crash Archives - usatoday24

AI solves equations and chops code, but continues to crash with PDFs: the explanation shows its limits

February 27, 2026 by usatoday24

It’s probably happened to you. You upload a PDF to an artificial intelligence chatbot in the hope that it will summarize a report, extract a table or find a specific piece of information for you in a matter of seconds. And, sometimes, he succeeds. But other times, the result is disconcerting: mixed columns, footnotes embedded in the middle of the text, tables converted into an illegible block or answers that do not faithfully reflect what the document says. The paradox is evident. Systems that already demonstrate clear advances in mathematics and programming They keep stumbling upon something as everyday as a PDF. And there is more than a simple punctual failure. Change of mentality. Although for us it is a document with well-defined paragraphs, titles and tables, for the system that processes it the situation may be very different. PDF is, first and foremost, a way to visually describe how a page should be rendered. And when a chatbot like Gemini either ChatGPT If you try to work with it, you do not always access an ordered structure, but rather a set of graphical instructions that you must first reconstruct before you can respond coherently. And that difference is better understood when we look at how a PDF “saves” information. How you actually organize information. Unlike a web page, where the content follows a logical order defined in the code, a PDF can store text as independent fragments placed at specific positions on the page. Many times, the file retains coordinates and placement instructions, but not necessarily explicit relationships between one sentence and the next. This implies that the order in which the text “appears” when extracted does not always coincide with the order in which we read it. If your document includes multiple columns, tables, or overlapping elements, the system must figure out how they fit together. And that deduction is not always trivial. {“videoId”:”x9hhg44″,”autoplay”:false,”title”:”The TRUTH of AI – This is how ChatGPT 4, DALL-E or MIDJOURNEY works 🤖 🧠 ARTIFICIAL INTELLIGENCE”, “tag”:”webedia-prod”, “duration”:”1173″} What happens with HTML. On a web page, the content is organized in an explicit hierarchy– There are tags that indicate what a title is, what a paragraph is, what a table is, and how those elements relate to each other. This structure is part of the file itself and makes it easier for other systems to read, index and process it. In a PDF, as we have seen, that semantic layer may not exist or be clearly defined. Therefore, in practice, extracting information from a website tends to be a more predictable process, while doing it from a PDF is more complicated. So what about OCR? It is the first solution that comes to mind. If the problem is that the text is not well structured or even “drawn” like an image, optical character recognition should convert it into something machine readable. And in part it does. OCR has been used for decades to transform images of words into text, but converting an image to text is not the same as reconstructing the logic of the document. When there are varied elements, the system can recognize each word without knowing exactly how they fit together. The result is not a failure in reading characters, but in the organization of information. In Xataka Dario Amodei founded Anthropic because OpenAI didn’t take the risks of AI seriously. Now you are going to give in to those risks Why don’t we abandon PDF? The answer is more pragmatic than technological. As reported by The Verge citing the person responsible for the PDF Associationthe format became established precisely because it allows a document to look the same today as it would in ten or twenty years, regardless of the device or software with which it is opened. A web page can change depending on the browser, an editable sheet can be modified or overwritten, but a PDF maintains its appearance and visual integrity. That stability is precisely what lawyers, engineers, public administrations and any organization that must maintain reliable records need. The challenge is not to replace the format, but to learn to interpret it better. Images | Xataka with Nano Bana In Xataka | Three AIs clashed in ‘War Games’. 95% of them resorted to nuclear weapons and none ever surrendered (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 AI solves equations and chops code, but continues to crash with PDFs: the explanation shows its limits was originally published in Xataka by Javier Marquez .

CRASH Clock is the proof

December 15, 2025 by usatoday24

Low Earth orbit is increasingly close to becoming the new space M-30. Every year more satellites are launched and the risk of collisions that end in catastrophe grows exponentially. We can see it in a much more indicative way thanks to a new indicator called CRASH Clock (Collision and Significant Damage Watch). This indicator warns that only 2.8 days would separate the current moment from a serious collision if we were suddenly left without evasion maneuvering systems. The abysmal difference compared to 2018. A team of researchers led by Professor Sam Lawler, from the University of Regina, Canada, has developed this indicator to measure the increasing risk of collision in low Earth orbit (LEO). The CRASH Clock is not a countdown to Kessler syndrome (that theoretical scenario where collisions generate exponential cascades of space debris), but it does reflect how congested orbital space is and how quickly everything could get worse in the event of any failure in the prevention systems. The most worrying fact: in 2018, before the massive deployment of mega satellite constellations, that same clock showed 121 days. In just seven years we have gone from four months of margin to less than three days. Why does it matter now?. The density of objects in LEO has skyrocketed with the arrival of megaconstellations. starlinkthe SpaceX satellite network, is the most visible example. According to a report filed with the United States Federal Communications Commission (FCC), its second-generation satellites perform an average of 44 evasion maneuvers per year. Researchers have calculated that this is equivalent to one maneuver every 1.8 minutes across the entire constellation. “There’s no magic here, you’re just avoiding collisions by moving a Starlink satellite every two minutes. This is bad,” counted Lawler in a Mastodon thread. Graph showing the probabilities of a catastrophic collision if the avoidance maneuvering systems were to fail In the densest area of LEO, currently occupied by Starlink, approaches of less than a kilometer occur every 15 minutes. It may seem like a safe distance until you remember that these objects travel at seven kilometers per second. We depend on technological perfection. The system works… for now. SpaceX applies an extraordinarily conservative maneuver threshold: its satellites take evasive action when the probability of collision exceeds 3 in 10 million, well below the industry standard of 1 in 10,000. But this efficiency comes at a price: an absolute dependence on automatic systems continuing to operate without failure. The real danger is not in everyday life, but in unexpected events. A major solar storm, a widespread software glitch, or simply a miscalculation could trigger chain collisions. The study’s authors warn that we are currently “well within the Caution Zone,” with a greater than 10% chance of collisions occurring in any 24-hour period if avoidance maneuvers were to cease. What the simulations reveal. The researchers used two methods to verify their calculations: analytical analyzes with data from public catalogs and simulations of bodies in orbit. In one of the simulations, by pure chance, the first collision occurred just three hours after the hypothetical cessation of the maneuvers. Before megaconstellations, the densest part of the orbit experienced a closer approach of less than a kilometer a little more than once a day. Now it happens more than once every 15 minutes. International coordination, key. Beyond Starlink, other megaconstellations are in the launch phase. OneWeb, Chinese projects, future Amazon deployments… they will all share the same orbital space. Therefore, communication between all agencies, governments and institutions is essential. But of course, “will China talk to Starlink?”, “will the secret satellites of the United States Government talk to OneWeb?” are questions that Lawler reveals. Beyond collisions. The risks are not limited to collisions between objects. The study also points out problems already present: astronomy disruption observational, pollution in the atmosphereand increased risks of casualties on land. “From these safety and pollution metrics, it is clear that we have already put LEO under substantial stress, and changes to our approach are required immediately,” the paper’s authors conclude. What’s coming now. The team has created a website where to periodically update the CRASH Clock and keep this alert visible. It is not about catastrophism, they clarify, but about ‘situational awareness’. “In the short term, a major collision would look more like the Exxon Valdez tanker disaster than an immediate end to Hollywood-style orbital operations. Satellite operations could continue, but with different operating parameters and a higher risk of collision damage,” counted Lawler. Cover image | POT In Xataka | Elon Musk has been refusing to take SpaceX public for 20 years. 1.5 billion dollars have changed his mind

For some reason, there are people trying crash against a wall. Makes more sense than it seems

March 24, 2025 by usatoday24

A Tesla Model and, a wall that simulates a road, a youtuber of 65.8 million followers, its reputation. The reputation of Tesla, its security systems and semi -autonomous driving. Another YouTube. A cybertruck tesla. Another wall with another painted road. Indeed, there are people by throwing their tesla against a wall: what is happening? The coyote game and the roadrunner. “I am in my tesla at 64 km/h with the activated highway launched against a false wall in the style of the coyote and the roadrunner.” This is how it starts Mark Rober’s videoknown NASA engineer with a YouTube channel in which 65.8 million subscribers are counted. Rober travels in his car thrown against a false wall in which he has simulated that the road on which circulates continues and perfectly simulates what it really is: a decoration. The objective is to check if your Tesla Model and equipped exclusively with cameras offers better or worse performance than a car equipped with a Lidar system As far as autonomous driving systems are concerned. The background. Rober uses his video to explain the differences between a car that bases its semi -autonomous driving in the cameras exclusively and those that use the Lidar system. In 2021, Tesla already advanced that would base its entire system of driving aid in the cameras. In 2022 it was confirmed that the company would stop riding radars and sensors. The company ensures that its software is able to analyze everything that happens exclusively using cameras with reliability that matches that of radars and sensors. Along the way, evidently, the cost of assembling these components that irremediably increase the final cost of the vehicle is saved. To check if this is true, Rober tests A Tesla Moder and that only uses cameras In its Autopilot system and a lexus RX equipped with a lidar sensor of the luminar company. The cars are launched with activated semi -autonomous driving functions (Autopilot in the case of Tesla and adaptive cruise control in the lexus) to verify what happens in different scenarios. Among them, children who appear in the trajectory of the car or very intense rains. The wall. During his tests, the vehicle equipped with Lidar December and saved a child from being hit if he was already on the road, if he appeared behind an obstacle, in cases of fog and intense rain or if he was dazzled by an intense light. And yes, also if there was a wall on the road painted with the simulation of a road. The Tesla Model and could not say the same and the cameras did not exceed the tests of the fog, the rain … and the famous wall. In the latter case, the car was cheated and thought that the road continued, not stopping at all. And has exploded the controversy. The video Mark Rober has exploded the controversy on social networks for various reasons. The first and most obvious: does this type make sense? When can a car find a wall that completely covers a road simulating that it continues later? Beyond that this has to value each other, what is true is that Rober activates the highway a few seconds before reaching the wall and in the video it is observed how the system is disabled when it is going to collide. Rober has been accused of turning off the system but has defended himself on social networks publishing a complete video in which it is seen how the system is disabled before the crash. A criticism that has been repeated To Tesla vehicles, accusing the company of washing their hands before the authorities. But not only that, Rober has also been criticized for using a vehicle with autopilot and not with Full Self Driving (FSD)the most advanced autonomous driving system in Tesla while Lexus was equipped with Lidar, one of the most expensive technologies at the moment. Cameras vs lidar. The experiment has something interesting although not a resolution. The truth is that one might expect that a cameras based system is less effective than one accompanied by a lidar. The single combination of mounting a camera and a radar has been demonstrated The most effective option So far both day and night. The cameras analyze with images what they have in front but a lidar sends pulses of light constantly around them. That light bounces on the object and returns to the lidar. With the time that has passed since the shipping of the pulse of light until it receives it again you can create a 3D map that, in the case of the wall, serves to verify that there is an obstacle on the road even if it can deceive the camera. Autopilot and FSD. The other great criticism that has been made to Rober is that he uses a car equipped with Autopilot but not with the FSDthe most advanced driving aid system of Tesla. The FSD is offered as an additional option through a subscription and allows the driver to disregard driving (even in urban environments) as long as it remains attentive to what happens in front of his eyes. For example, the Autopilot demands from the driver the confirmation of lane change but the FSD navigates between the streets by itself, without the need for driver’s validations. In fact, in the United States it allows The car travels 65 meters without a driver Inside. Thus, we can ask a tesla parked to come to pick us up from a distance. And there are people by throwing their cars against a wall. Rober’s experiment has given wings to other YouTuber who have come out to defend the company. In response, Kyle Paul, who has 237,000 subscribers on his YouTube channel and specializes in Tesla videos, He has published his own test. The goal has been the same: will your Tesla Cybertruck before a wall that simulates the continuation of the road? And his answer is that if the car has … Read more

Two videos show how 80 people from a plane have saved life in the last plane crash. There is triple explanation

February 20, 2025 by usatoday24

After the tragic accident last month Between an American Airlines plane and an United States army helicopter that left 67 dead, on Monday there was a similar scene when it occurred A forced landing of Delta Air Lines’s flight 4819 In Toronto. Two videos showed rawly the scene that the 80 people who went on board and who ended the plane turned. And yet, it did not end in tragedy. How could they be saved? The “miracle.” Delta Air Lines’s flight 4819, operated by Endeavor Air, made a seemingly normal landing at the Pearson International Airport in Toronto until, Suddenly, everything went wrong. As soon as the wheels began to touch the wheel, the plane slipped, hit its right wing, turned on its axis and ended upside down, leaving an absolute chaos scene. But as we said, and despite the violence of the accident, The 80 passengers and crew managed to survivealthough 18 were injured, including an adult and a child in critical condition, although already out of danger. The impact and the fight to escape. They counted at the Washington Post in a report report that the passengers described the impact as “collide with a wall” before the plane turned. When they were face down, many remained suspended by their seat belts, seeing how the fuel slid through the windows and the fire consumed part of the aircraft. Pete Carlson, a paramedic on board, managed to unbutch and drop to the roof of the planenow turned into the ground, before helping others free from their seats. Meanwhile, passengers, in shock, collaborated with each other to get out as quickly as possible. A video recorded by Pete Koukovanother passenger, showed the crew organizing the evacuation with firm and urgent instructions, guiding people through emergency doors in an inverted plane, a rare scenario in aerial accidents. Outside, firefighters sprayed fushes with foam, while black smoke and flames hindered visibility in full snowstorm. How tragedy was avoided. In front of many Other aerial catastrophes More or less recent, experts have been detailing these days how the “miracle” was possible. Actually and as we will see, there is not a single factor, but several that helped that finally there was not a single death. Three were the main factors that we describe. Plane design and safety measures. He Bombardier CRJ900like all modern commercial planes, it is built with standards called “Crashworthiness”which means that its structure, safety belts and seats are designed to maximize survival in case of impact. In this regard, the cabin absorbed much of the strength of the crash without disintegrating, allowing passengers to go alive. According to Graham Braithwaiteexpert in aeronautics at the University of Cranfield, these elements were fundamental for passengers to resist the impact of the overturn without fatal injuries. Professor Michael J. McCormickfrom the Embry-Riddle University, stood out in the post That engineering and aeronautical regulations have advanced so much that accidents that in the past would have been lethal can now result in total survival. Assessment of the crew. Equally important that the design of the aircraft was crew trainingdecisive to organize a rapid and effective evacuation. Despite confusion and chaos, they managed to maintain order and get everyone out of the plane in a matter of minutes, even with the aircraft dump. Here it is important to remember that the regulations of the Federal Aviation Administration (FAA) require that Any plane with more than 44 passengers can be evacuated in less than 90 seconds. However, in this case, the evacuation was more complex because the plane was overturned, hindering the orientation and exit of the passengers. Immediate response of emergency equipment. The third of the key legs. Toronto airport authorities acted rapidlyensuring the area and preventing the fire from extending to fuel deposits. The first air control reports confirmed the severity of the situation: “The plane is face down and in flames”, which led to a rapid mobilization of rescue equipment. According to the norms of the International Civil Aviation Organization (ICAO), rescue teams They must respond in less than two minutes. In this case, firefighters and paramedics arrived at the place in a matter of minutes, stifled the fire and helped evacuate passengers. Causes of the accident. Although the exact cause of the accident has not yet been determined, it is known that landing occurred under wind conditions up to 70 km/hyve accumulated On the track after two recent storms. Thus, the airport firefighters ruled out problems with the track, which suggests that other factors, such as mechanical failure or human error, could have influenced. Be that as it may, in a context where Recent air accidents They have resulted in fatal tragediesthe total survival of flight passengers 4819 is a real testimony of the advance in aeronautical engineering, security protocols and emergency response. Although The images of the airplane burned and overturned They suggest an inevitable disaster, the combination of these factors allowed to avoid the worst. As Carlson underlined In his interview, after leaving the plane I only thought about one thing: “I didn’t care about the cold, I didn’t care how much I had to walk, all we wanted to be out of the plane.” Image | Yyzbrennan In Xataka | The main suspect of the plane crash in South Korea is common in aviation. The problem is that it does not usually end like this In Xataka | We have always thought that the rear seats of the plane are the deadliest in case of accident. We were wrong

Log In

Sign In

Forgot password?

Your password reset link appears to be invalid or expired.

Log in

Privacy Policy

Add to Collection

No Collections