News

Their other news also has a lot to say

by

Android 17 It comes with a word written in large letters: Gemini. Google has taken advantage of its Android Show to show how it wants to integrate more artificial intelligence into the system, but the update is not just about that. Under that umbrella of Gemini Intelligence We also find more earthly news, the kind that may not shine as much in a presentation, but that can end up being much more noticeable on a day-to-day basis. Let’s get to know the most interesting ones, one by one. Quick Share and AirDrop One of the most practical changes in Android 17 has to do with something as simple, and as common, as sending a file to another person. Google had already taken a first step at the end of last year by making Quick Share work with AirDrop on some Pixel and Galaxybut now it wants that compatibility to stop being a rarity limited to a few models. Support will be expanded this year to Xiaomi, Honor and OnePlus, in addition to OPPO and Vivo, which were already announced. For phones that are not compatible, Google now adds an intermediate way: generate a QR code so that an iPhone user can scan it and receive the file directly in iCloud. Migration from iPhone Changing mobile phones always seems easy until messages, contacts, the eSIM and that home screen that each user has organized in their own way come into play. That’s where Android 17 wants to reduce a very well-known friction for those coming from an iPhone. Apple already added in iOS 26.3 the necessary support to transfer that data wirelessly to Androidbut the process still does not work because it requires a device compatible with Android 17. That piece will arrive this year and the rollout will begin with Pixel and Galaxy. Pause Point Pause Point is based on a gesture that we know all too well: opening an app almost without thinking about it. Google’s idea is not to block access, but to place a short pause before entering the applications that we have marked as distracting. Android 17 will show a ten-second timer, with suggestions to do breathing exercises or open a more productive app. It will also allow us to set how long we want to use that application in each session, and here comes the nuance: to deactivate Pause Point we will have to restart the mobile. Security Android 17 also reinforces a less showy, but very sensitive part of the system: security. Google is preparing several small improvements that target very specific scenarios, from calls that try to impersonate banks to apps with suspicious behavior. In the first case, protection will depend on collaboration with certain entities and having their banking application installed. Additionally, the system will improve the detection of malware by analyzing signals such as SMS forwarding or running in the background, while Chrome will check APK downloads for known threats. In case of theft, Google will also allow you to activate biometric protection remotely by marking the phone as lost. On-screen reactions Screen Reactions, which we can understand as reactions on the screen, looks directly at an increasingly common use of the mobile phone: recording ourselves while we comment on what appears on the screen. Android 17 will allow you to capture the video from the front camera and the content we are viewing at the same time, whether photos, videos, web pages or other elements. The person will appear cut out over that content, with a process that Google describes as something that can be done in a few touches. New emojis There are new developments that do not change how the mobile works, but they do change how we feel it every time we write. Emojis fall into that category. Google has redesigned Android’s 4,000 emojis to give them a slightly more three-dimensional look, with more depth and detail than the flatter, more cartoonish versions they replace. Custom widgets Here Android 17 returns to the territory of Gemini Intelligence, but with an idea that is quite easy to understand: create custom widgets without having to design them by hand. Create My Widget will allow you to build widgets for your home screen using natural language instructions. The company gives very specific examples, such as a meal planner that recommends protein-rich recipes, a weather widget designed for cyclists that prioritizes wind and rain, or a view of upcoming concerts in a nearby venue. rambler Rambler is another of Gemini Intelligence’s functions, but here the AI ​​is applied to a very specific situation: dictating a message and making the result not look like a raw transcription. This tool will work in real time to remove fillers, correct errors, and make text more concise. In a demo, Google showed how it could turn a spoken request into a shopping list and even understand a subsequent correction, such as ordering bananas and then discarding them. You will also be able to switch between multiple languages ​​within the same message. Task automation Another block of Gemini Intelligence looks at automations. Task Automation is already available for food delivery and transportation apps on the latest high-end models from Samsung and Google, and will now expand its support to more services. The company offers examples such as ordering groceries from a list saved in notes or preparing a travel itinerary from a photograph. Chrome auto browse will also arrive on Android at the end of June and Gemini in Autofill to complete forms more quickly. Material 3 Expressive Google also wants this intelligence to be noticeable in the interface itself, not just in specific functions. In the information it has shared about Gemini Intelligence, the company explains that the new visual language is based on Material 3 Expressive and that it not only seeks to be more attractive, but also more functional. The animations will have a purpose: to reduce distractions and help maintain focus on the task at hand. Availability The fine print is … Read more

It’s bad news for Google

by

If the question is which AI makes better images, the general answer would be Google’s Nano Banana 2. And if we talk about preparing reports rigorously, we would probably say that Claude is the one who takes the lead. But in the AI ​​race, just as important as being the best is appearing to be the best. And above all, make money with your model. And if the arrival of artificial intelligence to the labor market has felt like an earthquake in the shape of more or less related layoffs, barriers to entry to junior profiles and have to work more Against all odds, the reality is that in recent months the scenario of which AI is the favorite of companies has taken a turn. Visual Capitalist has published a graph that monitors month by month from January 2023 to March 2026 what percentage of US companies pay for each provider’s models. To prepare it, they used anonymized spending data from more than 50,000 companies on the platform. ramptaking only paid subscriptions, so free use is left out. The result is a clear picture of consolidation: the market is shrinking towards very few players at breakneck speed. The graph marks a clear winner from the start: OpenAI is the most widespread payment AI provider among US companies, reaching a share of 35.2% in March of this year. Just behind is Anthropic with a share of 30.6%. You have to look down a lot to find the others: Google, xAI and the rest of the providers are below 5%. But the most important thing when looking at the photo is not who the leader is but the trend: Anthropic’s is a meteoric rise. What AI model are companies paying for? The market closes and It only has room for two: OpenAI and Anthropic together account for nearly 66% of the AI ​​business payment market in the United States, meaning that two out of every three dollars that companies spend on AI models go to these two companies. The rest share the crumbs. This type of concentration is the fish that bites its tail: leading companies have more customers, more usage data and more resources to improve their products, so their pursuers have it increasingly difficult, although it is true that Google has muscle for a while. What AI models are companies paying for? Visual Capitalist with Ramp data January 2025 is a key date in the graph: OpenAI was present in 16.8% of companies and Anthropic barely had 4.1%, slightly below Google’s 4.2% share. In 14 months Anthropic has multiplied its presence sevenfold, while OpenAI has doubled it and Google has 4.3%. The takeoff coincides with the launch of Claude Code in February 2025, its scheduling assistant that became general availability in May of that year, and accelerates with the arrival of Cowork in January 2026, its workflow platform. That Claude be a rocket The graph has several explanations. Yes, it’s a good AI model, but Anthropic has been able to build concrete tools around that model that companies use every day and that make it difficult to switch vendors. According to Sacra estimatesas of October 2025, Anthropic had more than 300,000 business clients that represented approximately 80% of its revenue, which shows that those at Amodei were clear about their strategy from the beginning: their niche is the company and not so much the ordinary user. Google has been oscillating between 3 and 4.5% in business share for three years, a marginal advance compared to the budding duopoly and the investment made. Elon Musk’s xAI has gone from zero to 1.9% in March 2026, which means appearing on the map, but still very far from the competition. But the case of Google is the one that truly baffles: It has cutting-edge technology, one of the most powerful cloud infrastructures in the world and access to an amount of data like never before, but it doesn’t get companies to open their wallets. Everything indicates that the problem is in how it has packaged its products: dispersed among too many brands and platforms, which creates real confusion for the business customer. In Xataka | If the question is which of the big tech companies is winning the AI ​​race, the answer is: none In Xataka | The US’s problem in the AI ​​and humanoid race is not China: it is all of Asia and it is greatly disadvantaged Cover | Visual Capitalist

The PC market is mortally wounded because of RAM. Excellent news for Apple’s plans

by

If there was something missing from Apple’s catalog, it was undoubtedly the cheap MacBook. The non-Pro MacBook died a long time ago, the last attempt at a MacBook without a surname did not work and that role of “affordable” laptop fell into the Macbook Air. That laptop was still missing to stand up to the 800-900 euro market that Windows dominated at will and it turns out that Apple had the answer at home: the iPhone. Its processor, rather. Because that’s what he is macbook neo: the guts of a iPhone 16 Pro in a laptop chassis. In our analysis We lowered a bit what was being said about the MacBook neo, but pointing out that it was not only a very interesting device for a wide range of users, but a blow to the PC market. This is something that Apple does not want to miss and it seems that they have bent MacBook neo orders. However, they now face the “neo dilemma.” Stop or pay more, the neo dilemma To no one’s surprise, The MacBook neo worked like a charm in its first week. 699 euros for a perfect laptop for students, or for those of us who want a second computer, is an option that is difficult to reject. Because there are cheap laptops, but not with these battery features, system speed and, above all, build quality. For find something similar in Windows You have to go to more expensive models. In the midst of a memory crisis, furthermore, those 699 euros for the basic version seemed even more appealing. And it seems like Apple expected it to do well in the market, but maybe not so well. Tim Culpan is a former Bloomberg reporter, based in Taiwan and has a very interesting newsletter. Most importantly, you have some sources at the heart of the factories that produce components for these equipment. On your speaker, Blame point that Apple had planned a total shipment of between five and six million MacBook neo. Tim Cook described the reception of the laptop as “a demand through the roof”, showing himself very satisfied with its performance, and Apple was at a time when it had to take a decision to ensure the future of the device. The reason is that this laptop uses A18 Pro chips… different. They are the processors of the iPhone 16 Probut they were not suitable for the high standards of the iPhone. In this case, it implies that instead of six GPU cores, they had five. This happens with many other processors that are renamed or derived from more affordable products. They had a lot due to leftover shipments and they converted them into the guts of the laptop. These processors were practically “free” for Apple, but now Culpan points out that those in Cupertino had to decide whether to let the inventory run out or ask TSMC to manufacture a new batch. They have chosen the second. in a new publicationCulpan claims that Apple now aims to have a base of 10 million unitsdouble that initial forecast. But of course, ordering TSMC to manufacture a new batch of A18 Pro would mean having to pay a significantly higher price to build the laptop. This would greatly narrow the profit margin they have per unit sold. Although Apple to be TSMC’s second customerthe Taiwanese foundry does not work for free, obviously. A few days ago, Tim Cook pointed out to investors that Apple had been able to avoid the first wave of the RAM crisis due to the amount of stock accumulated, but that is over. After loading memory options both from Mac Studio as of Mac Miniit is evident that not even Apple is untouchable. Here, Culpan points to two scenarios. One is to eliminate the basic option of 256 GB of memory, which costs 699 euros, leaving only the 512 GB option for 799 euros. It would be the move they have already made with other products. The second letter is raise the price of both optionsbut giving some extra to “compensate”, such as extended free storage in the cloud for a period of time. We have already seen this strategy in the PC segment. The problem is that it doesn’t just increase the memory. Aluminum is also increasing and, no matter how little it increases, anything that increases the cost of a manufactured unit is something that will have an impact on the sales price. And there is another question. Since the MacBook neo was being manufactured with those A18 Pros that were not the best, when ordering a new batch you enter a scenario in which it is possible that the new MacBook neo are “better” than the ones we had until now. Simply because they have all six GPU cores intact. TSMC is not going to make them limited on purpose. Apple has the option of software limit one of the GPU coresbut in the end that is the least of the company’s problems at the moment. All components, including processors, have increased in price since the initial order a few months ago. If we are seeing something in the industry, it is that, in case it was not already clear, It is the user who ‘eats’ the problems either due to price increases or due to the impossibility of acquiring products because they simply do not exist. And something that we are also observing is that Apple is in that “neo dilemma” because they are seeing that the consequences of launching a product with an attractive price and a good value on a daily basis translates into they take it away like hot cakes. And all this within the context of the brutal component crisis that we are experiencing. In Xataka | Tim Cook optimized factories and processes, John Ternus builds things: what we can expect from the “new Apple”

50 years later, the Soviet fire of the “Gates to Hell” is going out. And it’s not good news

by

In 1971, in the heart of the Karakum Desert, a group of Soviet engineers observed how the ground was sinking under his feet after a failed drilling. What came next was not an immediate evacuation or closure of the area, but rather an improvised decision that, according to who witnessed itseemed like a quick solution to a specific problem. That choice, taken almost as another technical procedure, would end up having consequences that no one at that time was able to anticipate. The eternal fire goes out. During more than half a centurythe Darvaza crater has burned relentlessly in the middle of the desert, becoming an almost permanent image of inexhaustible fire that seemed to defy any natural logic. However, the most recent data show a clear change: the intensity of the flames has fallen drastically in recent years, losing more than 7% of its strength. What for decades was a constant spectacle begins to weaken, altering the perception of a phenomenon that many considered eternal. The origin between legend and Soviet heritage. The birth of the crater is still shrouded in all kinds of stories and uncertainty, although the most widespread and feasible version points to the accident. during Soviet drilling in search of gas in the sixties or seventies. According to this theory, the ground collapsed when it reached a pocket of natural gas and the engineers they decided to set fire to the site to prevent the release of toxic gases, convinced that it would be extinguished in a short time. Thus, what was going to last weeks lasted for decadesfed by an underground network of gas that never stopped flowing, giving rise to one of the best-known anomalies of the energy legacy of the former Soviet Union. From remote curiosity to global icon. Over time, the crater went from being a geological oddity to becoming a almost mythical destiny for travelers and explorers, despite the difficulties in accessing Turkmenistan. Its image, a gigantic burning cavity in the middle of nowhere, has fueled so much adventure tourism like internal propagandato the point of being used by country leaders as a symbol of power or control. The experience of approaching the edge and feeling the direct heat of the fire has reinforced its reputation as a unique place in the world. The attempt at control and doubts about its decline. For its part, the Turkmen government has years trying to control emissions from the crater, and attributes part of the recent weakening to new drilling nearby plants intended to extract gas. However, the independent analyzes They suggest that the loss of intensity could have begun before these interventions, which opens the door to natural causes that are not yet fully understood. This nuance introduces a key and dangerous uncertainty: it is not clear whether the end of the phenomenon responds to human action or to a change in the geological system itself. The unexpected twist: less fire does not mean less problem. Yes, because although At first glance, the reduction of the flames could seem like good news from an environmental point of view, the reality it is more complex. Fire acts as a mechanism that transforms methane (much more powerful as a greenhouse gas) into carbon dioxide, reducing its impact in the short term. If the flames subside, more methane could be released directly into the atmospherewhich would make progressive shutdown a potentially bigger problem. A fragile balance that is still active. Despite its weakening, the crater remains activewith visible flames and constant emissions that remind us that the phenomenon has not disappeared. The huge amount of gas accumulated underground suggests that the fire will not be completely extinguished in the short term, maintaining that strange balance between natural spectacle, industrial legacy and environmental problem. Thus, half a century later, the symbol of eternal fire begins to change, although its disappearance does not necessarily imply a more favorable end for the rest of the planet. Image | Stefan Krasowski, Tormod Sandtorv In Xataka | China’s first pipeline network is 4,000 years old and something revolutionary: it was built without the need for kings or nobles In Xataka | About to close, this remote mine in the Polar Circle has found a 2 billion-year-old yellow diamond that weighs 158 carats

Mistral has a new AI model. The good news is that it is absolutely European; the bad one, which is absolutely mediocre

by

The French startup Mistral has just launched Mistral Medium 3.5an open-weight AI model that is the great European exponent in an industry absolutely dominated by China—which competes directly with this type of projects—and by the US. And if this is the best they can do, it seems Europe has a problem. Mediocre. This is a “dense” model with 128 billion parameters and a context window of 256,000 tokens. While models with Mixture-of-Experts (MoE) architecture only activate a subset of the total parameters to achieve enviable efficiency and capacity, Mistral activates them all. That makes it much less efficient, but theoretically it should make its performance promising. And that’s the problem. Which it is not. Benchmarks. Pedro Domingos, professor of deep learning at the University of Washington, he expressed it very well: “Mainstream AI companies brag about how their model is much better in benchmarks. Soo Mistral brags about how their model is much worse.” It is true that the models with which it is compared are larger in total number of parameters, but as we will see later, even taking that into account, they are cheaper and theoretically more efficient thanks to the use of that MoE architecture in many of them. The model, however, unifies the previous catalog and follows the market trend of being able to establish the desired level of reasoning (reasoning_effort) as a parameter. Bad results. And he is somewhat right: Mistral does not seem to have problems showing the results of various benchmarks in which it performs poorly, but it also performs poorly with models that are by no means the most recent or powerful on the market. Thus, it is compared with Claude Sonnet 4.5/4.6, with Kimi K2.5, with GLM-5.1 or with Qwen 3.5 397B. In almost all cases (except GLM 5.1) there are already more recent and powerful versions of all of them. Not so far from local models. In fact Medium 3.5 scored 77.6% in SWE-Bench Verified, a programming test in which Qwen3.6-27b It reaches 72.4% with a fundamental difference: you can run it “for free” (with the appropriate hardware, and you paying the electricity bill) with a relatively affordable machine. More expensive (and somewhat more restrictive). If we use it via API, Mistral Medium 3.5 costs $1.50 per million input tokens and $7.5 per million output tokens. GLM-5.1 costs 1.4/4.4 respectively, and Kimi K2.5 costs 0.5/2.8 respectively. Its recent successor, Kimi K2.6, costs 0.95/4, and it is significantly better than Mistral being cheaper. There is a curious fact: Mistral uses a “modified MIT license” instead of the traditional Apache 2.0, and indicates that this model can be used commercially or non-commercially except for “high-income” companies. Chasing Anthropic. In addition to the model itself, the company has presented the so-called remote scheduling agents using Mistral Vibe CLI to, for example, send pull requests to GitHub in an automated way. It also has the so-called “Work Mode” in LeChat, allowing multi-step tasks to be managed autonomously. These are tools clearly intended to strengthen Mistral’s role as a base for scheduling agents, which is the path that has worked fantastically for Anthropic. Your advantage: being European. The only great strength of this model is that it has been developed by a European startup, and that gives it clear visibility at a time when many EU countries they talk about digital sovereignty. It is the only Western model that seems to want to compete with China in the field of open weight models, which is good news, but the truth is that in terms of performance it does not seem that the Mistral Medium 3.5 is going to perform competitively. The geopolitical security network. That, together with the fact that it costs more than its competitors, makes the decision to use it difficult unless for those who prioritize clearly that European origin. That is Mistral’s ace in the hole, and they are taking advantage of perfectly. The company has recently obtained financing to create data centers in Europe, and is nourished and fed by this new obsession with minimize dependency of North American Big Tech. In Xataka | The CEO of Mistral sends a message to Europe: the end of being the technological vassal of the United States

While the news says that it is dying between solar panels and expropriations, the data says the opposite

by

Solar panels that destroy olive trees, massive expropriations in Jaén, warnings that Spain will have to import oil… If we pay attention to the last few months of news about the world of olives, the conclusions are clear: these are bad times for the olive grove. And yet, the data does not confirm this. In fact, as they point out from Datadistathe surface area of ​​this crop has not stopped growing in the last 10 years. The olive grove does not stop growing. With the only exception of the small decline in 2022 (0.08% already recovered in 2023), the hectares of olive groves have grown every year. However, that does not mean that there is no problem. Almost the opposite. The olive grove grows, but it does so in a profoundly unequal way: irrigated land gains ground over dry land, the super-intensive olive tree in a hedge extends over land previously dedicated to cereal or cotton, and investment funds are concentrated in areas with more water. In this sense, the story is not about the disappearance of the olive tree. It’s about changing so much and so fast that it will soon be unrecognizable. What the data says. Apparently, the data is clear. According to provisional data from the Survey on Crop Areas and Yields (ESYRCE) 2025 From the Ministry of Agriculture, Fisheries and Food (MAPA), the olive grove area in Spain reached 2,873,396 hectares, 1.63% more than in 2024 and 5% more than in 2015. It’s just that if we look closely, that data tells a curious (and sometimes counterintuitive) story. For example: the olive tree is already the largest irrigated area in the country. And why does this change occur? Above all, because the irrigated olive grove is safer than the dry one. If it were possible, the entire Spanish olive grove would switch to irrigation regime overnight. Therefore, the interesting thing is to stop and think about why the accelerated change is occurring now. According to a February 2026 report from Datadistathe explanation has a first and last name: investment funds. In the last decade, these funds have gone from 45 to 1000, investment in “Iberian agribusiness” has tripled and this is converting many hectares into super-intensive olive trees (and abandoning the traditional one). And the situation feeds on itself. The growth of the super-intensive irrigated olive grove cushions the volatility of supply and, therefore, contains price spikes. That is, the sector becomes more attractive to investors. This is precisely what ensures the future of olive oil. Even if it is at the cost of changing it completely. Image | Vasilis Caravitis In Xataka | The very high prices of olive oil are just one symptom. The real problem is a sector on the way to disaster

The RAM crisis is very good news for someone. That someone is Samsung

by

The great supply crisis in 2026 is starring memoirs. Samsung, SK Hynix and Micron control 90% of global DRAM production, and can currently only cover about 60% of projected demand. This is bad news for consumers, and excellent news for giants that cannot keep up with selling memory. Tell Samsung. Samsung Electronics has published its financial report corresponding to the first quarter of 2026. The company recorded revenue of 133.9 trillion won and, so that we understand each other better, this is its all-time quarterly high, with a 43% increase compared to the previous quarter. Memories, memories, memories. The figure is even more surprising if we look closely at the Device Solutions Division, in which its memory business is located. It recorded a sales increase of 86% compared to the previous quarter, with a historical record in operating profit. Samsung itself details that this boom comes from the hand of much higher demand and, to no one’s surprise, a sales price that has increased in the industry. It is not something isolated. Sales related to memories and semiconductors will continue with strong demand throughout the second quarter, Samsung predicts. The company wants to continue capitalizing on demand for GPUs, CPUs and DRAMexpecting advances in agentic AI to continue accelerating demand growth. Why is it important. Samsung’s results are not only good news for the company’s shareholders: they are a reflection of a change in the industry that is here to stay. The RAM crisis will change forever the price of the products we buywill make companies that have never participated in the manufacture of memories have to start considering doing so, like teslaand positions manufacturers like Samsung in a position of power that they have not had for years. The new Samsung. Samsung has always been relevant in semiconductors and memories, but currently this division accounts for 94% of the company’s total operating profit. Virtually every won Samsung earns comes from its device solutions business (RAM and chips). And what about mobile phones?. Although Samsung’s near future will be led by a single division, the company gives enough clues about its future in a territory that touches the average user very closely: mobile phones. Its DX division (in which smartphones are found) grew 19%, with more sales and more profit compared to the previous quarter. Samsung expects a slight drop in revenue next quarter, although it will continue to focus on three clear pillars: high end, folding and series A. In Xataka | There is a company that has grown 3,000% in the stock market, even beating the performance of Nvidia: Sandisk

DeepSeek V4 is here. It’s good news for efficiency and bad news for the myth

by

DeepSeek has published its V4 model under MIT license, with notable improvements in code and architecture designed for Chinese chips. It has also admitted, in its own technical report, that it is three to six months behind leading Western models. For a laboratory that A little over a year ago the global narrative of AI changedthat is much more than a nuance. Why is it important. DeepSeek became a symbol in January 2025. Its moment shook the markets, questioned the logic of the American technology stock market and convinced half the world that China could compete head-to-head on the frontier of AI, at a fraction of the cost. It’s not that V4 destroys that story, but it does complicate it a bit. China’s most important laboratory in AI arrives with a model that its own engineers describe as a step, not a leap. The context. V4 has taken longer than expected to arrive. According to sector sources collected for 36KrDeepSeek suffered a serious training failure in mid-2025 while trying to migrate its infrastructure from NVIDIA to Huawei’s Ascend chips. Internal opinions on technical direction were not aligned, and the founder, Liang Wenfengimposed conditions that were difficult to execute. The result: months of delay and a model that, furthermore, is still not multimodal, postponed due to lack of computing capacity and cash. Between the lines. The most interesting thing about V4 is in its architecture. The model introduces TileLang, a domain-specific language that allows low-level code to be decoupled from CUDA (the NVIDIA standard) and compile it for different chips. It also incorporates MegaMoE, a kernel designed to reduce latency in expert parallelism that already runs on Ascend hardware. But V4 training has continued using NVIDIA GPUs. Independence is, for the moment, more of an aspiration than an accomplished fact. turning point. While DeepSeek looked inward, the Chinese market has been reorganizing itself without it: Doubaofrom ByteDance, has become the most downloaded chatbot in China. MiniMax and Z.ai They have gone public. Alibaba has achieved great adoption thanks to vertical applications. DeepSeek never wanted to build a consumer product, and the market hasn’t waited for it. The internal bill has also arrived: the laboratory has lost key talent to Tencent, ByteDance and Xiaomi in practically all areas. Liang Wenfeng refused to give up 20% to an unidentified large investor. And now, for the first time, DeepSeek opens an external funding round. Main loser? The narrative of open source Chinese as a real alternative to the Western closed model has taken a hit. A Qwen employee has told 36Kr that “the golden age of nonprofit AI development is over.” The big question. It’s whether DeepSeek can regain lost ground. That depends largely on Huawei, whose Ascend 950 promises to scale well with V4, but 750,000 units are equivalent, adjusted for quality, to a week of American production. The gap is not closed with ingenious architectures. It is closed with silicon. In Xataka | Companies around the world face an irresolvable dilemma: either they are with China or with the US, with both it is no longer possible Featured image | Solen Feyissa

AI enters the era of CPUs. To no one’s surprise, this is bad news for the consumer.

by

The current state of data centers is redefining the production lines of the main technological players. What seemed like a specific crisis in the price of RAM and SSDs has ended up becoming a tsunami that is sweeping away products and the consumer market. The data centers They need the same components as consumers and the rest of the industry, and there was one component that seemed safe: the processors. It’s over. Danger! During his presentation of results In the first quarter of 2026, Intel gave a worrying piece of information: the ratio of CPUs and GPUs in data centers could soon reach 1:1. So far, we’ve talked about memory and GPUs as the primary hardware in data centers, but there needs to be a CPU running the show, and currently, there was one CPU for every eight GPUs. However, things are starting to change due to agentic AI. How to know the components of your PC (RAM, Graphics, CPU…) and the state they are in Training is still important, but now we seem to be entering the era of inference, and that’s where CPUs excel. David Zinsner, Intel’s chief financial officer, described in the call with investors that the CPU/GPU ratio had already gone from 1:8 to 1:4 (one CPU for every four GPUs), but that this agentic AI was exploding the memory of the CPUs, approaching the aforementioned 1:1. Change of course. I think you can now anticipate where the shots are going to go. As we read in Tom’s Hardwarethis action has caused a reaction: that Intel begins to move its production lines to begin to reduce its capacity in consumer products and increase the output of Xeon processors, which are indicated for servers and continuous work. All because, currently, delivery times for server CPUs are about six months away and they cannot allow AMD, which also has its products for servers, to beat them in that race. Consequences. Price increases. Without palliatives. Xeon CPU prices are estimated to have risen in March by 10% to 20% due to that shortage, but consumer CPUs have also increased in price by 5% to 10%. It is not going to stop there, as another 10% price increase is expected for the second half of the year. In fact, it is the same as with mobile phones, with Intel pointing out that the consumer PC market will decrease by double digits this year. Here it happens as in the rest of the segments: manufacturers are raising prices for everyone (hyperscalers and consumers), but they allocate their production to data centers because they are the ones that will buy the most volume from them. Intel doesn’t care if there is no CPU for users because it is not the bulk of its current market, but what it cannot afford is to have hyperscalers for more than six months expectingespecially if data centers are expected to have to mount more CPUs in the short term. The objective: the great foundry. Intel has been in the doldrums for a few years. In the consumer segment, AMD’s Ryzens have beaten them to the punch, the ARC GPUs have not finished coming together and things were not going well. A series of poor results led the United States government to invest $2 billion to ‘rescue’ the company with one goal: to make it the largest American foundry. Because, even if things did not go well, they are still one of the few companies in the world with the capacity to create chips, like TSMC or Samsung. They have some of the best machines on the market and that government bailout soon began to bear fruit with clients such as Apple and Nvidia. During the presentation of results A few days ago, Intel declared a net loss of $3.7 billion, but something happened nonetheless: the stock rose 20%. The reason is that investors are not looking at Intel’s present, but rather its future, and the changes applied in recent months seem to be going in the right direction. They are not the only ones. This change of direction and production lines is not exclusive to Intel. We have seen it in other companies, but it is true that here they directly advocate leaving out the consumer to prioritize the large customer: Big Tech. Something similar happened with Samsung a few days ago, when it was reported that the company had begun to move your LPDDR4 memory production lines to LPDDR5. This type of memory is better, but also more expensive, which will cause devices that previously mounted LPDDR4 memory (low-end miniPC or entry- and mid-range mobile phones) to have to go directly for LPDDR5 memory that is faster, but also more expensive. In the end, the translation is the same as always: as users, we are going to have to tighten our belts and hold on with the devices we have for a while longer. How long? Until 2027 if you ask some, 2028 if you ask others or if that… 2030. Images | Intel In Xataka | There is no energy for so many data centers and the consequence is clear: half of those planned for 2026 in the US are in danger

There are people reselling tickets to the World Cup final for 2.3 million dollars. Great news for FIFA

by

It is still too early to know if the 2026 World Cup will be a success, a failure or will be added without pain or glory to the extensive chronicle of FIFA. What we can say at this point is that enjoying the tournament in situ it won’t come cheap. Especially if you aspire to see the final, which will be played on July 19 at MetLife Stadium in New Jersey. The cost of your tickets it takes months embroiled in controversy, but the debate has soured after some positions have come to light resale market for the price of a 200 m2 apartment in the center of Madrid. All with the veiled pleasure of FIFA. What has happened? That although there is still more than a month until the opening match, the World Cup in North America (to be played between Mexico, Canada and the USA) is already earning the dubious honor of being the most expensive of history. The fans screamed in the sky last decemberwhen the first tickets were launched, but the rates that were offered then seem like a ‘bargain’ when compared to those that are now being achieved in the purchase and sale market. In this secondary trade, channeled through FIFA, there are passes that are offered for the same What does a 200 m2 apartment in Madrid cost? Does it sell so expensive? Yes. The news has advanced it the Associated Press (AP) agency, but it comes with taking a look at the buying and selling platform of tickets hosted on the FIFA website to verify it. If we look for available passes for the final on July 19, we will see that there are people reselling them for more than two million dollars. To be precise, there are at least four seats on sale in the lower stand (behind the goal) for a whopping $2,299,998.85. Not all tickets cost the same, but resale prices are generally not affordable for everyone. The cheapest seats, 3rd category, are offered for $10,900. If you want a position with better views and more comfort, you can add a few thousand more to that figure and purchase higher category passes for $16,100, $33,800, $43,200 or even $207,000. The prize goes to the entries of 2.3 million and 991,500, which is what a seller asks for seats located in the front area. On Wednesday FIFA itself put up for sale a new block of tickets on its direct sales platform, where it was possible to find seats for the final by $10,990. Who controls these rates? Direct sale tickets are launched by FIFA itself, but things change when we talk about the secondary market. There, in the so-called “Resale/Exchange Market” the federation does not control prices, although it does take a considerable part of the business. For each transaction you pocket a commission which is divided into two parts. One, 15%, is applied to whoever purchases the ticket. Another, of the same value, is borne by whoever detaches from the entry for resale. As they explain in Guardianthat means that if one of the tickets that cost 2.3 million is finally sold, FIFA would deposit $690,000 into its account. But… How is that possible? In other editions of the World Cup, the resale price of tickets was limited at face value, but this time FIFA has changed the approach. The reason? First, adapt to the market of the host countries, especially the United States, which is the one will host more games of the tournament. Secondly, FIFA hopes that by channeling the buying and selling itself, the use of portals such as StubHub will be discouraged. “FIFA has established a ticketing and secondary market model that reflects standard ticket market practices for major sporting and entertainment events in host countries,” alleges in a statement cited by the Associated Press. “Resale facilitation fees are aligned with industry standards in the North American sports and entertainment sectors.” Is it an isolated controversy? The controversy has now arisen due to the prices that are being reached in resale, but the truth is that the cost of the tickets has been a matter of discussion since the first phase of sale, activated in December 2025. The focus has been on both the prices themselves and the system applied by FIFA in the sale, the ‘variable pricing’similar to dynamic rates. Consumer organizations like the OCU have already raised their voices for that same reason. For reference, in December tickets for the final were already being sold for prices ranging from 4,185 and 8,680 dollars. And this despite the initial promise to offer them for 60 dollars in the group stage. “They only exist as ridiculous green splotches on the edge of seating maps, little more than mirages of inclusion,” ironizes Bryan Armen, from Guardian. Does it only happen with tickets? No. The tickets are so expensive because, FIFA allegesare one of their main sources of income. However, passes to matches are not the only thing that is valued at a gold price. In recent days, another controversy has arisen around the celebration of the World Cup in the US that revolves around something that has little to do with sport: public transportation. The New Jersey rail operator has decided that those who want to buy round-trip tickets to travel from Manhattan to MetLife and watch the July 19 final there will have to pay 150 dollars. It is almost 11 times more than what the same service costs on a normal day, when it is around $12.9. Images | FIFA and Wikipedia In Xataka | Mexico City is already noticing the economic effect of the World Cup: it is losing homes and gaining Airbnb apartments

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.