usatoday24

The “richest treasure in America” ​​is not a pyramid, but the imposing collection of a Mixteca princess of 800 years ago

by

Mexico is an inexhaustible source of pre -Hispanic treasures findings. We continue discovering vestiges of the past practically every week and it is something that He has shot with the Mayan train project and with the latest technology tools. Thanks to the latter, we have been able to find hidden pyramids, Underground areas And it is something that can give a push in the Study of old structures. However, there are pieces that, although we discover a few years ago, are still of great value. An example is the treasure of the tomb 7 of Monte Albán, which some continue to consider “the richest treasure in America.” Not only are hundred pieces, but they have a really enviable state of conservation. The Mixteca Treasure of Tomb 7 In January 1932, Dr. Alfonso Caso discovered at the archaeological site of Tomb 7 of Monte Albán one of the biggest pre -Hispanic treasures in history. At the time of its finding, a first restoration was carried out, but care was also carried out both in 1944 and a few years ago, with the objective not only of studying the pieces, but of restoring those that were precise and carrying out an adequate conservation process. In total, it is a set of more than 230 pieces made with sacred materials in the pre -Hispanic era, such as gold, silver, rock glass and precious stones such as turquoise, obsidian and green stone. The state of conservation of many of these objects is imposing. There are also less striking objects such as those made from bone or an alabaster pot. But … Where do these luxury objects come from? INAH researchers They offered context to the pieces, affirming that their origins date from the union of a lord of the dynasty of Zaachila and a Mixtec lady at some point between the 1,200 and 1,400 after Christ. Skull decorated with turquoise mosaics and shells Rock glass glass Locked bone with turquoise embedding Specifically, from the National Institute of Anthropology and History they comment that it was a union between “Mr. 5 Flor, of the Zaachila dynasty, with the princess of the Mixteco kingdom of Teoxacualco, Mrs. 4 Rabbit ‘Quetzal’”. The princess would have used the tomb 7 as a sanctuary to deposit the Tnaniwrappers in which the relics of the ancestors such as effigies of great value and other luxurious objects were kept. And, judging by the photos of the articles, the princess’s treasure was not small, precisely. Now, when these articles were found, they did not look as well as in the photographs that accompany these lines. According to INAH, it was necessary and meticulous restoration process so that we can admire the details of each of these pieces. Mask that represents the god Xipe Totec made in molten and soldier gold A pot with support manufactured in alabaster Bezote that represents the head of a jade stone foisan. The posterior plug is gold sheet This includes the recovery of the properties of its color, the polished finish of some objects and the reflective power of others. In whatever it may be, work was worth work, since the collection has an imposing presence and, beyond the photographs that INAH himself Share On its website (where we can see descriptions of each object), you can see live in room III of the Museum of Cultures of Oaxaca. And, in addition to admiring the goldsmithing of each piece, it is interesting to note that this treasure is important because it demonstrates not only how certain materials already worked on what they gave importance, but some of them were achieved thanks to the exchange and trade networks between different Mesoamerican sites. This is something extremely significant that, little by little, we are knowing and that tells us about the cultural wealth of some sites. A recent example is the study of Channels from Mexico City That, precisely, it was an imposing cultural center thanks to those commercial activities. Images | INAH In Xataka | The Maya played football. And now we know that under the courts they buried a hallucinogenic surprise *An earlier version of this article was published in August 2024

Depseek has just launched something that will bitter the day and the US chips companies: it is called Depseek-V3.1

by

There was a day when Depseek surprised half the world by demonstrating that you could go far with less. Today returns with V3.1 And a message that does not go unnoticed: the model has prepared for the next Chinese chip batch. We are not talking about an automatic market overturn, but about a concrete bet that points in an awkward direction for Nvidia and company. If that technical tune with the Chinese hardware It translates into performance, conversation about who feeds AI in China is going to sound very different. According to the company’s own noteV3.1 opens a hybrid inference in the purest style GPT-5: the same system with two routes, Think (deep reasoning) and Non-Think (Quick response), Sygons from your website and app. The formulation is clear: “Hybrid Inference: Think & Non-Think, a model, two models.” The company also underlines that the version Think “Reach answers in less time” than your predecessor. That is, not only do pesos change, the inference modes that are already in service also change. The phrase that frames everything: an FP8 “thought for national chips” In a comment set in his latest publication in Wechat, Depseek writes: “EU8M0 FP8 is for the next generation of national chips.” That is the point that tense the rope: it suggests that the company has adjusted the data format, apparently a FP8 which label as EU8M0, to the next wave of Chinese processors. Bloomberg andReuters collect that message And they synthesize it: v3.1 is “personalized to function with next -generation AI chips Chinese. ”In other words, optimization oriented to the local ecosystem. The original comment in Chino (left) and its Spanish translation with Google Translate (right) FP8 is an 8 -bit format that weighs half that FP16/BF16. With native support, it allows more yield per cycle and less memory, provided that the climb is well calibrated. In the official Model Card of Hugging Face It is read that Depseek-V3.1 “has trained using the EU8M0 FP8 scale” format, which indicates that it is not only a packaging of weights, but that training and execution have been expressly adapted to that precision. The delicate part, and it is convenient to be prudent, is that everything points to a chips remittance that will be displayed in the future, since they can take advantage of this scheme natively. So is this bad news for Nvidia? The data of the fiscal year that expired on January 26 indicates that China represented approximately 13% of the company’s revenues led by Jensen Huang. If part of the computation of AI in China Classic duo muta NVIDIA GPU + CUDA ECOSYSTEM To domestic solutions that work with the UE8M0 FP8 format and give good results (presumably chips ascend of Huawei), the demand for Western solutions could be eroded over time. China meant about 13% of Nvidia’s income in the last fiscal year All this happens on the US export controls board: restrictions that sought to stop China’s access to leading chips and that have also accelerated their commitment to self -sufficiency. This year the Trump Administration rehabilitated with conditions the export of H20a chip cut for China. Since then, the state of the H20 has been oscillating: among permits, Chinese regulatory pressures and Nvidia plans to present Blackwell -based alternatives. The background message is that the framework is political and changing, and any route that allows China to depend less on these windows becomes strategic value. You have to remember another fact that helps to calibrate expectations. According to Financial TimesDeepseek tried to train his future R2 model with Huawei chips ascend to official instances and found persistent technical problems. He ended up returning to Nvidia for training, while he was still working on the Compatibility for inference. That episode does not invalidate the current strategy, but puts the bar: to completely migrate its processes is not simple, it requires, among other things, months of engineering. V3.1, therefore, it must be read as iteration. Now the company states that it has prepared its model for the next Chinese chips. Matherena models scores And here we have another interesting fact. Matharenaa platform linked to Zurich Federal Polytechnic School which evaluates models in real and recent mathematical competitions, places GPT-5 as a leader, with 90% in final response tests, already deepseek-v3.1 (Think) something behind although among the best models of the moment. This helps to locate the context: V3.1 Compete above. Images | Xataka with Gemini 2.5 | Matharen and Deepseek screen catches In Xataka | Tiktok stole the searches, Depseek beat them in Ia: Baidu discovers that being “the Chinese google” is no longer enough

Hundreds of Microsoft employees have shared their salaries and we already know how it is paid in AI or azure

by

It is one of the most anticipated times of the year for Microsoft employees: where they are received Double increases and payments. And as it is almost a tradition in great technological ones, transparency (sought or not) emerges. Hundreds of employees of the Redmond company They are sharing their compensation data on a spreadsheet Collaborative that already has more than 850 entries and has been analyzed by Business Insider. An anonymous and voluntary document. This initiative has emerged from the employees themselves so that everyone knows the salary reality of their company. And also to see The big salaries in Silicon Valley. This is filled without anyone knowing what an employee charges from the company. And what is included in the spreadsheet offers a fascinating look at Microsoft’s salary structure, detailing the base salary, bonuses and actions of shares. And he does it at a crucial moment for the company. The war for talent in the era of AI. These data are not used in a vacuum. Microsoft is in the middle of An aggressive campaign to attract and retain the best talent in artificial intelligencean effort that has promoted its income to record figures. The company does not hesitate to get the checkbook, reaching try to sign rivals talent as a goal With millionaire offers. Although, on the other hand, beat record in layoffs while getting large amounts of benefits. Something a bit incongruous. In fact, Microsoft has applied new salary guidelines this year with specific exceptions to be able to compete more aggressively in the AI ​​market. This occurs in a paradoxical context: despite having fired thousands of workers, the company insists that its global workforce will remain stable, indicating that the hiring plans in strategic areas remain at full speed. The key salaries exposed. Although compensation in the technological sector depends largely on the actions, the base salary is a key indicator to compare the strategic importance of the different teams within a company. Based on the analysis of almost 300 software engineers in the United States, within the spreadsheet these are the average base salaries in one of Microsoft’s most important divisions: Cloud + AI: 204,135 dollars Commerce + Ecosystems: 191,597 dollars Security: 189,285 dollars Azure: 176,035 dollars Experiences and Devices (includes Windows and Surface): 175,123 dollars Microsoft AI: 170,456 dollars Xbox: 168,831 dollars Koreai: 167,759 dollars Record extract made by Microsoft employees with their salaries. A fascinating look, but with nuances. Now, you have to take this data cautiously, since they are not official or exhaustive at all, since it represents a very small sample Of the total of 219,000 employees that Microsoft has right now. In addition, it is completely voluntary and anonymous information, so it could be adulterated. On the other hand, it is likely that employees with the highest wages or higher positions are less likely to participate in this type of collaborative documents, so the figures could be lower than the real average of the entire company. An interesting fact that underlines the source consulted is that all these salary ranges self -declared by current employees tend to be lower than those that appear in the official guides of Microsoft itself for new hiring. This could confirm that bias or simply reflect the aggressive policy of the company to attract new talent with very competitive initial offers. Great salaries in the technological environment. It is known that if you are a good programmer or technological expert the salary will not be bad. In the case of Spain it is known that Salaries can reach 170,000 eurosbut outside our borders, the engineers specialized in AI They are becoming the best paid in history. Therefore, there are already many future options for those who want to enter this world, being interface designers who will surely be better paid. But above all the departure to Europe is almost mandatory to have a good salary. Images | Boliviainteligent In Xataka | Microsoft has secured its future with a very simple AI strategy: copy what Amazon did with the web

We had always thought that after 35 our performance collapsed. Is the opposite

by

Working in the technology sector involves facing the “Curse of the 35“That idea, halfway between myth and the harsh reality, that when you reach that age Your career stagnates, The hiring opportunities collapse And you become a “obsolete” piece in front of younger colleagues. This perception is fed by data such as The average age of employees in technological giants: 29 years on Facebook and LinkedIn, 30 on Google and 33 in Microsoft. Science comes to doubt it. A recent one Scientific study published by academics From the South China University of Technology and the University of New South Wales has rigorously analyzed if this barrier has some empirical justification. And the conclusion is resounding: no, you don’t have it. The myth collapses: there is no abrupt fall of performance. To get to this conclusion, the researchers used A known methodology as ”. In simple terms, they sought a sudden and vertical fall in the performance of professionals, just at the time they turn 35. Analyzed a huge amount of data from an important Professional Networking Platformtogether with patent records and Salaries of thousands of engineers and scientists. The result was unequivocal: there is no ‘significant discontinuity’ in the innovative performance or economic productivity of professionals when crossing the threshold of 35 years. In other words, the idea that an engineer’s ability vanishes from one day to another is simply a myth without a scientific basis. What we do have is a productivity peak. That there is no chopped drop in productivity does not mean that the capacity for innovation is constant. The study reveals what the true professional career is, and has an inverted ‘u’. According to the data, the A professional ability to innovate (measured in the number of patent applications that the investigated have made) increases rapidly during the twenties, it reaches its maximum point around 34 years, and from there it begins a soft and gradual decline. A peak of innovation that varies in the sector. Depending on the specialty you have, the peak of ‘maximum success’ is different. In the case of software and communication engineers, the peak reaches its maximum at 32, while chemical engineers achieve it at the age of 31. Those who take longer to reach this are mechanical engineers, standing at 37, where accumulated experience plays a more important role. Why does prejudice do not exist then. If the performance does not fall according to this study, the question we can ask ourselves is why that ‘myth’ arises that at 35 there is a barrier. The study points to a possible economic explanation and for this they introduced a novel metric: the performance-sailor (PIR) ratio that measures how many patents an employee produces for each euro that the company invests in its salary. Here the curve is different. It has ‘u’. The analysis shows that this ratio is decreasing until it reaches its lowest point between 38 and 39 years. This is because, in that age strip, wages can continue to rise by seniority, while patent production has already begun its soft decrease. It is the time when an employee, from a purely numerical perspective, is ‘more expensive’ for the company. After touching background, it will rebound again. But when a person is at the bottom of the ‘U’, he later returns to ascending. Experience, management capacity, mentoring and accumulated tactical knowledge become very valuable assets that, although they are not always reflected in a patent, are crucial for the long -term innovation of a company. It is good news for professionals. The professional career does not end at 35, according to this study. In fact, if you are here, you are at your best. It is the ideal stage to maximize your innovative production. Facing the 40, it is strategic to start pivoting towards management, leadership or mentoring roles, where vast experience may be better to take advantage of it. It is also a warning for companies. There are companies that can be reluctant to hire someone to exceed 35 years, but they are in a big mistake. They are losing talent that is in their performance peak. The study suggests that companies should reconsider their salary structures so that they are linked to the real performance that at ancient times. But in addition, they also advocate promoting more diverse teams with employees from different age ranges, where the most experienced can collaborate with the youngest. The unemployment problem among the older ones. This is something that in Spain is suffered in the first person. Recently, it was seen that 47% of the unemployed in Spain are over 50 years oldwith the incentive that many will not work until they retire. Therefore there are mechanisms such as Subsidy for over 52 years to give them an economic livelihood. But the reality is that the Spanish labor market faces that In 2030 40% of workers will be over 50 years old for the general aging of the population. This makes it right now back a qualified person just because he is olderNot the best idea. And more when we see that its productivity does not look so diminished. Images | Agefis Vlada Karpovich In Xataka | China has a huge youth unemployment problem. So much, that some people pretend that they work

Microsoft has put co -pilot in Excel. And you have also notified that you do not use it if you need the results to be correct

by

Artificial intelligence has finally landed In one of the most famous spreadsheets in the world: Excel. But he hasn’t done it as many imagined. Talk to the program in natural language and automatically solve everything for us? Only in part. Copilot’s shortcuts allow to summarize, classify and order databut they do not turn Excel into a agent able to take control of a project from beginning to end. In addition, the novelty comes with clear limits: the function can make mistakes. Microsoft is testing co -pilot as a native function within the Excel calculation engine. The idea is simple: write an instruction in natural language within the sheet and obtain a result that is placed directly in the grid, with the same behavior as any Excel exit. Integration allows you to continue working with defined ranges, tables and names without changing the structure of the file, and the results are updated when the data changes, without resorting to scripts or external accessories. Copilot is landing in Microsoft Excel As we see, it is a useful function, but with clear limitations. At the moment we cannot ask in natural language things such as “fuses the books of the subsidiaries, make everything to euros to the official change, eliminate duplicates and enter me a quarterly consolidated balance with comparative graphics.” Copilot, for now, plays in another land: Understand instructions in natural languageYes, but it does it within Excel and always about the data you have on the sheet itself. Accepted that Copilot is not an automatic pilot, it is worth exploring what it is practical. Microsoft points to four daily fronts that are resolved without abandoning the sheet: Rain of ideas without leaving the sheet. Ask for a list of concepts, titles or keywords from a brief description. If we need it, we can ask that you rewrite a text in a lighter or more formal tone and leave the result ready to polish. Summaries that go to the point. Point out a wide range and request a short text with trends, peaks and falls. This is useful when a table must be converted into an understandable paragraph for a report. Direct classification in the grid. Deliver a column with comments, tickets or survey responses. We can ask that you return labels such as “positive/neutral/negative” and a brief category. Everything would turn to new columns to continue working. Lists and tables on the fly. From a description, which proposes a table with the columns that we indicate (task, responsible, date, state). It is a basis for organizing without wasting time mocking. ⠀In the official noteCatherine Pidgeon (Excel team) offers a very clear example of what Copilot can do today. Imagine a sheet with a column full of opinions about a new coffee maker in the office. Some value flavor, others complain about noise or deposit size. Copilot can read that column and return, right next to us, a table with two fields: the feeling of each comment (positive, neutral or negative) and a category Cut that helps you group (“taste”, “noise”, “capacity”). In minutes we would go from a messy list to something analyzable. There we decided: filter for the most frequent problems, count a category or prepare a summary. And how is it used in practice? Exactly like any other Excel function: writing the sign equal to the beginning of a cell and then the name of the function. In this case, Simply put = co -pilot () and add inside the instruction in natural language or the range of data that we want to use as a context. There are no tricks or hidden menus, it is invoked as invoked a = sum () or un = search (), which makes learning minimal and experience is familiar from the first moment. It is time to put the caution on the table. Copilot works within the book itself: it does not access the web or documents of the company, and its results must be reviewed and validated before including them in reports or business decisions. Microsoft is clear about it: it is not appropriate for high -risk scenarios (with legal, regulatory or compliance implications). There are also operational limits: the function has quotas of use (up to 100 calls every ten minutes and 300 per hour). Microsoft is clear about it: it is not appropriate for high -risk scenarios. What we are seeing should not surprise us either. Artificial intelligence has advanced by giant steps, but still has a weak point: it can be wrong. These inaccuracies are usually called “hallucinations”And, although technology improves with each generation, the companies that develop it warn that It is not convenient to re -one hundred percent. It is almost a courtesy reminder that we have already normalized: every time we open a chatbot as COPILOT, Chatgpt either GeminiWe find a warning that reminds us that what we read can contain errors. Will the day come when these failures disappear? There is no clear answer. Even paying for the most advanced versions, the fastest or most expensive, absolute security does not exist. That is the reflection that also applies in Excel: Copilot is useful and opens possibilities, but never replaces the tranquility it gives for oneself that the numbers walk. Copilot does not reach everyone yet. Microsoft is gradually display They are on the beta channel, both in Windows and Mac. The company ensures that in the coming months it will also be available in the Excel Web version through the Frontier program. Nothing must be done to activate it: the function appears on the sheet when the requirements are met, and its use is completely optional. Images | Microsoft | WIRESTOCK In Xataka | If the question is whether there is an AI bubble, Sam Altman has just answered. One with which he is winning

No need to install solar panels everywhere

by

Our strategy in the solar energy deployment could be much smarter. This humble editor does not say, but an advanced computational model developed by Harvard, Rutgers and Stony Brook universities in the United States after processing five years of electricity network data. Climate profitability. That is the key to the recent research published by the magazine Science Advances. The authors have identified the regions of the country where each new solar panel offers the highest performance in terms of carbon dioxide reduction. The conclusion is overwhelming: to maximize the reduction of emissions, instead of Install solar panels Everywhere, it is better to install them in the right places. The “where” before the “how many”. Not all solar energy megawatts are the same. The impact of adding photovoltaic to the electricity grid depends strongly on the energy matrix that already exists in the region. The study makes it clear After processing with AI The generation, demand and emission data of 13 regions of the United States between 2018 and 2023. The regions where solar energy has a huge impact are those that still depend largely on very polluting fossil fuels, such as coal. In places like California, Florida, the West, Texas and the Southwest, each kilowatt-moor of solar energy that is injected into the RED directly replaces energy generated by coal or gas plants. The result is a massive and immediate reduction of CO₂. What happens in the other regions. In places like New England, the central area of ​​the United States and Tennessee, the effect is minimal. The reason? They already have a cleaner energy matrix, with a strong presence of nuclear, hydroelectric or natural gas energy (which, even if it is a fossil fuel, emits approximately half of CO₂ than coal). In these regions, add more solar energy barely moves the needle of emissions, because the energy that solar panels move was already relatively clean. In a world with limited resources, this type of optimizations can help investments in clean energy as effective as possible. But there are still patterns that can only detect AI. The contagion effect. One of the most fascinating contributions in the study is that the model was able to quantify effects that until now were difficult to measure, revealing counterintuitive dynamics in the electricity grid. For example, the study shows that installing solar panels in one region can clean the air of another. He California case It is paradigmatic: an increase of 15% in its solar capacity not only benefited the State, but reduced daily emissions in the northwest region in 913 tons and in the southwest in impressive 1,942 tons Conclusion: Investing in solar energy in the Arizona desert could be one of the most efficient ways to reduce emissions in Oregon. Image | Rawpixel In Xataka | China broke the solar panel market. Now their companies have had to say goodbye to a third of their employees

China has no rival in batteries and solar panels. It is already ready to also destroy Western pharmacists

by

China’s quota in the world photovoltaic cell market Broken 80%and for the moment only India seems to have the ability to disturb him in the long term. Its manufacturing capacity is 17 times greater than the rest of the planet togetherwhich has encouraged the Chinese administration to support the manufacture of More than 1,000 GW in N -type cell capacity As the other countries fulfill their net emission commitments. On the other hand, China is also a world leader in The production of lithium batteries. If we stick to electric cars the country led by Xi Jinping Fabrica 57% of batteries that these vehicles use. Catl and Byd are the largest lithium batteries manufacturers on the planet with A market share in 2023 34% and 16% respectively. Its absolute leadership in these two sectors is the result of a strategy that prioritizes investment in R&D and large -scale production to shoot competitiveness. Now Chinese biotechnological companies plan to apply this successful formula to medicines. China has the ability to drastically reduce the cost of medical care “We can reduce the costs of medical care and benefit more people through technological innovation and the improvement of efficiency (…) the recent achievements of China in the field of biotechnology show that it is possible to do it,” says Da Liuthe CR-CP Managing Director Life Science Fund, a risk capital investment group created by the Chinese state and the Thai chain Pokphand Group conglomerate with the purpose of investing in biotechnology companies. “We can reduce medical care costs and benefit more people through technological innovation and efficiency improvement” Chinese pharmaceuticals and biotechnology are living a “Deepseek”. This means simply that the medications they develop are every time most demanded by multinational corporations. What the latter pursue is to obtain the necessary licenses to market, produce or distribute the medications developed in China without elaborating them from scratch. Just three weeks ago the American investment bank Jefferies published a report in which he argues that Chinese companies and assets are extraordinarily competitive thanks to “their efficiency in costs, their small deadlines and the quality of the resulting product.” Interestingly, in seven of the ten main agreements carried out by the global pharmaceutical industry during the first semester of 2025, Chinese licenses were involved, According to the Pharmcube consultant. The success of the Chinese pharmaceutical industry is supported by a strategy very similar to that which has led this Asian country to lead the markets of solar panels and lithium batteries. However, China does not have it as easy as it seems. And it is that the geopolitical tensions that mainly support the countries led by Donald Trump and Xi Jinping are very difficult for Chinese biotechnological companies their international expansion through mergers with foreign companies. According to Liu What they need to transform into great multinational actors in the pharmaceutical industry is to achieve a dominant position in at least one or two fields of biotechnology. Image | Edward Jenner More information | SCMP In Xataka | The comfort of swallowing a pill against the power of an injection: the dilemma that will mark the future of the drugs to lose weight

They are Russian gas stations

by

Petroleum prices will rebound again in recent weeks. Brent rose 2.7% and the West Texas intermediate 1.1% this week, According to Reuters. Behind this volatility are the low, presumable expectations, of a peace agreement between Russia and Ukraine. This uncertainty is immediately reflected in the markets, which already discourage tougher sanctions against Moscow and a long war. In fact, while Vladimir Putin and Donald Trump were gathered in Alaska, thousands of kilometers, Ukrainian drones reached a Russian refinery. Suggesting that war is no longer freed in Donbás. Now hits where Kremlin hurts most: oil. Shortage in front of the bombings. In Russia, wholesale gasoline prices have shot at record levels. In a Financial Times report It is detailed That the most common fuel, the A95, reached 82,300 rubles per ton in the St. Petersburg Stock Exchange, 55% more than at the beginning of the year. The climb responds to a double pressure: the seasonal rebound in demand and damage to the Russian energy infrastructure caused by Ukrainian attacks with drones. The population begins to suffer consequences. The drivers are already evidencing it: long lines, empty suppliers and rationing in regions such as Zabaikalsky, Crimea and the East end. According to the Moscow TimesRussia has lost about 13% of its refining capacity since the beginning of August, after attacks that forced the closure of at least four refineries. In Vladivostok, motorists wait up to two hours to refuel. “The suppliers are covered with posters of ‘out of service’,” said a driver to the local press. Scarcity also has a historic echo. In some areas they have introduced “coupons” of fuel, a rationing mechanism that evokes the memories of the last years of the USSR. Crimea, annexed by Russia in 2014, suffers supply cuts. Sergei Aksyonov, regional chief designated by Moscow, has admitted to FT “Interruptions in some stations” and requested patience “until the end of the special military operation.” However, despite trying to contain the crisis with subsidies and prohibiting the export of gasoline on July 28, domestic demand remains without satisfying. The drone offensive. According to the countryUkraine has launched in August a systematic offensive against the Russian oil industry, hitting refineries and distribution centers to hundreds and even thousands of kilometers from the border. On August 10, Ukrainian drones reached a Lukoil installation in the Republic of Komi, more than 2,000 kilometers, a record record. The attack stopped for days the production of one of Rosneft’s key refineries in Saratov. These attacks are not new, but now they are more frequent and coordinated. The Financial Times He has stressed that unlike 2023 – when the damages were repaired quickly – in 2024 the campaign aims to leave several plants simultaneously for longer. The strategic objective, As the country explainedIt is double: show the Kremlin that it is vulnerable in its own territory and hit its main source of income, hydrocarbons, in the middle of the peace negotiations. The chaos of the drones. Kyiv’s weapon are swarms of low and long -range drones, produced by mass. As my partner has detailed in XatakaUkraine manufactured 2.2 million drones in 2024 and wants to double that figure in 2025. But the battlefield has also become an electronic chaos: up to 60 drones can fly at just a kilometer in front, interfering mutually. Sometimes, Ukrainian countermeasures themselves against Russian drones leave their devices inoperative. In others, operators connect unwittingly to enemy transmissions. Even so, the strategic message is clear: drones allow Ukraine to hit away and erode the Russian war economy. The forecasts. Analysts expect that gasoline prices in Russia will continue high at least until September, According to the Financial Times. Although a generalized national crisis is not anticipated – because the low demand after summer and part of the damaged capacity can be recovered – it is expected that the problems in remote and poorly connected regions are expected. Moscow could resort to imports from refined products from Belarus to cover part of the deficit. However, Ukraine seems determined to maintain his campaign. “There will be more,” warned the Ukrainian General Staff after an attack on the Syzran refinery in Samara. A war laboratory. Beyond the economic, the conflict reveals two long -range trends. On the one hand, the war in Ukraine has become a military innovation laboratory, where drones resistant to interference, platforms with artificial intelligence and new electronic warfters are rehearsed, As we have pointed in Xataka. These lessons learned will mark the armies of the future. On the other hand, within Russia, the appearance of rationing by coupons connects the current war with collective memories of Soviet scarcity, an uncomfortable reminder that the costs of the invasion also reach everyday life. A new front. The war in Ukraine has displaced its most visible forehead: it is no longer only in the trenches of Donbás, but in the tails of Russian gas stations. While Putin insists that Moscow maintains the military initiative, citizens face rations, queues and rising prices. History suggests that shortage hits stronger than any projectile. The tsaries learned him at the beginning of the 20th century, when popular discontent undermined his power. Today, more than a century later, Kremlin see how the internal front can become its greatest vulnerability: if drones continue to attack refineries, the price will not only pay the economy, but also the population. Image | Unspash and National Police of Ukraine Xataka | Less oil ships are scrolled, and there is only one reasonable explanation: Russia’s ghost fleet

They have burned 3.3 billion euros in developments that will never see the light

by

There are times that it is better to stop, recapitulate and, in the worst case, throw all the work in the trash. Because there are occasions in which it is better to throw the work in the trash than to move forward and continue trying into a project condemned to failure. That is what should be completed within Stellantis last year. And at the same conclusion this year has been reached. The result: 3,300 million euros lost in developments that will never see the light. In the trash. A few weeks ago, Stellantis published the results of his first half of 2025. The most striking figure was, without any doubt, The loss of 2.3 billion euros. A figure that is still much more striking if we take into account that last year, in the same period, a benefit of 5,647 million was thrown. The figure is explained by the tariffs that the company has had to face in the United States and for a descent in sales. Billing has fallen more than 12%, explained by a 6% sales fall in Europe (with some cars to review, such as Citroën C3) but, above all, 25% in the United States where they had a huge accumulated car stock In the dealers. But, above all, it is explained by money that has gone, directly, in the trash. Hydrogen. With the Presentation of the results The Cancellation of plans of Stellantis related to hydrogen. In its results account, losses of 733 million euros were attributed to the performance of this item in the first half of 2025. That figure was built with investments in Symbioa joint company with Michelin and Forvia for the production of hydrogen -moved vans. The production plant took more than 300 million euros that have not finished anywhere. However, it is estimated that stellantis losses directly related to hydrogen investments have been much higher and They reach about 1.8 billion euros. Keep in mind that Stellantis mounted, for example, a program for convert electric vans in hydrogen versions. Maserati. But hydrogen, which seemed like a future investment to take positions, seems almost the least of the problems when it is observed what is past with the Italian luxury vehicles. The Maserati MC20 is barely being sold and The company has canceled the plans To develop a completely electric version. The result is that in the first half of 2025 they have lost 552 million euros. But the problem is another. The problem is in 2024 when 1.5 billion euros went to drain. China is not the way. The big black point is that in 2024 Stellantis had the production of Electrical versions of the Maserati Quattroporte and Levant. The idea was to sell those cars, mainly, in China where the electric car is triumphing but the process was paralyzed at the last minute. The result is seen in your results account: 1.5 billion euros lost in the development of these cars. Before the Bad results of foreign companies In China, someone decided that it was better to brake and assume losses before the result was even worse than reflected. The intention was that these electrical versions They will occur in Mirafiori (Turin) So investments in personnel and machinery were already made. At least, we had to try to patch the matter. The last piece of dominoes. That obligation to do “something” with the Mirafiori plant was taken advantage of by Stellantis to try to boost the sales of Fiat 500. The electrical version has been added to modify the vehicle’s platform since the car was born as a purely electric vehicle. This last money is not explicitly reflected in Stellantis accounts but it is unquestionable that it has had to assume another stone on the road although it has been awarded to the rest of the investment items since, unlike cars such as Peugeot 3008 (born to house all kinds of technologies)the Fiat 500E was born as a car without combustion versions on the horizon. This last expense, in fact, seems to have been enough for Stellantis does not make the same mistake with the Peugeot 208 And think, from the beginning, to set up the car on a multienergy platform and, in this way, not have to turn back in the project. Even if he had not been born prior loss of 1,500 million euros with Maserati. Photo | Maserati and Stellantis In Xataka | After years of crisis, Stellantis has definitely retired from China: his joint business has entered bankruptcy

How to improve your Tinder profile using artificial intelligence

by

Let’s explain How to improve your Tinder profile wearing artificial intelligencewith several tips that you can also use in other similar appointments applications. Because we all like to cause good impression, and in this type of apps extra help can always be welcome. We are going to start by giving you several previous tips and telling you things to take into account before starting. But then, we will divide the ways in which you can help you from AI in three parts, one that is about your photographs, another about your biography and another on the chats themselves. Before starting, things to take into account Before starting with the work to improve your profile using AI, it is convenient to take into account some tips. For a start, Keep authenticity above all. It is very good to want to make your profile better, but if you do things like putting a photo taken by AI or too formal texts, you run the danger of causing more rejection than attraction. Given this, everything is about Use AI to show your best versionand not to invent another person or personality. Do not alter personal traits or present yourself with another way of being, because if you start talking to someone these things will be noticed right away. With this, you must remember that if you use AI to improve a photo with its sharpness or its light, or to cut it or make a funny assembly without the intention of going through you, it is worth. But if you modify your features, factories impossible scenarios or try to pass a false image by real, you will only generate distrust. Be careful with privacy In the event that it is something that worries you a lot. All the photos and data that you raise to an AI to modify them can be stored on the servers, so it is convenient not to upload very sensitive content. In addition, use only photos of yours so that, even if you modify them, they cannot report you for using other people’s content. Choose and improve your profile photos You can use it to help you improve your profile picture in Tinder. To start, the application has a tool called Selector Photowhich uses its own artificial intelligence and specifically trained to look in your photo reel and find those images that you consider better or more attractive. In addition to this, both Android and iOS have small artificial intelligence options in their photographic applications that will help you improve the image, Delete unwanted elementsor optimize colors. In the event that you have a photo that you consider very good, but that for some reason it has come out a little blurred or little clear, you can also Use AI to improve it. In these cases, you can use services such as Remini or others of Restoration by Ia To improve them. Maybe you can also use chatgpt to create an image from your photo, since you can get it maintaining your face proportions. However, Do not use this image as the mainbut it is simply better to use it within your photo collection to contribute a fun spark, especially if you appear doing something you like and do not pretend to be a realistic image. Finally, there are also some tools such as Photofeelerwhich are used to Give yourself a feedback on which of your photos are better. It is oriented to appointment apps, and takes several parameters such as looking at the camera, positions and general quality of the photo to tell you which of those present are better. Not bad for when you are not sure what to use. Improve your biography and your presentation Artificial intelligence offers you Several ways in which you can improve your biography of Tinder. It all depends on what you want, if you rewrite what you already have done or if you prefer that I am writing everything from scratch for you. If you have no idea where to start, you can tell your artificial intelligence assistant like Chatgpt something like this: “I want you to help me create a biography for Tinder. Go by asking me step by step everything you need of me and then write it to me.” When you do this, the AI ​​will begin to ask you a series of questions about how you want biography, but also about your personal data. And when you have written everything, then it will take into account how those of Tinder are usually and create one for you from scratch. You can also hit your biography and Ask him to improve it to Tinder. So, you will go to a rewriting process of what you already have. With this you will ensure that it is better written, and you can even ask you to write it in a specific tone, as more formal, informal, flirtatious, selfless, or whatever you want. Another of the options you have is translate your biography to another language to add it below in the event that you also want it to be in English. And if you are going to ask the AI ​​to believe it from scratch or rewrite it, you can ask you to do it in the language you want. Here, just keep in mind that it is very important Do not describe someone who is not you. It is not about creating a character with qualities or a personality that you don’t have, but knowing how to better convey how you are. Because if someone makes you matching that you are in one way and then find that you are from another, it will immediately lose interest. It is possible that in some biographies you see that four letters are included that describe personality. If you want to have the same, you can ask the AI ​​as MAKE A MBTI personality test Ask as a question. Then you will ask you a series of questions to answer, … Read more

Log In

Forgot password?

Forgot password?

Enter your account data and we will send you a link to reset your password.

Your password reset link appears to be invalid or expired.

Log in

Privacy Policy

Add to Collection

No Collections

Here you'll find all collections you've created before.