precision

China brought humanoid robots to the country’s biggest television show: it made them practice kung-fu with millimeter precision

by

Every year, hundreds of millions of people in China sit in front of the television to watch the Spring Festival Gala, recognized by the Guinness Book of Records as the most watched annual program on the planet. It is not only a music and dance show, but also a showcase where the country decides what image it wants to project of itself. In this scenario of maximum visibility, the presence of humanoid robots ceases to be a simple technological curiosity and begins to function as a public declaration about the place that innovation occupies in the national narrative. What happened there was not just an artistic number, but a clear clue as to where the Asian giant is looking when it thinks about its technological future. Kung fu, choreography and coordination. To present their robots to millions of spectators, the organizers turned to a deeply recognizable symbol: martial arts. In the CCTV broadcast available on YouTube We can see robots using traditional weapons such as swords and nunchucks, as well as doing tricks and jumping from trampolines, always in sequences shared with human performers. The choice of kung fu provided more than just visual spectacularity, it can also be interpreted as a close way of reading technological advancement within a tradition known to the public. The magnitude of the event. The Spring Festival Gala has been broadcast since 1983 and is an inseparable part of the New Year celebration in hundreds of millions of homes. Reuters also describes it as an event comparable, in terms of media scale, to the American Super Bowl, capable of concentrating popular culture, political message and industrial ambition in a single night. What appears in that scenario entertains and, at the same time, projects a message and indicates priorities. A gateway for the industry. Behind the staging there were specific names and a visible strategy. They participated in the gala companies known in the West such as Unitree, but others less known such as MagicLab, Galbot and Noetix. The immediate precedent helps to understand the moment: Unitree’s robot performance in the previous edition went viral and, in a way, brought this technology closer to the general public. So the idea of ​​betting on a similar show again is reasonable. From the stage to the factory. The public display of these systems fits with a line of industrial policy that places robotics and AI at the center of the next Chinese manufacturing stage. In recent years we have seen how the Asian giant has invested heavily in this sector. According to OmdiaChina accounted for around 90% of the nearly 13,000 humanoid robots shipped worldwide last year, a global shipping metric that does not go unnoticed. Morgan Stanley also projects that Chinese sales could exceed 28,000 units this year, which would point to a notable expansion phase. In Xataka There are people sharing their court cases with AI. The problem is when a judge considers the conversations as evidence In the end, what was seen on that stage went beyond well-executed choreography. Behind each movement appeared a country narrative that combines technological ambition, industrial policy and cultural projection in the same television image. The question is no longer whether these robots can perform in front of millions of people, but rather how much their presence will grow in the coming years and into what spaces of daily life they will end up integrating. For now, its massive presence is destined for this type of spectacle. Images | CCTV In Xataka | While technology companies dispense with juniors to replace them with AI, IBM is doing the opposite: catching bargains (function() { window._JS_MODULES = window._JS_MODULES || {}; var headElement = document.getElementsByTagName(‘head’)(0); if (_JS_MODULES.instagram) { var instagramScript = document.createElement(‘script’); instagramScript.src=”https://platform.instagram.com/en_US/embeds.js”; instagramScript.async = true; instagramScript.defer = true; headElement.appendChild(instagramScript); – The news China brought humanoid robots to the country’s biggest television show: it made them practice kung-fu with millimeter precision was originally published in Xataka by Javier Marquez .

Today we know how to solve crimes with scientific precision. And we owe it all to a lady with a dollhouse

by

You imagine the scene and smile. It is perfect to become the seed of a script that ends up winning an Oscar. You, along with dozens of other male criminology students in the room, are greeted by a woman in her 50s who looks like an endearing grandmother. You go to the next room and in it you find a doll house which, they order you, you have 90 minutes to examine. “They sent me here because they told me it would be a training visit,” you protest internally. When you look at the recreation you begin to perceive the macabre aura of the event. The mental image of what a toy house must be is broken inside you when you see that the cloth and porcelain romper is a headless prostitute thrown on the bathroom of a room to which over the years, you notice, the same squalor that has characterized the life of its guest has been attached. There’s more: one thing catches your attention, just one of hundreds. Some lines drawn in chalk on the miniature ironing board in the corner of the room. They mark the price that the utensil must have had when they bought it several years ago. You look at the whole and the level of detail The entire room is such that you start to get dizzy from the puzzle you have to solve. But there is no room to be stunned. The strange grandmother at the beginning had already warned you: you have 90 minutes and there is not a single one to lose. Frances Glessner Lee, dollhouses and the true origin of CSI Frances Glessner Lee, whom we know today as “the mother of forensic science,” he didn’t have it easy to get to where he did. If it were not for a concatenation of circumstances, it is likely that this police branch would have lost one of its most valuable and, of course, curious pedagogical milestones that we have known. Born in Chicago in 1878, Lee was the daughter of John Jacob Glessner, owner of the successful International Harvester company. Motivated by her childhood and teenage readings of Sherlock Holmes, she longed to dedicate herself to the exciting world of homicide investigation. At the end of the 19th century, the typical thing for society ladies was not to go to Harvard and then dedicate themselves to solving crimes, but get married and start a family. They forced this on him when he was 20 years old. He divorced and waited for his father and brother to die so he could inherit the family fortune and finally be able to make his own decisions. In all that time his concerns never went away. He studied criminology in Boston, He donated part of his inheritance to Harvard to open a brand new forensic medicine department there and he got to work. At 52 years old. Throughout her life, Lee founded the Harvard Medical School and served as an advocate for absolute rationalization in police investigation. among many other thingsbut the great work of his life was another: his Limited studies of unexplained deathsa series of 19 dioramas or small miniature doll houses that represented complex crime scenarios that would be analyzed by future students of Criminology or Forensic Investigation. As a socialite, Lee used her money and social skills to make your way into the world of men and convince them to participate in your proposal. The cottage seminar by day, opulent parties at the Ritz Carlton by night. Fourteenth State Police Homicide Investigation Seminar, November 17-22, 1952, at the Department of Forensic Medicine, Harvard Medical School. Read to the right. Harvard Medical Library and Francis A. Countway Library of Medicine. Complete photograph of your death: every object around you counts If something stands out all the biographies of Lee It is his unhealthy passion for detail. The passages about his works are an endless collection of nouns and adjectives. The cans of food in the cupboards, the fogged-up mirrors, the half-peeled potatoes, the overflowing ashtrays, the unmade beds, the open oven faucets, the pieces of wood under the nails, the violet stains observable on the subject’s face. The backyards and fire escapes hidden from view of investigators and that Mrs. Lee ordered from specialized carpenters just so that no element of the room imagined in her mind escaped her control. Only one thing escaped him: it is not possible to distinguish rigor mortis in a doll. For all this only Lee can know how many months or years was able to dedicate to each of the jewels in his gloomy collection, which cost as a result of 3,000 or 4,000 dollars of those of that time piece by piece and that involved both work and love: all the textile elements that we see in the dioramas were made by her herself. The Bounded Studies are not only a pedagogical tool, but also a theoretical proposal on the tangible, material aspect that surrounds the reality of human death. As if each object, each frayed blanket and each photograph on the front page of the newspaper that has the figure of a murdered man at its feet, also became part of the same concatenation of events that led to his death. Pure chaos theory applied to forensic architecture. As we have learned later, the houses became so complicated that many students were not able to find the appropriate solution as to how the hypothetical crime had occurred, but rather that they were simply left unable to give a single answer. A pedagogical legacy still to be resolved Of the 18 dioramas that Mrs. Lee made, we only know the answer to 13 of them. Over the years, no one has been able to solve five of these simulations that are still considered some of the most arduous homicide scenes in history. Its creator took the secret of those five rooms to the grave: she agreed if the participant proposed the correct … Read more

Gemini 3 promises more quality and precision than ever in its responses. The question is whether we will really notice the difference

by

Google has announced the launch of Gemini 3its new artificial intelligence model. in the company They claim it is their most advanced reasoning model because it is “designed to understand depth and nuance.” Gemini 3 will also be available as standard as part of AI Mode in the renewed Google search engine (in this case and for the moment, only in the US). It is the first time that Google offers the benefits of its AI model from day one in the search engine, but it also reaches the Gemini app and the developers who work with AI Studio and Vertex AI. Behind him success of Gemini 2.5 Pro and Flashthe new version arrives in 30 new languages, including Catalan, Basque and Galicianand as we say you can start testing today in the United States… or outside of there via a VPN. Gemini 3 promises. At least in the tests Google highlights how the model’s behavior has been outstanding in various synthetic tests. Thus, Gemini 3 leads the LMArena classification with 1,501 points—the first to overcome the 1,500-point barrier. According to Google, the Gemini 3’s test results put it ahead of all its competitors in virtually all scenarios. In fact, he manages to reason “at the level of a PhD” according to the tests of Humanity’s Last Exam (exceeds 37.5% of the test without tools) and GPQA Diamond (91.9%). It also makes spectacular progress in mathematics, as demonstrated by the 23.4% on the MathArena Apex test: GPT 5.1 scores 1.0% and Claude Sonnet 4.5 1.6% on the same test, for example. The model also wants to be more direct: his answers are more “concise (…) and he prefers to offer valuable information instead of resorting to clichés and flattery. Tells you what you need to hear, not just what you want to hear“. Gemini 3’s ‘Deep Think’ mode goes even further in tests: in Humanity’s Last Exam it achieves 41.0%, but it also in the demanding ARC-AGI 2 It achieves 45.1% (with code execution), which also demonstrates progress in abstract reasoning and visual understanding. Gemini 3 explains the world to you in a simple way The model has a context window of up to one million tokens, which allows it to be used, for example, to analyze huge repositories of code or text and then work on that data. Its multimodal support allows you to analyze all types of information. For example, Gemini 3 can decipher and translate handwritten recipes in different languages ​​to create a family cookbook that you can share. Or analyze your pickleball games (we assume the same thing happens in other sports) and identify areas where you can improve and generate a training plan. Or scrutinize the data from a research paper and from it generate code for an interactive guide that helps us better understand those studies. In fact, integration with Google Search is an especially important part of Gemini 3, which being “embedded” in AI Mode It has the capacity to generate interactive visual elements (widgets, calculators, simulations) in real time. At Google they want the search to be more interactive than ever, and that will mean that sometimes the answers will not be just text, but rather a small interactive webapp that allows us to better understand the answer. Programming (and agents) to power The other crucial element of the model is its capacity in the area of ​​programming. Its results in tests of this type are once again outstanding, and for example it tops the WebDev Arena leaderboard with a score of 1,487 ELO. The model now behaves much more powerfully in the visual part. It also scores 54.2% on Terminal-Bench 2.0, which evaluates a model’s ability to use tools and operate a computer through a terminal. Additionally, it far outperforms 2.5 Pro in SWE-bench Verified (76.2%), a benchmark test that measures the effectiveness of scheduling agents. These Gemini 3 programming capabilities are intended to be used in a new agent development platform called Google Antigravity. The developer experience is using a “conventional” AI integrated development environment (IDE), but your agents can have access to the editor, terminal, and browser. That means these agents can autonomously plan and execute complex software tasks and validate their own code, making it easier for human developers to review and audit that code than ever before. The real challenge of the most recent models On paper Gemini 3 is postulated as a model that can really make a difference compared to its competitors. The test results and Gemini’s own trajectory make us think that the behavior of this model will indeed be remarkable. However The question is whether we will really notice the difference. In recent months we have seen how other AI companies have launched new models, but the impact for a large majority of users has been discreet: the previous models already performed really well, and although the new ones undoubtedly provide improvements, for many consultations these improvements allow us to perceive that jump in performance. Here we see two ways for Google to effectively demonstrate the capabilities of these models. The first opportunity for Gemini 3 will likely be in the area of ​​programming, and it will be these professionals who will likely be able to get the most out of those additional capabilities. But for the rest of the users, it will be that new AI Mode and the Gemini app that will have to make us notice those features. We are intrigued by this ability to respond with small interactive elements —graphics, widgets—, and perhaps with them we will really discover this new capacity of this chatbot. In Xataka | Let’s say goodbye to Google Assistant a decade later. Google has begun to delete its code to leave only one option: Gemini

To build an “artificial sun” we need to be able to move the weight of ten elephants with millimeter precision. This is what China has just done

by

In Chinese mythology, Kuafu was a giant who challenged the gods when trying to catch the sun to give light and heat to their people. Centuries later, China re -pursues that same ambition, but now with avant -garde science: to create a “Artificial sun” that provides clean and unlimited energy. And in that way, the engineers have just presented a new protagonist worthy of legend: a colossal robot. The arm for fusion. The Asian giant has developed a remote manipulation platform for future fusion reactors. It is a system with three robotic arms, whose main manipulator can raise up to 60 tons – the weight of ten African elephants – with a millimeter accuracy, According to South China Morning Post. Meanwhile, the two secondary arms stand out for even more extreme precision: ± 0.01 millimeters, which makes it the most advanced remote management system in the field of fusion. Closer to the “artificial sun.” The objective of this whole project is to achieve stable nuclear fusion, that almost inexhaustible energy that mimics the process that occurs in the sun’s core. In fact, China has been breaking records for years in its East experimental reactor, which this year has achieved Maintain a confined plasma for 1,066 seconds, a world record that exceeds 403 seconds Realized in 2023. But for this energy to become commercial, it is necessary to resolve a major challenge: maintenance. The internal components of a reactor, such as coating or the diving, are constantly damaged by heat, radiation and magnetic fields. And this is where this new robot comes into play: no human being could work in these extreme conditions. The in -depth project. The robot is part of the craft (Comprehensive Research Facity for Fusion Technology), an installation in Hefei, Anhui, nicknamed “Kuafu” in honor of the mythical giant. More than 300 scientists and engineers participate in this project, According to SCMPunder the supervision of the Institute of Plasma Physics of the Chinese Academy of Sciences. “We have developed a machine capable of meeting extremely demanding requirements by overcoming obstacles in materials, sensors and control,” explained Pan HongtaoInstitute researcher. The idea is to use craft as a test bank to develop and validate key fusion technologies, including those that will be applied in the future Chinese experimental fusion reactor (CFETR) and in the International Iter project In France. Ready to go into action? For now, we are not talking about an operational robot in a reactor, but of an experimental platform. According to China Dailythe system has already exceeded the evaluation of experts and will serve as an engineering verification platform to ensure that, when reactors enter into operation, remote maintenance is safe and precise. Craft, where it is housed, plans to be completed in the late 2025. Beyond fusion. Although the immediate objective is to maintain fusion reactors, technology is not limited to that field. According to CGTNthe advances achieved in this robot could also be applied in inspection of nuclear plants, aerospace industry, operations with heavy machinery or even emergency rescues. A global career for the artificial sun. The Kuafu robot does not arise in a vacuum. Other countries also develop remote maintenance systems, although with much lower capabilities. The most advanced arm of Mitsubishi Heavy Industries (Japan) supports only 2 tons. In contrast, the Chinese robot can manipulate 30 times higher loads. At the international level, the Iter project in France – in which 35 countries participate – will have a system Able to handle up to 45 tons. The new Chinese system already exceeds it in load capacity, although both remain development platforms rather than operating systems. The road map is clear but slow: Chinese experts They calculate that they are still 30 to 50 years to see commercial fusion reactors. And the European Union, with its Eurofusion projectdoes not expect to start testing with plasma until the end of this year. Forecasts The Chinese commitment to nuclear fusion advances with firm steps. The development of a robot capable of lifting 60 tons with surgical precision is not a simple engineering achievement: it is an essential piece for someday fusion reactors to maintain and function stable. Humanity He has been trying to replicate the energy of the sun on earth. With advances like this, China shows that it is determined to be the protagonist in that race. Decades may be missing to see fusion plants in operation, but every step we bring us a little more to that utopia to capture the sun. Image | Freepik Xataka | The largest nuclear fusion project on the planet has survived the setbacks. This is the date on which Iter should be ready

A new generation of robots promises precision and efficiency. It also opens the door to cyberspage risks

by

The movement of the robotic arm seems impeccable: each turn, each clamp, each displacement occurs with the accuracy of a metronome. However, while these actions convey confidence in a hospital or a factory, another story is drawn in the background. The commands, although encrypted, let rhythms and pauses In the traffic that travels through the network. These patterns, invisible to the naked eye, in many cases allow to deduce the task executed. The same accuracy that we applaud in the robot can become a trace for external observers. Over the last years the interest in collaborative robots has shot himself. Hospitals are used as surgical assistants for their ability to make fine movements without fatigue, and in factories they have become allies for repetitive or risk tasks. Not only do productivity improve, they also reduce accidents by replacing the operator in hostile environments. The connectivity that drives its expansion, however, is also the one that can expose them to new vulnerability scenarios, According to a study in the University of Waterlooin Canada. Precision that dazzles in hospitals, a trail that can be interpreted The investigation did not focus on real -time robots by means Preprogrammed scripts. These systems receive an orders sequence and execute them with minimal human intervention, which reduces direct supervision and expands automation possibilities. At the same time, the way in which these systems structure high -level commands generates regular traffic patterns, and that regularity opens opportunities for analysis. The work was designed with a very concrete scenario: a passive attacker, someone who only observes the network traffic between the robot and its controller without deciphering it. The experiment was carried out with an arm Kinova Gen3a light robot usually used in research environments. The controller executed preprogrammed commands and the communications were protected with TLS encryption. With this assembly, the researchers registered 200 network traces corresponding to four different actions, looking for a varied and representative set. The authors began by converting network catches into temporary signals: instead of looking at the content of the packages, they analyzed when each one was sent and with what separation. Those time series were treated as acoustic signs, which allowed to apply classic signal processing techniques, such as correlation and convolution, which seek similarities and patterns in pulses and rhythms. From this transformation they trained a classifier that, in the closed environment of the test, assigned an action to each trace. The experiment used 200 traces on four actions and showed that, even with activated TLS, temporary subpatrones were detectable. Kinova Gen3 The trials showed that the method worked with remarkable efficiency: in most tests the system was able to identify the robot action with a success level close to 97%. This means that, although the orders travel encrypted by TLS, the observation of the intervals and the cadence can allow to rebuild what task it is executed. In a hospital, that could give clues about the nature of an intervention; In a factory, on the production sequence. A complete deduction is not always achieved, but the finding shows that the encryption alone is not enough. Although orders travel encrypted by TLS, the observation of the intervals and cadence can allow to rebuild what task it is executed. The finding acquires relevance to extrapolating it to real environments. In health, an attacker could identify details of a surgical intervention without the need to access the medical history, only observing the robot communication flows. In the industry, the patterns themselves could reveal assembly steps or Characteristics of a patented process. It is not an isolated failure of a specific model, but an alert signal on how connectivity multiplies the exposure. Each connected robot becomes a possible observation point. The researchers did not limit themselves to pointing out the problem: they also explored possible defenses. One of them is to modify the timing of the robot programming interface, so that the commands do not follow such a regular and predictable pattern. Another is to apply package filling and timing manipulation to hide the real rhythms. These measures could reduce the inference capacity of an attacker, although with a cost: lower network efficiency and, in some cases, more latency in the execution of the robot. Technological innovation always advances in parallel to the need to protect it. Cobots exemplify that balance: they promise efficiency and new forms of work, but also force rethink defense measures. It is not about stopping its adoption, but about doing it with a conscious look of the risks. Security and development are not opposite paths; They must travel together if you want the future of robotics to be sustainable and reliable. Images | Kinova Robotics (1, 2) | Freepik In Xataka | Alibaba is becoming the Ai Open Source sponator. Your family of Qwen models is putting the market above

If Spain wants to save the countryside, you already know what you have to do: precision agriculture

by

Increase in costs, water and climatic problems, European regulations, ecological demands, pests, retirement, lack of labor, industrial crisis, international competition … The Spanish field accumulates the problems and, instead, the solutions are scarce. Above all, because the only medium term is clear, but carrying it out is devilishly difficult: precision agriculture. What is precision agriculture? As I remembered a couple of days ago, Manuel F. HerradorProfessor at the UDC Civil Engineering School, he always says that, in the future, experts will be horrified how much of our way of building was summed up in two words: Brute Force. They are the same words that perfectly define agriculture of the last decades. ‘Precision agriculture’ is an elegant way to say that we are trying to change it. And it is not easy. Especially since the main problem of standard agriculture is that it is an activity that continues to develop in open systems. That is, environments “almost unpredictable and full of risk in which Two nights of frost mean losses of up to 9 million euros“ Efforts to improve agriculture have historically focused on “break that opening“. Since 1850, when in the Netherlands they began to use greenhouses at the productive level, the key strategy has been to create physically closed environments. In Spain, we know a lot about this. The problem is that the greenhouse strategy is not enough. Is there alternative? The other option is to inform the ecosystems in which we work. Today, thanks to agronometric satellites, We have capable technology to monitor in real time pests, diseases or water distribution. We can do it, in fact, with a 30 centimeter resolution (and with drones we could improve it to almost ridiculous limits). We can also manage crops efficiently and personalized. And the promise of that management is enormous: “A PWC study for the Business Association for Plant Protection (AEPLA) estimates that boosting precision agriculture could increase agricultural production by more than 54,000 million euros up to 2050”. We talk about a productivity eight times higher than the current one. So much? Well, as I say, that is “the promise.” As Roberto Ruiz saidBBVA agricultural business, it is “better controlling supplies, to better control fertilizers, phytosanitary products, etc.” The potential is huge; The real impact depends on many factors. But it is important to keep in mind that we are not talking about a futuristic. We carry more than A decade with solutions of this type and the results are beginning to arrive. And why don’t we move towards that world? That is the big question, right? Spain is an agroganadero giant and has been going through more than important problems for years … Why is the evolution of the field so slow? Nobody see that time runs out? And the truth is that they see it. But the situation is very uncertain and the lack of generational relief atenza a good part of the sector in a stagnation that is difficult to leave. From there arise phenomena like Lto financing the field: phenomena capable of solving many of these issues, but that drag your own (and huge) problems. Spain plays its agricultural future in the coming years. The good news is that we know what the way is. Image | Jed Owen In Xataka | Murcia is being filled with “ghost agricultural companies”: they arrive, exhaust the resources of the territory and go

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.