Tag: Control

  • Ground Control provides architectural analysis of spaceflight facilities

    Ground Control provides architectural analysis of spaceflight facilities

    [ad_1]

    Ground Control: A Design History of Technical Lands and NASA’s Space Complex by Jeffrey E. Nesbit | Routledge | $39.19

    The morning of June 17, 1985 was just like any other humid morning on the Florida coast. It was just around sunrise, and from my parents’ car, I could see watery marshes reflecting the clouds aflame in blooms of pink, orange, and magenta. The air was thick, and through it I saw other cars, a long line of them, stretching into the distance, all pointed in the same direction to a point somewhere faraway. I stood up on the car roof to get a better look at the other visitors. Some had binoculars, many had cameras. So did I, and shortly before 7 a.m., I mounted my father’s 35mm Nikkormat to a small reflecting telescope. The car roof was slippery with condensation, but once I got some more secure footing, I made some quick calibrations on the camera rangefinder before focusing on the bronze, ogive-shaped external fuel tank belonging to the Space Shuttle Discovery. Even with my relatively powerful telescope, the spacecraft seemed small. And yet there it was: black leading-edged, delta-winged, and snub-nosed, poised for yet another jaunt into low Earth orbit. The world was still and quiet in these final moments. I stared at the morning sky reflected on the marshes, at the grasses rustling in a gentle breeze, and heard the countdown echoing in the distance when suddenly the air seemed to be torn apart with a terrific, guttural report. In those few seconds, I panned the telescope and snapped a handful of images of Discovery ascending into the sky on top of a curving column of fire and smoke. It was 7:33 a.m.

    Launch Complex 14 in Cape Canaveral
    Launch Complex 14, blockhouse, constructed in July 1956, Cape Canaveral Air Force Station, Merritt Island, Florida (Roland Miller)

    I was born in the aphelion of the Nixon administration. I am old enough to say my earliest heroes were people like Neil Armstrong, Michael Collins, and “Buzz” Aldrin. I had been to NASA’s Kennedy Space Center eight years earlier, in 1977, where my parents bought me all the Apollo, Gemini, Mercury, and SkyLab mission patches. We toured the mammoth Vehicle Assembly Building (VAB) with its giant star commemorating the United States Bicentennial, saw some rockets, ate in a cafeteria, tried on space helmets. Nothing was even remotely as cool as the Lunar Excursion Module simulator that gave kids like me a chance to peer through a triangular window and see the surface of the moon from an Apollo astronaut’s point-of-view.

    Launch Complex 39 crawler way construction, Kennedy Space Center
    Launch Complex 39 crawler way construction, Kennedy Space Center (Launch Operations Center), Merritt Island, Florida 1963 (NASA Archives)

    In 1984, I continued this nerdy trajectory when my parents moved to Seabrook, Texas, not far from the Johnson Space Center. I bought Space Shuttle mission patches. I even got to sit in a chair from a Gemini capsule. I went to nearby Clear Lake High School, and taking the bus to and from there, I will always remember our route along Space Center Boulevard, which curved along the eastern edges of the Johnson Space Center, seeing astronauts of all stripes jogging or playing softball.

    For the architect and educator Jeffrey E. Nesbit, these landscapes, buildings, artifacts, and purviews that played such a huge part in my upbringing are “technical lands.” The term is one that Nesbit and urbanist Charles Waldheim have explored in their edited volume from 2023, Technical Lands: A Critical Primer to describe spaces “united by their ‘exceptional’ status—their remote location, delimited boundary, secured accessibility, and vigilant management.” “Technical lands” is a more or less critical armature applicable to spaces such as proving grounds, research facilities, or even landscapes that are appropriated, sequestered, and hidden in the name of scientific and political agendas. Whether as a catch-all term that incorporates objects of various scales, from the industrial object to infrastructure, or even as a spatial category, technical lands had yet to receive any kind of sustained monographic inquiry—until now. In Ground Control: A Design History of Technical Lands and NASA’s Space Complex, Nesbit continues the agendas from his 2023 volume to provide a nuanced analysis of the buildings and landscapes at the Kennedy and Johnson Space Centers. More specifically, or at least, according to Nesbit’s own calculus, it is an “infrastructural history of the U.S. rocket launch complex.”

    view of launch complex in florida through periscope
    Launch Complex 14, periscope view downrange to rocket pad, Cape Canaveral Air Force Station, Merritt Island, Florida, March 8, 2021 (Jeffrey E. Nesbit)

    Nesbit presents a compelling case for a thorough architectural analysis of NASA’s primary human spaceflight facilities in a series of concise chapters. These chapters are enriched with his personal photographs from visits to the Space Centers, as well as archival materials and period ephemera. Drawing upon an extensive visual archive from the Historical American Buildings Survey and newly declassified documents, Nesbit illustrates how a potent blend of Kennan-inspired geopolitics and Eisenhower-era pragmatism shaped the design and construction of America’s spaceflight complexes.

    Launch Complex for space rockets in Florida
    Launch Complex, 34, rocket stand, Apollo 1 Memorial Ceremony, Cape Canaveral Air Station, January 27, 2018 Merritt Island, Florida (Jeffrey E. Nesbit)

    His work clearly reflects an affinity for previous scholarship at the intersection of architectural history and theory and history of science. In this vein, Ground Control weaves concepts from an enviable roster of works, from Keller Easterling’s theoretical explorations of infrastructure in Organization Space (1999), to Paul Virilio’s writings on the militarization of vision in War and Cinema: The Logistics of Perception (1986). Other writings by historian Felicity Scott and architect Fred Scharmen on the utopian and architectural ambitions of Gerard O’Neill and NASA’s Space Settlements program were clearly referenced (featured in the recent Emerging Ecologies show at the Museum of Modern Art.) And finally,  Jean-Louis Cohen’s 2011 work, Architecture in Uniform: Designing and Building for the Second World War, and its claim that war “disciplined” architectural modernism, emerges as the most frequently cited reference in this discourse. Together, these works illuminate the complex interplay between architectural thought, scientific advancement, and military-industrial spatial organization that shaped the “Space Complex,” a catchall term that describes NASA’s spaceflight architectures.

    When sifting through the layered references and armatures that propel Ground Control along its theoretical itineraries, it’s worth noting that the book is about much more than the Space Complex. Sure, there are blockhouses, bunkers, towers, and office buildings that tell a story of how NASA’s land acquisitions resulted in a new infrastructural type—a Spaceport—for connecting Earth to outer space. As Nesbit writes, Ground Control  “turns its lens back towards the range of scales—from the interior semi-buried objects to the occupation of vast and remote reaches across the globe.” His spaces of inquiry—the Space Center control rooms, highways, spacecraft assembly facilities, and administrative centers—are ordinary enough to escape the discerning eye of one expecting a flirtation with normative histories of aesthetics or styles. Nesbit never lets readers forget that NASA’s architecture is ubiquitous, but not generic. In other words, and despite HABS’s description of NASA’s administrative facilities as examples of International Style, this is a book where the buildings and structures are nameless and faceless. Without pedigree or distinction, the architecture of these complexes have had an outsized impact in the way the United States fashioned its self-image in the postwar world.

    Launch Complex 37 Blockhouse, Cape Canaveral,
    President John F. Kennedy, NASA director Kurt Debus, Wernher von Braun, and other primary officials are briefed by George Mueller, Launch Complex 37 Blockhouse, Cape Canaveral, Florida, November 16, 1963. (Courtesy NASA Archives)

    It’s a bit of a paradox. What kind of history does Ground Control tell? The book’s title suggests that it is a “design history,” a term that echoes the way Cohen’s Architecture in Uniform focused on “designing and building,” and not just architecture. Nesbit critiques NASA’s land consumption practices in Florida and Texas. Indeed, Nisbet and Waldheim have argued previously that the identification of “technical lands” is a political act in itself. And yet there is an uneasy case of Stockholm Syndrome here, for despite these critical readings of NASA’s facilities as technical lands, the methods employed by Nesbit in this book are doing the same thing—expanding and neutralizing an entire realm of inquiry under the banner of a theoretical category.

    As interesting as this can be, it leads to some historical blind spots. By limiting its inquiry to NASA’s Space Centers, Ground Control overlooks spaces that have arguably contributed to the Space Complex. Buildings and facilities such as the Jet Propulsion Laboratory in Pasadena, California; the testing facilities at Edwards Air Force Base; and the Full-Scale Wind Tunnel at the Langley Memorial Aeronautical Laboratory (built by NASA’s precursor, the National Advisory Committee on Aeronautics, or NACA) are crucial players in any history of the U.S. Space Programs.

    Launch Control Center in Kennedy Space Center
    Launch Control Center, lower firing room, Kennedy Space Center, Merritt Island, Florida, 1968 (NASA Archives)

    There are other moments throughout the book sure to raise a historian’s hackles. Nisbet’s emphasis on the ubiquity and uniformity of NASA’s architectures rightly mentions the role played by the U.S. Army Corps of Engineers (USACE). But there are no substantive investigations on how USACE culture contributed to formation of “technical lands,” a theme that is central to Architecture in Uniform as well as Todd Shallat’s Structures in the Stream: Water, Science, and the Rise of the U.S. Army Corps of Engineers (2010). Nisbet also ventures into some overreaching claims about modern architecture. This becomes especially evident when he discusses enclosures at the Space Centers along Victor Gruen’s Southdale Center (1956) Joseph Paxton’s Crystal Palace (1851), and even Gerard O’Neill’s Space Settlements. By inverting cultural historian Leo Marx’s concept of the “Machine in the Garden” into a “Garden in the Machine,” Nesbit claims that the Space Centers constituted a new garden typology, a “synthetic environment situated between terrestrial wilderness and extraterrestrial imagination.” This is a very promising argument, one that would have benefited from more discussions about the material culture of NASA’s Space Centers. Instead, it is abandoned for discussions about representation and visuality.

    This is certainly important, but it relies on a kind of unbalanced optic, as evidenced in the book’s final chapter. There, Nisbet reveals fascinating details about the terraforming needed to build the paved roads used by the crawler-transporters ferrying rockets from the VAB to the launchpads. This becomes yet another instance of “an infrastructure producing images of wasteland for departing Earth.” But it overlooks the roads’ inability to withstand the crawler-transporters’ immense weight. This problem led to expensive repaving every three years, steering NASA engineers to an innovative solution: to cover the asphalt with Alabama river rock, which would be crushed into a smooth surface with every journey of the crawler-transporters.

    Mercury Capsule technicians
    Mercury Capsule hoisted into Altitude Chamber by McDonnell technicians, Missile Assembly Building S, Cape Canaveral Industrial Area, Merritt Island, Florida, September 7, 1962 (NASA Archives)

    Ground Control is also marred by missing footnotes (including a very significant one that was supposed to reference Architecture in Uniform in chapter 5) and some image citations that are wanting more explanation. Also missing from this book is any discussion of NASA’s future Manned Spacecraft Center as a vital part of exurban development in the Houston areas. Adam Higginbotham makes such a point in Challenger: A True Story of Heroism and Disaster on the Edge of Space (2024), reminding readers that the location of the Johnson Space Center had as much to do with NASA’s aspirations as it did with its proximity to Congressional districts home to important appropriations committee members, and to the planning and design of Clear Lake City.

    Ground Control begins at the Cape Canaveral Air Force Base on July 24, 1950, with the launch of Bumper 8, a hybrid rocket made of pieces from a WAC Corporal missile and a propulsion stage from a captured and reconfigured Nazi V2 rocket. It is an auspicious beginning for the book as this launch ties America’s entry into outer space with Wernher von Braun’s continuation of his World War II–era work at Fort Bliss, Texas, and Redstone Arsenal in Alabama, which spurred the development of ICBMs and the headlong drift towards global nuclear war. I also remembered that morning in 1985 when I photographed the launch of shuttle Discovery from a vantage point not far from the site of Bumper 8’s ascent. But as a writer, historian, and critic, I could help but think of Thomas Pynchon’s Gravity’s Rainbow (1973), a book that begins and ends with the launch of a V2 rocket. The rocket becomes a cipher for modernity and a talisman that conjures a dark vision of military technologies as fetish objects. I wonder if the technical lands of Kennedy and Johnson Space Centers are operating on a similar register. Does the design culture at NASA have a metonymic relationship to larger patterns of institutionally sponsored land-use practices? With every moment spent poring over Ground Control, the more I realized that the design, construction, and development of the Kennedy and Johnson Space Centers are more space oddity than space complex.

    Enrique Ramirez is a historian of art and architecture. He lives in Brooklyn.



    [ad_2]

    Source link

  • Meet the Far-Right Constitutional Sheriffs Ready to Assert Control if Trump Loses

    Meet the Far-Right Constitutional Sheriffs Ready to Assert Control if Trump Loses

    [ad_1]

    Tim Marchman: This is rooted as you write in white supremacist beliefs. Can you unpack that a little bit?

    David Gilbert: It is, and you can trace it back from the late-1960, early-1970s to a movement called Posse Comitatus, which was founded some say by a guy called William Potter Gale. He was at the time a minister in this militant anti-Semitic white nationalist quasi-religion, kind of known as Christian Identity. He believed that the sheriffs were these protectors of the citizens and that they had the power to call up militias and that they should be enshrined in law as the ultimate power law enforcement anywhere in the country. We’ve seen across the years that these far-right or Constitutional Sheriffs, no matter what they’ve done in terms of the extreme actions they’ve taken, if they have a base of supporters in their locality or in their county who believe in what they’re doing, they will be voted back into office for decades at a time.

    Tim Marchman: The mandate of the public is pretty powerful, but some of these sheriffs are citing a higher source of authority. They say their power derives from God, which seems pretty unconstitutional given the separation of church and state in America. How do they respond to that?

    David Gilbert: Well, they respond by saying that the separation of church and state is not something that really exists. They say that, that again is a misreading of the Constitution, and the entire Constitutional Sheriff’s movement is deeply infused with Christian nationalist beliefs and ideology. Most of the Constitutional Sheriffs who I’ve spoken to over the last six months or so are eager for the US to return to being a nation rooted in Christianity, where Christianity is at the center of all aspects of life, be that law enforcement or education or government or culture. They believe that in that society because they believe they got their power from God, that they will be the most powerful law enforcement individuals across the country.

    Tim Marchman: Under this constitutional order as they understand it, is there a role for constitutional governors or constitutional mayors, or are these powers unique to sheriffs?

    David Gilbert: They seem to believe that these powers are unique to sheriffs. In all the time I’ve been covering this, I’ve never heard any of them speak about other figures, whether in government or law enforcement that would hold similar powers to a sheriff. Again, that comes back to the idea that this is somehow enshrined in the Constitution. As we said, it’s not, but in their belief system, in their ideology, they can trace the sheriff. It’s one of the oldest law enforcement offices in the world. It goes right back to the UK where the sheriff did the bidding of the local magistrates and collected taxes and stuff like that. It’s obviously been exported from England to the US and it has persisted since the beginning of the US nation. They believe that, that is key to giving them the power that no one else in the US has because at a local level, they’re there to protect their citizens, and the citizens are the ones who elect them, and therefore, that is their duty. Even if other positions like a governor is elected by the people, they don’t seem to believe that, that position should have the similar kind of constitutional protections.

    [ad_2]

    Source link

  • Best parental control software in 2024 (UK)

    Best parental control software in 2024 (UK)

    [ad_1]

    This content originally appeared on Mashable for a US audience and has been adapted for the UK audience.

    We all have the best intentions when it comes to entertaining the kids — outdoor activities, board games, sports classes, wood whittling — but it doesn’t always turn out that way. Screens are just too handy and easy. Not to mention parents are generally exhausted from the reality of the working week and, well, parenting.

    Indeed, it’s only a matter of time before you’ll turn to technology to save the day. Nothing captures the attention of a child like a desktop, tablet, laptop, or mobile, and there’s nothing you can do about it. It’s time to just accept it. And let’s be honest: Modern kids have never known a world without the Internet and screens.

    The problem, of course, is that the internet is full of danger for kids, including explicit materials, bullying, and cyber criminals. Fortunately, there is a solution in the form of parental control software. If that sound a bit new and scary, here’s a quick explainer and a selection of the best options to keep your kids safe.

    What is parental control software?

    In the most basic terms, it’s a software that helps you to manage all the devices your kids use — as well as the way they use them. It’s all about making sure the online world is a safe space for your kids to explore. Though you should chat with your littles ones about the importance of safety before you let them loose online or implement any software.

    How does parental control software work?

    Parental control software lets you keep tabs your child’s phone, tablet, or computer via an app. You can monitor internet searches and browsing history, as well as block inappropriate apps (or block distracting apps during homework hours) or even limit device usage all together. It’s less invasive than physically going through their phone or computer. And when kids know that their activity is being watched or managed, they’re more likely to practice better cyber safety habits.

    What’s a safe amount of screen time for kids?

    It’s easy to get sucked into scaremongering over kids’ screen time. It’s not the end of the world if kids have more screens than parents would like, and parents should give themselves a break when it comes to relieving the stress of parenting with a little screen time. The American Academy of Pediatrics (AAP) suggests very limited screen time for children under two years old, an hour a day between the ages of two and five, and two hours between the ages of five and eight. Beyond that it’s important to talk to your child about responsible screen use. Problems associated with too much screen time include weight gain, sleep problems, and general inactivity.

    What is geofencing?

    Parental control software can also help keep tabs on your kid when they’re not at home. Though most have GPS and some sort of basic “Where’s my child?” function, only a select few software options offer geofencing, which is a big help for parents juggling the schedules of multiple children.

    This location-based service lets you set up virtual boundaries around where a child should or shouldn’t be, as well as a specific time that the child should be there. Let’s say that your child goes straight from school to a sports practice three times a week. The geofencing feature will monitor their phone’s location and will alert you if your child doesn’t show up to the scheduled area on time. Teenagers may even enjoy not being bombarded with “Where are you?” texts. Geofencing can also be used to monitor web time when your child is in a designated location. For instance, some parents may disable games or social media during school.

    What is the best parental control software?

    There are plenty of parental control software options out there for you to consider, with something for every family. We have researched everything on offer and lined up a selection of the very best services, including software from big names like Kaspersky, Qustodio, and Norton. Basically, we’ve done all the hard work for you.

    These are the best parental control software options for 2024.



    [ad_2]

    Source link

  • Anthropic’s new AI model can control your PC

    Anthropic’s new AI model can control your PC

    [ad_1]

    In a pitch to investors last spring, Anthropic said it intended to build AI to power virtual assistants that could perform research, answer emails, and handle other back-office jobs on their own. The company referred to this as a “next-gen algorithm for AI self-teaching” — one it believed that could, if all goes according to plan, automate large portions of the economy someday.

    It took a while, but that AI is starting to arrive.

    Anthropic on Tuesday released an upgraded version of its Claude 3.5 Sonnet model that can understand and interact with any desktop app. Via a new “Computer Use” API, now in open beta, the model can imitate keystrokes, button clicks, and mouse gestures, essentially emulating a person sitting at a PC.

    “We trained Claude to see what’s happening on a screen and then use the software tools available to carry out tasks,” Anthropic wrote in a blog post shared with TechCrunch. “When a developer tasks Claude with using a piece of computer software and gives it the necessary access, Claude looks at screenshots of what’s visible to the user, then counts how many pixels vertically or horizontally it needs to move a cursor in order to click in the correct place.”

    Developers can try out Computer Use via Anthropic’s API, Amazon Bedrock, and Google Cloud’s Vertex AI platform. The new 3.5 Sonnet without Computer Use is rolling out to Claude apps, and brings various performance improvements over the outgoing 3.5 Sonnet model.

    Automating apps

    A tool that can automate tasks on a PC is hardly a novel idea. Countless companies offer such tools, from decades-old RPA vendors to newer upstarts like Relay, Induced AI and Automat.

    In the race to develop so-called “AI agents,” the field has only become more crowded. AI agents remains an ill-defined term, but it generally refers to AI that can automate software.

    Some analysts say AI agents could provide companies with an easier path to monetizing the billions of dollars that they’re pouring into AI. Companies seem to agree: according to a recent Capgemini survey, 10% of organizations already use AI agents and 82% will integrate them within the next three years.

    Salesforce made splashy announcements about its AI agent tech this summer, while Microsoft touted new tools for building AI agents yesterday. OpenAI, which is plotting its own brand of AI agents, sees the tech as a step toward super-intelligent AI.

    Anthropic calls its take on the AI agent concept an “action-execution layer” that lets the new 3.5 Sonnet perform desktop-level commands. Thanks to its ability to browse the web (not a first for AI models, but a first for Anthropic), 3.5 Sonnet can use any website and any application.

    Claude 3.5 Sonnet new
    Anthropic’s new AI can control apps on a PC. Image Credits:Anthropic

    “Humans remain in control by providing specific prompts that direct Claude’s actions, like ‘use data from my computer and online to fill out this form,’” an Anthropic spokesperson told TechCrunch. “People enable access and limit access as needed. Claude breaks down the user’s prompts into computer commands (e.g.. moving the cursor, clicking, typing) to accomplish that specific task.”

    Software development platform Replit has used an early version of the new 3.5 Sonnet model to create an “autonomous verifier” that can evaluate apps while they’re being built. Canva, meanwhile, says that it’s exploring ways in which the new model might be able to support the designing and editing process.

    But how is this any different than the other AI agents out there? It’s a reasonable question. Consumer gadget startup Rabbit is building a web agent that can do things like buying movie tickets online; Adept, which was recently acqui-hired by Amazon, trains models to browse websites and navigate software; and Twin Labs is using off-the-shelf models, including OpenAI’s GPT-4o, to automate desktop processes.

    Anthropic claims the new 3.5 Sonnet is simply a stronger, more robust model that can do better on coding tasks than even OpenAI’s flagship o1, per the SWE-bench Verified benchmark. Despite not being explicitly trained to do so, the upgraded 3.5 Sonnet self-corrects and retries tasks when it encounters obstacles, and can work toward objectives that require dozens or hundreds of steps.

    Claude 3.5 Sonnet new
    The new Claude 3.5 Sonnet model’s performance on various benchmarks. Image Credits:Anthropic

    But don’t fire your secretary just yet.

    In an evaluation designed to test an AI agent’s ability to help with airline booking tasks, like modifying a flight reservation, the new 3.5 Sonnet managed to complete less than half of the tasks successfully. In a separate test involving tasks like initiating a return, 3.5 Sonnet failed roughly a third of the time.

    Anthropic admits the upgraded 3.5 Sonnet struggles with basic actions like scrolling and zooming, and that it can miss “short-lived” actions and notifications because of the way it takes screenshots and pieces them together.

    “Claude’s Computer Use remains slow and often error-prone,” Anthropic writes in its post. “We encourage developers to begin exploration with low-risk tasks.”

    Risky business

    But is the new 3.5 Sonnet capable enough to be dangerous? Possibly.

    A recent study found that models without the ability to use desktop apps, like OpenAI’s GPT-4o, were willing to engage in harmful “multi-step agent behavior,” such as ordering a fake passport from someone on the dark web, when “attacked” using jailbreaking techniques. Jailbreaks led to high rates of success in performing harmful tasks even for models protected by filters and safeguards, according to the researchers.

    One can imagine how a model with desktop access could wreak more havoc — say, by exploiting app vulnerabilities to compromise personal info (or storing chats in plaintext). Aside from the software levers at its disposal, the model’s online and app connections could open up avenues for malicious jailbreakers.

    Anthropic doesn’t deny that there’s risk in releasing the new 3.5 Sonnet. But the company argues that the benefits of observing how the model is used in the wild ultimately outweigh this risk.

    “We think it’s far better to give access to computers to today’s more limited, relatively safer models,” the company wrote. “This means we can begin to observe and learn from any potential issues that arise at this lower level, building up computer use and safety mitigations gradually and simultaneously.”

    Claude 3.5 Sonnet new
    Image Credits:Anthropic

    Anthropic also says it has taken steps to deter misuse, like not training the new 3.5 Sonnet on users’ screenshots and prompts, and preventing the model from accessing the web during training. The company says it developed classifiers to “nudge” 3.5 Sonnet away from actions perceived as high-risk, such as posting on social media, creating accounts and interacting with government websites.

    As the U.S. general election nears, Anthropic says it is focused on mitigating election-related abuse of its models. The U.S. AI Safety Institute and U.K. Safety Institute, two separate but allied government agencies dedicated to evaluating AI model risk, tested the new 3.5 Sonnet prior to its deployment.

    Anthropic told TechCrunch it has the ability to restrict access to additional websites and features “if necessary,” to protect against spam, fraud, and misinformation, for example. As a safety precaution, the company retains any screenshots captured by Computer Use for at least 30 days — a retention period that might alarm some devs.

    We’ve asked Anthropic under which circumstances, if any, it would hand over screenshots to a third party (e.g., law enforcement) if asked, and will update this post if we hear back.

    “There are no foolproof methods, and we will continuously evaluate and iterate on our safety measures to balance Claude’s capabilities with responsible use,” Anthropic said. “Those using the computer-use version of Claude should take the relevant precautions to minimize these kinds of risks, including isolating Claude from particularly sensitive data on their computer.”

    Hopefully, that’ll be enough to prevent the worst from occurring.

    A cheaper model

    Today’s headliner might’ve been the upgraded 3.5 Sonnet model, but Anthropic also said an updated version of Haiku, the cheapest, most efficient model in its Claude series, is on the way.

    Claude 3.5 Haiku, due in the coming weeks, will match the performance of Claude 3 Opus, once Anthropic’s state-of-the-art model, on certain benchmarks at the same cost and “approximate speed” of Claude 3 Haiku.

    “With fast speeds, improved instruction following, and more accurate tool use, Claude 3.5 Haiku is well suited for user-facing products, specialized sub-agent tasks, and generating personalized experiences from huge volumes of data like purchase history, pricing or inventory data,” Anthropic wrote in a blog post.

    3.5 Haiku will initially be available as a text-only model and later as part of a multimodal package that can analyze both text and images.

    Claude 3.5 Haiku
    3.5 Haiku’s benchmark performance. Image Credits:Anthropic

    So once 3.5 Haiku is available, will there be much reason to use 3 Opus? What about 3.5 Opus, 3 Opus’ successor, which Anthropic teased back in June?

    “All of the models in the Claude 3 model family have their individual uses for customers,” the Anthropic spokesperson said. “Claude 3.5 Opus is on our roadmap and we’ll be sure to share more as soon as we can.”

    [ad_2]

    Source link

  • Anthropic Wants Its AI Agent to Control Your Computer

    Anthropic Wants Its AI Agent to Control Your Computer

    [ad_1]

    Demos of AI agents can seem stunning, but getting the technology to perform reliably and without annoying (or costly) errors in real life can be a challenge. Current models can answer questions and converse with almost humanlike skill, and are the backbone of chatbots such as OpenAI’s ChatGPT and Google’s Gemini. They can also perform tasks on computers when given a simple command by accessing the computer screen as well as input devices like a keyboard and trackpad, or through low-level software interfaces.

    Anthropic says that Claude outperforms other AI agents on several key benchmarks including SWE-bench, which measures an agent’s software development skills, and OSWorld, which gauges an agent’s capacity to use a computer operating system. The claims have yet to be independently verified. Anthropic says Claude performs tasks in OSWorld correctly 14.9 percent of the time. This is well below humans, who generally score around 75 percent, but considerably higher than the current best agents—including OpenAI’s GPT-4—which succeed roughly 7.7 percent of the time.

    Anthropic claims that several companies are already testing the agentic version of Claude. This includes Canva, which is using it to automate design and editing tasks, and Replit, which uses the model for coding chores. Other early users include The Browser Company, Asana, and Notion.

    Ofir Press, a postdoctoral researcher at Princeton University who helped develop SWE-bench, says that agentic AI tends to lack the ability to plan far ahead and often struggles to recover from errors. “In order to show them to be useful we must obtain strong performance on tough and realistic benchmarks,” he says, such as reliably planning a wide range of trips for a user and booking all the necessary tickets.

    Kaplan notes that Claude can already troubleshoot some errors surprisingly well. When faced with a terminal error when trying to start a web server, for instance, the model knew how to revise its command to fix it. It also worked out that it had to enable popups when it ran into a dead end browsing the web.

    Many tech companies are now racing to develop AI agents as they chase market share and prominence. In fact, it might not be long before many users have agents at their fingertips. Microsoft, which has poured upwards of $13 billion into OpenAI, says it is testing agents that can use Windows computers. Amazon, which has invested heavily in Anthropic, is exploring how agents could recommend and eventually buy goods for its customers.

    Sonya Huang, a partner at the venture firm Sequoia who focuses on AI companies, says for all the excitement around AI agents, most companies are really just rebranding AI-powered tools. Speaking to WIRED ahead of the Anthropic news, she says that the technology works best currently when applied in narrow domains such as coding-related work. “You need to choose problem spaces where if the model fails, that’s okay,” she says. “Those are the problem spaces where truly agent native companies will arise.”

    A key challenge with agentic AI is that errors can be far more problematic than a garble chatbot reply. Anthropic has imposed certain constraints on what Claude can do—for example, limiting its ability to use a person’s credit card to buy stuff.

    If errors can be avoided well enough, says Press of Princeton University, users might learn to see AI—and computers—in a completely new way. “I’m super excited about this new era,” he says.

    [ad_2]

    Source link

  • New system lets X-62A aircraft adjust flight control behavior in the air

    New system lets X-62A aircraft adjust flight control behavior in the air

    [ad_1]

    Lockheed Martin’s famous-yet-secret Skunk Works is conducting live flight tests of its Smart Adaptive Flight Control Environment (SAFE) autonomous AI system on the X-62A. The system can be reconfigured in real time while the X-plane is still in the air.

    The US Air Force’s X-62A Variable In-flight Simulation Test Aircraft (VISTA) is one of the world’s premiere flying test beds for autonomous flight systems It’s a modified F-16D Block 30 Peace Marble Il aircraft that has been fitted with Block 40 avionics. Operated by the U.S. Air Force Test Pilot School at Edwards Air Force Base in California, it began life as the NF-16D and was renamed the X-62A in June 2021.

    Its purpose is to use the VISTA Simulation System (VSS), developed by Calspan, and Lockheed Martin’s Model Following Algorithm (MFA) and System for Autonomous Control of the Simulation (SACS) to provide the fighter aircraft with the ability to carry out complex flight tests of autonomy and AI technologies.

    Along with advanced sensors and Getac tablet displays in both the forward and aft sections of the cockpit, the X-62A now has the ability to not only reconfigure itself at the baseline level, but to do so in real time while in flight. As a result, flight testing is vastly sped up because the aircraft can shift to a new iteration without having to land and head back to the hangar.

    In addition, SAFE uses AI and machine learning to produce a learn-to-fly capability that, in its latest version, uses online learning to correct control errors. It also has a bolt-on capability that allows the system to be quickly installed in a wide variety of autonomous aircraft.

    “The SAFE system has the potential to revolutionize flight tests, operational aircraft performance and safety,” said Jeff Hakes, director, Revolutionary Air Vehicles, Emerging Concepts and Technologies, Lockheed Martin Skunk Works. “Our work advancing state of the art adaptive technologies continues to evolve and deliver more flexible, capable and collaborative capabilities across current and future platforms.”

    Source: Lockheed Martin



    [ad_2]

    Source link

  • In latest move against WP Engine, WordPress takes control of ACF plugin

    In latest move against WP Engine, WordPress takes control of ACF plugin

    [ad_1]

    The dispute between WordPress founder Matt Mullenweg and hosting provider WP Engine continues, with Mullenweg announcing that WordPress is “forking” a plugin developed by WP Engine.

    Specifically, Advanced Custom Fields — a plugin making it easier for WordPress users to customize their edit screens — is being taken out of WP Engine’s hands and updated as a new plugin called Secure Custom Fields.

    Mullenweg wrote that this step was necessary “to remove commercial upsells and fix a security problem.”

    The Advanced Custom Fields team responded on X, describing this as a situation where a plugin “under active development” has been “unilaterally and forcibly taken away from its creator without consent,” which it said has never happened “in the 21 year history of WordPress.”

    “This essential community promise has been violated, and we ask everyone to consider the ethics of such an action, and the new precedent that has been set,” the ACF team wrote.

    Both Mullenweg’s blog post and a reply from WordPress claim that similar situations have, in fact, happened before, though Mullenweg added, “This is a rare and unusual situation brought on by WP Engine’s legal attacks, we do not anticipate this happening for other plugins.”

    They also pointed to WordPress’ plugin guidelines, which give WordPress the right to disable or remove any plugin, remove developer access, or change a plugin “without developer consent, in the name of public safety.”

    Some background: WordPress is a free, open source content management system used by many websites (including TechCrunch), while companies like WP Engine and Mullenweg’s Automattic offer hosting and other commercial services on top. 

    Last month, Mullenweg published a blog post criticizing WP Engine as a “cancer to WordPress.” His criticisms covered everything from WP Engine’s lack of support for revision history to its investor Silver Lake, but he also suggested that its “WP” branding confuses customers, making it sound like the company is officially connected to WordPress.

    Cease-and-desist letters have gone both ways, with WP Engine claiming Mullenweg threatened to take a “scorched earth nuclear approach” unless the company paid to license the WordPress trademark.

    WordPress banned WP Engine from accessing WordPress.org, briefly lifted the ban, then imposed it again. This essentially prevents WP Engine from updating the plugin through WordPress.org — so it can’t offer automatic updates to address security issues.

    WP Engine has, however, published a workaround for users who want to update the plugin and continue using ACF. (It says the workaround is only necessary for ACF’s free users, as pro users will continue to receive updates through the ACF website.)

    Moving forward, Mullenweg wrote that Secure Custom Fields will be available as a non-commercial plugin: “If any developers want to get involved in maintaining and improving it, please get in touch.”

    [ad_2]

    Source link

  • iOS 18 Control Center: 18 apps that add useful actions to your iPhone

    iOS 18 Control Center: 18 apps that add useful actions to your iPhone

    [ad_1]

    Apple’s iOS 18 software update rolled out earlier this month, and it brought significant changes to the iOS Control Center, which provides convenient access to commonly used functions with a simple swipe down on the Home Screen. Now, in addition to the existing controls like Wi-Fi, Bluetooth, Flashlight, and so on, Apple allows third-party app developers to create controls for their own apps. 

    Below, we’ve compiled a list of iOS 18-ready apps that users can try in order to take advantage of the redesigned Control Center.

    Bulletin

    Image Credits: Screenshot by TechCrunch

    Bulletin, the AI-powered news reader app that summarizes articles, notably has a total of nine Control Center widgets (the most we’ve seen so far). These include new options to quickly perform app actions, such as going straight to the “For You” tab without opening the app. Additionally, options to go to different news categories, including “Business,” “Entertainment,” “Fashion,” “Sports,” and “Technology.” There’s also an option to pull up the app’s “Smart Summary” feature, which uses AI to recap the top articles of the day.

    CARROT Weather

    Image Credits: Screenshot by TechCrunch

    CARROT Weather, the app that delivers snarky weather forecasts, is also taking advantage of the iOS update with two Control Center widgets. Now, users can add shortcuts to quickly open the Weather and Maps tabs.

    Crouton

    Image Credits: Screenshot by TechCrunch

    Recipe organizer and meal planner app Crouton launched a helpful Control Center widget for users looking to import new recipes with one tap. 

    Cheatsheet Notes

    Image Credits: Screenshot by TechCrunch

    Cheatsheet is a widget that lets users write down important reminders such as hotel room numbers, license plate numbers, and lock combinations to quickly access from their iPhone home screen. The app now has a Control Center widget where people can swipe down to access one of their most important notes. There’s also a new option to add a new “Cheat” or note.

    FitBee

    Image Credits: Screenshot by TechCrunch

    FitBee is a new nutrition-tracking app that lets you log a meal by taking a photo of it, and AI can estimate the meal’s calories and macros, according to the app’s developers. FitBee recently released two features that take advantage of the new Control Center, including the ability to scan barcodes and snap photos of food items, letting users log even faster.

    Focused Work

    Focused Work iOS 18 app control center
    Image Credits: Screenshot by TechCrunch

    Focused Work, a time management app that helps users focus, has new control widgets that allow people to start a focus session and access the “Scratch Pad,” or notes section for writing down thoughts that can be revisited after the session. 

    LookUp

    LookUp iOS 18 control widget
    Image Credits: Screenshot by TechCrunch

    LookUp is a dictionary app that teaches users one word a day. It introduced several actions to the Control Center, such as the ability to search for a word, scan a word with the phone camera to see its definition, or set up the app’s “Reading Mode,” which turns on a Live Activity to quickly look up a word from the Lock Screen.

    Mango Baby

    Image Credits: Screenshot by TechCrunch

    The Mango Baby app, designed for newborn care, offers parents five useful widgets that can be added to their Control Center. The app now allows users to log diaper changes, and it provides the option to set up different timers such as a sleep timer, pumping timer, bath timer, and tummy time timer to help parents keep track of their baby’s daily routine.

    MediSOS

    MediSOS iOS 18 app control widget
    Image Credits: Screenshot by TechCrunch

    The most notable feature that MediSOS offers is its emergency siren that activates when users press the SOS button in the app, notifying people around them that they’re in need of medical attention. Now, people can start the medical emergency alarm from the Control Center thanks to the iOS 18 update. 

    Dark Noise

    Dark Noise, the app that plays ambient sounds, added multiple options to the Control Center, so users can choose a type of noise, start playing rain sounds, and set a timer all without opening the app. 

    Pedometer++

    Image Credits: Screenshot by TechCrunch

    Pedometer++ lets people track their step count, walking distance, active calories, and heart rate. The app’s new widgets let users instantly start a walk, run, and hike. They can also open the app with one tap. 

    Saturn

    Image Credits: Screenshot by TechCrunch

    Saturn is a recently launched app dubbed the “Mastadon for Everyone” and works with the decentralized social network to show popular content. Now, users can access a series of actions for their iPhone Control Center, such as the ability to open tabs like “Communities,” “Highlights,” “Latest News,” “Popular Pages,” “Suggested Accounts,” and “Trending Tags.” Users can also create a new post or open the search function.

    Sofa 

    Sofa app iOS 18 Control Center
    Image Credits: Screenshot by TechCrunch

    Sofa allows users to create lists of different things, from apps to download to things to watch, read, play, and listen to. There are four new control widgets to choose from: “The Pile,” which has everything saved in one place; “Pinned items,” which takes users to their favorite items; “Logbook,” for things they’ve already completed; and “List,” which quickly pulls up a go-to list.  

    Spark

    Image Credits: Screenshot by TechCrunch

    Spark is an AI email assistant designed to help people handle and sort emails while connecting multiple accounts into one inbox. The app offers four widgets that can be added to the Control Center, allowing users to easily access the app for composing emails, conducting searches, accessing attachments, or checking the calendar.

    Tripsy

    Image Credits: Screenshot by TechCrunch

    Tripsy is a travel planner app that lets people plan a list of activities for their next trip, receive real-time flight updates, and even track expenses. Its new control actions allow people to quickly access their trip itinerary or create their upcoming travel plan.  

    TV Remote

    Image Credits: Screenshot by TechCrunch

    TV Remote is a fully customizable remote app for LG, Samsung, Roku, and Sony televisions. One of the most helpful new actions for the Control Center is “Remote Command,” letting users quickly access the remote whenever they want without having to launch the app. 

    Waterllama

    Waterllama app control center widgets
    Image Credits: Screenshot by TechCrunch

    Waterllama, the app that tracks daily water intake to ensure people stay hydrated, has a new Control Center widget that lets users quickly add a custom amount of water or other beverages to the daily total. Notably, the widgets aren’t static, so when people tap on them, they include animations. Users can also customize the animation from six different options: splash, big splash, drizzle, droplets, circles, and a pixelated splash. 

    Zenitizer 

    Image Credits: Screenshot by TechCrunch

    Zenitizer allows users to create meditation timers, listen to relaxing sounds, track daily goals, and set reminders. Now, people can add the app’s widget to their Control Center to set a meditation timer with one tap. 

    [ad_2]

    Source link

  • Tiny implanted magnets make for better, simpler control of prosthetic hand

    Tiny implanted magnets make for better, simpler control of prosthetic hand

    [ad_1]

    While a number of groups are developing thought-controlled prosthetic hands, most of the devices require electrodes or other relatively fiddly electronics to be implanted in the amputee’s body. Such is not the case with a new hand, however, which works with tiny inert implanted magnets.

    For most thought-controlled prosthetic hands, electrodes are implanted in the user’s residual arm stump, or even in the brain.

    As that person thinks about moving the fingers of the hand in a given fashion, those electrodes detect the nerve signals that travel from the brain towards the hand. Integrated electronics translate those signals into commands that are relayed to the hand’s servos, causing the fingers to move as desired.

    Led by Prof. Christian Cipriani, scientists at Italy’s Scuola Superiore Sant’Anna (Sant’Anna School of Advanced Studies) set out to develop a simpler yet better-performing alternative. The resulting system was trialled on a 34-year-old test subject named Daniel, who lost his left hand in an accident in September of 2022.

    Daniel uses the hand to pour water from a cup
    Daniel uses the hand to pour water from a cup

    Sant’Anna School of Advanced Studies

    Utilizing MRI scans and electromyography (the monitoring of electrical activity in muscle tissue), the scientists started by determining how the muscles in Daniel’s left arm stump contracted when he thought about moving the fingers of his missing hand. Part of what made Daniel a good recipient was the fact that he was already experiencing phantom limb sensation, a common phenomenon in which amputees still feel the presence of a missing limb.

    Based on their observations, the researchers implanted six small magnets – each just a few millimeters in size – into some of the key muscles. They then fitted Daniel with a Mia-Hand prosthetic hand, which included a carbon fiber cuff that was worn over his arm stump.

    As the muscles in that arm contracted in response to his finger-movement thoughts, magnetic field sensors in the cuff detected the corresponding movements of the implanted magnets. Depending on which magnets/muscles were moving in which way, the fingers of the hand were prompted to move in the associated fashion.

    The hand was used to remove a lid from a glass jar
    The hand was used to remove a lid from a glass jar

    Sant’Anna School of Advanced Studies

    In tests performed so far, Daniel has used the hand to perform tasks such as opening a jar, using a screwdriver, closing a zip-lock bag, cutting with a knife, plus picking up and moving various objects. He was also able to modulate his grasping force when handling fragile items.

    “The trial on the first patient was successful,” says Cipriani. “We are ready to extend these results to a broader range of amputations.”

    A paper on the study, which is part of the European Research Council’s MYTI project, was recently published in the journal Science Robotics. Daniel can be seen using the hand, in the video below.

    Scientist at MIT are developing a similar system, in which prostheses are controlled by implanted magnetic beads.

    La protesi del futuro

    Source: Sant’Anna School of Advanced Studies



    [ad_2]

    Source link

  • $599 robot mower handles tough terrain with RTK and cruise control

    $599 robot mower handles tough terrain with RTK and cruise control

    [ad_1]

    It’s a competitive field when it comes to robot mowers, but this one may be a cut above most, with a pleasing price point, supreme weather detection, lawn mapping and a host of other features that mean you won’t have to keep watch over it as it goes about its work.

    From a company that’s been working on mastering the auto-mower since 2016, the Anthbot Genie promises to be easy to set up, and with four cameras and RTK – real-time kinematic positioning, or high-precision satellite positioning – it identifies and veers around obstacles such as animals and furniture in real time.

    ANTHBOT Genie: Plug and Play

    If GPS positioning is weak in the yard, the Genie will use the backup VSLAM (visual simultaneous localization and mapping) system, which is handy if it needs to venture out of sight, such as under trees or eaves.

    Perhaps one of its best features is its adaptive cruise control mapping, which stores the coordinates of an area to be mowed and identifies the boundaries, so you won’t have to babysit it while it goes about its work.

    Your hedgehogs will be safe when the Genie is at work
    Your hedgehogs will be safe when the Genie is at work

    Anthbot/Kickstarter

    The brushless motor powers blades that spin at 3,000 RPM, and will adjust in height to cut the top third off grass, even on uneven lawns, and those blades will lift up when it encounters anything it shouldn’t be trying to mow through, such as small branches or rocks. It can also mow up slopes of up to 45 degrees.

    The Genie can also be remotely controlled by an app, which is also used to schedule mowing times and can be adapted to your home’s lawn size, seasons, weather conditions and time of day. If it starts to rain, it’ll simply sense this and head back to its docking station.

    The mower is also 4G-enabled, so you can monitor its location in real time and, if it happens to be taken from your yard, you can track its location in real time with the built-in GPS system. You can also set a PIN code to ensure no-one else can operate the Genie.

    It’s creators say it’s passed 2,400 safety tests to date, and comes with a two-year warranty.

    There are several models of the Genie, starting with the 1000 model, which comes with a three-blade disc and has a maximum mowing range of 0.25 acres (1,012 sq m) and a capacity for two hours of work. The top-end 5000 model boasts a five-blade disc and can cover 1.25 acres (5,059 sq m) and will keep powering on for four hours.

    There are some other added extras, like night vision lighting and ride-on edges so it can straddle paths and grass to mow right to the lawn’s edges.

    As part of its Kickstarter campaign, which has already reached nearly 7,000% of its starting pledge, the Genie 1000 is available for US$599, 45% off its MSRP of $1,099. The Genie 3000 comes in at $699 (46% off MSRP) and the 5000 model is $799 (47% off MSRP). There are also a bunch of different add-ons, like extra cutting blades, wall-mounting kits and extended battery warranties.

    You can check out one of its other safety features – its emergency stopping capabilities – in this video below.

    ANTHBOT Genie: Collision Emergency Stop Test

    Source: Kickstarter



    [ad_2]

    Source link