive data. Altman’s claims are as always striking, but the lack of clear evidence makes it difficult to believe these data. Other recent studies are more useful when it comes to reflecting this increasingly lower cost both in energy and water from the use of AI, but there are no accepted standards or a consensus on the true impact of energy and water consumption when using chatgpt or other AI models. Image | Lukáš Lehotský | Village Global In Xataka | The light price is again negative: it is a sign that the system needs a redesign

If the problem is too difficult, they give up immediately

by

Machines do not think, that’s an illusion. We do not say it, a group of Apple researchers who have just published a revealing study entitled precisely (‘The illusion of thinking‘). In it these experts have analyzed the performance of several AI models With the ability to “reason”and their conclusions are striking … and worrying. Puzzles for the “reason”. The normal thing when evaluating the ability of an AI model is to use benchmarks with programming or mathematics tests, for example. Instead, Apple created several Tests based on logical puzzles that were totally new and that therefore could not be part of the training of these models. Claude Thinking, Deepseek-R1 and O3-mini participated in the evaluation. Models that crash. In their tests They checked Like all these reasoning models, they ended up starring Bruces against a wall when they faced complex problems. In those cases, the accuracy of these models fell resorted to 0%. It was also not matched that you granted more resources to these models when trying to solve those problems. If they were of some difficulty, they could not with them. They get tired of thinking. In fact, something curious happened. As the problems became more complicated, these models began to think no more, but less. They used less tokens to solve them and riddled before they could use unlimited resources. Not with help. Apple researchers even tried to give the models An exact algorithm that guided the models to find the solution step by step. And here, another capital surprise: none of the models managed to solve problems despite having those guided solutions. They could not follow instructions consistently. These graphs show the differences between models that do not reason (Deepseek-V3) with those who do (deepseek-r1) in low complexity (yellow), medium (blue) and high (red) problems. There are only advantages for “reasoning” in medium difficulty problems. In the high models they simply collapse. Source: Apple. Three types of problems. In their evaluation they divided the problems to be solved in three classes and verified if the reasoning models really contributed something to the traditional models that do not “reason.” Low complexity problems: reasoning models effectively surpassed those who did not have that reasoning capacity. Of course, they often think too much to solve these simple problems. Average complexity problems: there was still some advantage over conventional models, but not too much. High complexity problems: All models ended up starring these problems. Thinking, nothing. According to these researchers, the reason for this failure when reasoning in complex problems is simple. These models do not “reason” at alland all they do is use advanced patterns recognition techniques to solve problems. That does not work with complex problems, and there the foundations of these models are completely falling apart. Given these problems, if a model is given clear instructions and more resources should improve and be able to try to solve them, but this study demonstrates otherwise. Far from AGI. What these results suggest is that the expectation that these models have generated is undeserved: the current reasoning models simply fail to move from a certain barrier by adding data or computing. Some pointed to how reasoning models could be a possible way Towards the search for the AGIbut the conclusions of this study reveal that in fact we are not closer to achieving models that can be considered general artificial intelligence. They do not find solutions, they memorize and copy them. In fact, the study corroborated something that others defended in the past: These models simply have knowledge, and reproduce the solution they already had memorized when they find corresponding patterns that lead to that solution. Thus, these models could solve the famous problem of the Hanoi towers From many movements because they once know the solution can be applied systematically. However, in other puzzles they failed to the few movements. Stochastic parrots. Many of the critics of the AI ​​always They have defended That the generative models, reason or not, are basically parrots that repeat what has been taught. In the case of AI they detect patterns and are able to find/predict the following word/pixel when generating text or images. The result is usually convincing, but just because they have become extremely good when detecting these patterns and responding properly and coherently. But it is not new knowledge: it is to repeat the queya. They don’t think. Other critical experts of these expectations have been alerting us to alert us for the dangers of anthropomorphism of the IAS. I explained it Subbarao Kambhampti, from the University of Arizona, which, for example, analyzed the “reasoning” process of these models and their “chain of thought”. We use verbs like “think”, when they don’t think. They do not understand what they do, and that contaminates all the assumptions we do about their capacity (or lack of it). Do not trust what the AI ​​tells you. The behavior of these models confirms what is known since Chatgpt appeared on the scene. As convincing that these models may seem – “reason” or not – the reality is that they can make serious mistakes and make mistakes, although others certainly right. In fact there are cases in which these models do surprise by their ability to solve problems: In Scientific American A group of mathematicians were overcome by an AI model that managed to solve some of the most complex mathematical problems that they failed to solve, or that took longer to solve. Image | Puzzle Guy In Xataka | Copilot, Chatgpt and GPT-4 have changed the world of programming forever. This is thought of programmers

The world wants more and more matcha tea. The problem is that Japan is not being able to give it to him

by

For centuries Matcha tea It has been one of the most recognizable symbols of Japanese culture. Now is that and something else: a drink popularized by social networks and coveted in the international market to such an extent that, in Japan itself, there are stores that have been forced to limit sales by customer. The world is thirsty for Matcha. Increasingly. And that has triggered a deep imbalance in the market that will not be easy to correct in the short term. The big question is … why? Limited purchases. The market of Matchaa popular Japanese powder tea that is made with Técha leaves And it stands out for its bright green color, faces a complicated scenario: a considerable (and growing) decoupling between the supply and demand that has even led some suppliers to limit the amount of product they sell to their customers. Recently Nikkei Asia He spoke of online stores without stock or UJI and Kyoto stores in which only a single Matcha product per day is allowed to acquire. Nothing else. “Due to the strong and continuous increase in the demand for products in recent months, the current demand has already exceeded our production capacity,” Recognize The store Marukyu Koyamaen In a message posted on its website. “This translates into an extremely low inventory of all Matcha products at this time.” If the scenario were not complex in itself, it is added that there are producers who want to prevent their historical clients, such as temples, sanctuaries and places where the drink is used for ceremonial purposes, they are left short. A figure: 4,176 tons. The figures always help to better understand trends. And that of the Matcha boom is no exception. Although throughout the last decades the consumption of green tea and Matcha has been losing bellows in Japan, its success among foreign consumers has triggered the production of the ground TE. If in 2010 the country produced 1,471 tonsin 2023, according to the data of the Ministry of Japanese Agriculture (Maff), that volume had shot up to the 4,176 tons. Almost triple. The Japan Times Precise In addition, in the last five years, the sector has experienced a key change, orient more and more towards the international market: today more than half of the Japanese matcha ends up exporting. In general Nikkei Calculate That last year the flow of green tea exports reached 8,798 tons, ten times more than a couple of decades ago. Of that volume, powder tea, especially Matcha, represented more than half. The reflection in prices. The increase in demand has not long to move to prices, which draw an upward curve. After years down, the value of the tancha began to rebound after the pandemic and It has been consolidating little by little. In early May the Japanese newspaper The Asashi Shimbun revealed That, in the first auction of the year of green tea held at the Ja Zen-Noh Kyoto distribution center, the kilo reached an average of 8,235 yen, almost 67 dollars and 1.7 times more than a year ago. The value far exceeds the previous record, of 2016. Looking out of Japan. Interestingly, matcha fever comes after decades in which both the consumption of powdered tea and green tea in general has been losing strength In Japan’s households, which also explains that has decreased production. The causes of their renewed boom must be sought beyond, in an international demand that According to some analysts It moves to historical maximums. Only American buyers absorbed in 2024 44% of shipments International TEA powder. Second, quite a distance, Germany and Malaysia are. There are estimates that calculate that Matcha’s world market will be around 5,000 million of dollars in 2028, a more than considerable amount if one takes into account that in 2023 it stood at 2.8 billion. “In spite He recounts in Bloomberg K. Oanh ha. What is that boom? There is not a single factor that explains the growing world thirst for Matcha. When analyzing the phenomenon, analysts usually point out a key: the visibility it has reached in networks, both for its photogenic bright green hue and by the videos and comments that extol their health benefits, a message that seems to have penetrated above all After the pandemic. To that growing interest are added the changes in the consumption of the Matcha, which has ceased to be an exclusive drink of homemade ceremonies and meetings to join bottled drinks or even accompany chocolates and ice cream. And why not grow anymore? The tancha production has grown over the last years, expanding beyond the traditional areas of Japan. In addition, their farmers have seen how new competitors from China and South Korea arrived. So … why not that mismatch between supply and demand is not remedied? The key is in the limitations that the sector itself is found to expand its crops. The country has land, but plantations are not created overnight. It takes years In being lists and tancha leaves need to pass a storage and processing phase. Especially if farmers want to wear stone mills. The sector is not alien to the shortage of collectors and the lack of generational relief in a country that dealt with a serious birth crisis. Either to A record tourism flow that further tense demand. With that backdrop, farmers also face a fear that complicates the investments in the plantations: that the Matcha boom stays alone in that, a punctual fashion. Images | Matcha & Co (UNSPLASH) and T.TSeg (Flickr) In Xataka | Japan’s economy depends more and more on a very Japanese phenomenon: fans absolutely delivered to its idols

It doesn’t give you infinite options so you don’t even think

by

Mercadona sells more choosing for you. While chains like Carrefour Abruman with up to 20,000 references, Mercadona triumphs in quota and profitability, among other reasons, thanks to their less than 10,000 carefully selected references. The Valencian chain has turned the selection and reduction of catalog into an advantage: by offering less options, eliminates decision fatigue and increases its margins. Why is it important. Mercadona has achieved A net margin that touches 3.9%well above the rest of his rivals. Applies what we could call “spotify formula” to retail: Select and screen instead of accumulating. That also allows you to get more rotation, more control and more margin. The panoramic. Eroski, Carrefour, Alcampo … bet on the variety. Mercadona bets on the selection. It operates with what we could call an “efficient assortment”, based on their own brands developed with what they call their “Totaler suppliers“ It controls so much what is manufactured, as the “how” is manufactured, such as what finally reaches the linear. Spotify avoids navigating among millions of songs with very specific and refined recommendations. Mercadona filters between thousands of products and chooses by the consumer reducing their options. That translates into less than 10,000 products per store in front of competitors that double that amount. Assorted reduction is a general trend, but Mercadona is the one who has opted the most in the last five years (-45% assorted compared to -31% of Eroski, -20% of Carrefour or -14% of Lidl), according to a Kantar study cited by Interempresas. The context. Psychology supports it. Barry Schwartz showed that having too many options can be an initial hook, but When executing they overwhelm us. Is The choice paradox. Reducing them also reduces friction and decision -making time. Less products, less decision fatigue, more satisfaction. In detail. Where you used to choose between eleven oils or six snacks, now choose between two or three. The client, if he trusts, does not have to compare. Do not spend mental energy. Just buy, and continue. Yes, but. The model also has costs. Mercadona withdraws products quite frequently if they do not fit their expectations. Some would be pleasantly popular on other surfaces, but they are not enough in the Mercadona model. It is something similar, since we talk about technology, to what happened with the iPhone Mini: they had around 5% of the annual sales share of the iPhone, something celebrated for many other manufacturers, but Not for Apple, who discontinued him After two generations. Mercadona has not invented the selection and screening, but it has refined the technique of applying it to the supermarket. Deepen. Choosing for the client can be seen as a paternalistic movement, but Mercadona has made it efficiency. When there is overabundance in the environment, well -designed shortage is power. In a world that usually rewards accumulation, Mercadona wins simplifying. In Xataka | Mercadona earns more and more money selling money, no food: the effect of interest rates on their results Outstanding image | Mercadona

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.