us-technology

Why You Can Trust CNET Our expert, award-winning staff selects the products we cover and rigorously researches and tests our top picks. If you buy through our links, we may get a commission. Reviews ethics statement Xbox Game Pass subscribers can play Gears of War: Reloaded on Day 1. Zachary McAuliffe Staff writer Zach began writing for CNET in November, 2021 after writing for a broadcast news station in his hometown, Cincinnati, for five years. You can usually find him reading and drinking coffee or watching a TV series with his wife and their dog. Expertise Web hosting | Operating systems | Applications | Software Credentials Apple software beta tester, “Helps make our computers and phones work!” – Zach’s grandparents Gears of War is one of the most iconic Xbox series, spawning sequels, spinoffs and books. I fell in love with the series almost 20 years ago when the original landed on the Xbox 360. So I can’t wait to play the remastered edition of the game, called Gears of War: Reloaded, when it launches on Xbox Game Pass on Aug. 26. Xbox Game Pass Ultimate, a CNET Editors’ Choice award pick, offers hundreds of games you can play on your Xbox Series X, Xbox Series S, Xbox One, Amazon Fire TV, smart TV and PC or mobile device for $20 a month. A subscription gives you access to a large library of games, with new ones, like Doom: The Dark Ages, added monthly, plus other benefits such as online multiplayer and deals on non-Game Pass titles. Here are the games Microsoft is adding to Game Pass soon. You can also check out other games the company added to the service recently, like Assassin’s Creed Mirage. Herdling Game Pass Ultimate and PC Game Pass subscribers can play now. Get ready for a grand alpine adventure as you shepherd some giant Calicorns — which look like Appa from Avatar: The Last Airbender — up into a mountain pass. You’ll encounter eerie dangers, puzzles and more on your adventure. And before you ask, yes, you can pet these wonderful creatures. Blacksmith Master (game preview) Game Pass Ultimate and PC Game Pass subscribers can play now. Adventuring in a medieval world can be fun and challenging but it can also be difficult work. If you’d rather run and manage a shop in a medieval setting, give Blacksmith Master a try. You run your own forge and have to manage resource acquisition, production and sale of goods. You’ll build your own forge, craft everything from utensils to weapons and become one of the most skilled craftsmen in the kingdom — all without fighting dragons or other dangerous creatures. Void/Breaker (PC) Game Pass Ultimate and PC Game Pass subscribers can play now. You’re trapped by a ruthless AI in an endless loop where you’re forced to fight hostile enemies, die and then repeat. In this roguelite game, each time you die, you’ll face new challenges. Fast gunplay, quick movements and all-out destruction are your friends in this adrenaline-filled adventure. Who knows? Maybe you’ll get out one day. Goat Simulator Remastered New for Game Pass Standard subscribers. Chaos. That’s the best way to describe this game. You play as a goat and your mission is the total destruction of everything in sight. Can’t stand the look of that car? Run it off the road. Don’t like the town hall’s placement? Demolish it. Want that kid’s ice cream? Eat it. Do what you want in this sandbox-style game where anything is possible.  Game Pass Ultimate subscribers got access to this game in November and Game Pass Standard subscribers can get in on the fun soon, too. Persona 4 Golden Game Pass Ultimate, PC Game Pass and Game Pass Standard subscribers can play now. A string of serial murders set this game in motion, where you’ll travel between the real world and the world within TVs as you try and find out who — or what — is behind the murders. You’ll explore your town and carry on daily activities, as well as travel to the TV World to explore different dungeons and battle monsters in this RPG. Gears of War: Reloaded Game Pass Ultimate and PC Game Pass subscribers can play on Aug. 26. The iconic Gears series is back with a new, remastered edition of the original game. This game maintains the original story of Marcus Fenix and his squad as they take on the alien Locust hordes and it includes a bonus campaign act, every multiplayer map and mode, plus characters and cosmetics unlocked through progression.  This game is also coming to PlayStation 5 and PC, and thanks to cross-play functionality, you’ll be able to play with and against your friends on those platforms. Dragon Age: The Veilguard Game Pass Ultimate and PC Game Pass subscribers can play on Aug. 28. Step into the enchanted world of Thedas, a land full of untamed wilderness, glittering cities and strife. A pair of corrupt ancient gods has broken out of their hold and is wreaking havoc across the globe. It’s up to you to gather a team and stop them, but that’s just the main quest. You can also embark on side quests and companion quests that shed light on who they are and how the world is changing around you. Veilguard changes the Dragon Age formula with a more action-oriented approach, though it retains familiar aspects like branching dialog choices and skill trees. So get ready to dive into a wild adventure, make new friends and become a legend. Games leaving Game Pass on Aug. 31 While Microsoft is adding those games to Game Pass soon, the company is also removing five others from the service on Aug. 31. So you still have some time to finish up your campaign or any side quests before you have to buy these games separately. Ben 10: Power TripBorderlands 3Paw Patrol Mighty Pups Save Adventure BaySea of StarsThis War of Mine: Final Cut For more on Xbox, discover other games available

In the 10 years since releasing its first original film Beasts of No Nation, Netflix has evolved into a power player in the movie scene, earning honors at award shows like the Oscars and Golden Globes.  The streaming giant received 18 Oscar nominations in 2025 and took home three: two for Emilia Perez and one for best documentary short with The Only Girl in the Orchestra. Popular action films and comedies like KPop Demon Hunters, The Old Guard and Hit Man have helped make it a reliable destination for hits.  Whether you love drama, comedy, blockbusters or quiet, independent films, check out what we consider to be the best movies you can watch on Netflix right now. And if a film is leaving soon, we’ll let you know so you can watch it before it gets pulled from the platform — some of our favorite films on the platform, like Mad Max: Fury Road, Sicario and the 2022 indie horror Barbarian, are all leaving soon — catch them before they go! If you’d rather go with a series, peruse our picks for the best TV shows to watch on Netflix. Read more: Where to Watch All the 2024 Oscar Winners Netflix KPop Demon Hunters (2025) The Netflix original animated film KPop Demon Hunters premiered in June and has already risen through the ranks to become the platform’s second most-watched original film of all time. The film, about a KPop group that performs by day and hunts demons in their down time, features tons of great original music and stars Arden Cho, May Hong and Ji Young Yoo as the titular girl group tasked with fighting off a demonic boy band.  Jasin Boland Mad Max: Fury Road (2015) Charlize Theron has become one of the most reliable action stars around, and her turn as Furiosa in Mad Max: Fury Road helped her earn that reputation. The 2015 film from writer-director George Miller is set in a post-apocalyptic desert and is essentially a nonstop battle sequence where hero Max Rockatansky (Tom Hardy) teams up with Theron’s Imperator Furiosa to wage battle against warlord Immortan Joe (Hugh Keays-Byrne) and his army. Nicholas Hoult, Rosie Huntington-Whitely and Zoe Kravitz also star. Catch the film now before it leaves the platform on Sept. 8. Screenshot by Meara Isenberg/CNET Barbarian (2022) Filmmaker Zach Creggers’ latest movie, Weapons, is currently dominating the box office, and if you haven’t seen his debut feature, 2022’s Barbarian, do yourself a favor and catch it before it leaves Netflix on August 31. The film stars Georgina Campbell and Bill Skarsgard as guests at an Airbnb that turns out to be nothing like the advertisement.  Lionsgate Sicario (2015) Sicario, the 2015 thriller featuring Emily Blunt as an FBI special agent tasked with apprehending a lieutenant in a Mexican drug cartel, was the first screenplay penned by Yellowstone showrunner Taylor Sheridan, before the Duttons were even a glimmer in his eye. The film, directed by Denis Villeneuve, co-stars Benicio del Toro, Daniel Kaluuya and Josh Brolin, which adds to its star power. Netflix just revealed that the film will be leaving the platform on Sept. 1 so there are only a few weeks left to watch. Universal Pictures Jurassic Park (1993) With this year’s release of Jurassic World Rebirth, there’s no better time to revisit the original Jurassic Park film directed by Steven Spielberg. The film, which just arrived to Netflix along with The Lost World: Jurassic Park and Jurassic Park III, was an immediate blockbuster when it first came out in 1993 and has become a part of the pop culture lexicon in the years since. (And where would Jeff Goldblum be without it?) Warner Bros. The Departed (2006) It’s almost impossible to choose which of Martin Scorsese’s films is his best, but The Departed is the only one that (finally) won him an Academy Award for best director. Loosely based on real-life Boston criminals like Whitey Bulger, the film stars Jack Nicholson as Frank Costello, a mob boss who plants an informant within the Massachusetts State Police. Matt Damon, Leonardo DiCaprio, Vera Farmiga and Martin Sheen all co-star in the film that just arrived to Netflix this month. Paramount Pictures Clueless (1995) Clueless, which premiered in 1995, has become one of the great teen films of all time. The outfits! The music! The endlessly quotable script! The comedy, which stars Alicia Silverstone as the seemingly vapid LA teen Cher Horowitz, is based loosely on Jane Austen’s novel Emma and arrived to Netflix on Aug. 1; we’re totally buggin’.  Universal Pictures Jaws (1975) It’s a big year for Jaws; the Steven Spielberg classic is celebrating its 50th anniversary and to commemorate, the film (and its three sequels) arrived on Netflix last month. The film, which starred Roy Scheider, Richard Dreyfuss and Robert Shaw as three men on the hunt for a killer shark off the beaches of a New England town, has it all: thrills, impeccable performances and a score by John Williams that might be one of cinema’s most memorable ever.  Netflix The Old Guard (2020) Based on the comic book by Greg Rucka, the action-fantasy The Old Guard stars Charlize Theron as one of a group of immortal mercenaries who are able to regenerate even after being killed. When they’re tracked down and hunted by a pharmaceutical company that intends to study them, they fight back in an effort to protect themselves. The first film, released in 2020, was such a success that a sequel, The Old Guard 2, premiered earlier this summer and features Uma Thurman as Theron’s newest adversary. David James/Paramount Pictures Mission: Impossible – Rogue Nation (2015) Netflix just released the first five movies in the Mission: Impossible franchise and while we really enjoy them all, 2015’s Rogue Nation is one of our favorites. Rogue Nation was the first film in the franchise to be directed by Tom Cruise’s frequent collaborator Christopher McQuarrie and it’s the film that brings Cruise’s Ethan Hunt together with ex-MI6 agent Ilsa Faust,

Plus: Google’s new phones are AI-first This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. On the ground in Ukraine’s largest Starlink repair shop Starlink is absolutely critical to Ukraine’s ability to continue in the fight against Russia. It’s how troops in battle zones stay connected with faraway HQs; it’s how many of the drones essential to Ukraine’s survival hit their targets; it’s even how soldiers stay in touch with spouses and children back home. However, Donald Trump’s fickle foreign policy and reports suggesting Elon Musk might remove Ukraine’s access to the services have cast the technology’s future in the country into doubt. For now Starlink access largely comes down to the unofficial community of users and engineers, including the expert “Dr. Starlink”—famous for his creative ways of customizing the systems—who have kept Ukraine in the fight, both on and off the front line. Together, they have repaired and customized more than 15,000 terminals since the war began. Despite the pressure, the chance that they may lose access to Starlink was not worrying volunteers at the time of my visit; in our conversations, it was clear they had more pressing concerns than the whims of a foreign tech mogul. Russia continues to launch frequent aerial bombardments of Ukrainian cities, sometimes sending more than 500 drones in a single night.  The threat of involuntary mobilization to the front line looms on every street corner. How can one plan for a hypothetical future crisis when crisis defines every minute of one’s day? Read the full story. —Charlie Metcalfe This story is from our forthcoming print issue, which is all about security. If you haven’t already, subscribe now to receive future issues once they land. This article is also part of the Big Story series: MIT Technology Review’s most important, ambitious reporting. The stories in the series take a deep look at the technologies that are coming next and what they will mean for us and the world we live in. Check out the rest of them here. NASA’s new AI model can predict when a solar storm may strike NASA and IBM have released a new open-source machine learning model to help scientists better understand and predict the physics and weather patterns of the sun. Surya, trained on over a decade’s worth of NASA solar data, should help give scientists an early warning when a dangerous solar flare is likely to hit Earth. Solar storms occur when the sun erupts energy and particles into space. They can produce solar flares and slower-moving coronal mass ejections that can disrupt radio signals, flip computer bits onboard satellites, and endanger astronauts with bursts of radiation.  While there’s no way to prevent these sorts of effects, being able to predict when a large solar flare will occur could let people work around them. Read the full story. —Peter Hall Why recycling isn’t enough to address the plastic problem I remember using a princess toothbrush when I was little. The handle was purple, teal, and sparkly. Like most of the other pieces of plastic that have ever been made, it’s probably still out there somewhere, languishing in a landfill. (I just hope it’s not in the ocean.) I’ve been thinking about that toothbrush again this week after UN talks about a plastic treaty broke down on Friday. Nations had gotten together to try and write a binding treaty to address plastic waste, but negotiators left without a deal. Plastic is widely recognized as a huge source of environmental pollution—again, I’m wondering where that toothbrush is—but the material is also a contributor to climate change. Let’s dig into why talks fell apart and how we might address emissions from plastic. —Casey Crownhart This article is from The Spark, MIT Technology Review’s weekly climate newsletter. To receive it in your inbox every Wednesday, sign up here. The must-reads I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology. 1 Google is betting that AI can help you take better photos So long as you don’t try zooming in on someone’s face, that is. (WP $)+ Gemini is getting a new audio model capable of detecting tone. (TechCrunch)+ Google’s AI efforts are certainly outpacing those of its hardware rival Apple. (Bloomberg $) 2 Meta’s AI hiring spree is on pauseInvestors are increasingly concerned by the mad sums being bandied about. (WSJ $) 3 China is preparing to show off its hypersonic missilesThe world will be watching its military parade closely next month. (FT $)+ Meanwhile, India has tested a missile that could hit deep into China. (The Guardian)+ Taiwan’s “silicon shield” could be weakening. (MIT Technology Review) 4 RFK Jr. wants to send you MAHA food boxesBut concrete details are light on the ground. (The Atlantic $)+ How MAHA is shaking up packaged goods’ supply chains. (Fortune $) 5 Extreme heat is driving cases of Legionnaire’s disease in NYCOlder air conditioning infrastructure is helping to spread dangerous bacteria. (Vox)+ A fifth person has died in connection with the current outbreak. (ABC News) 6 What it’s like to vibecode for a massive startupManaging AI coding apps is a whole lot like herding interns, supposedly. (Wired $)+ What is vibe coding, exactly? (MIT Technology Review) 7 Starship’s rocket launch could delay flights in FloridaEven after the launch is completed. (TechCrunch) 8 This app will help you find the sunniest spots in Paris ☀️The community-driven Jveuxdusoleil is updated in real time. (The Guardian) 9 The world’s only public diamond mine is in Arkansas 💎Visitors have unearthed more than 35,000 precious gems since it opened. (Ars Technica) 10 It turns out Uranus had a hidden moon all alongAnd many more may be discovered in the future. (NYT $)+ It’s the 29th known satellite to orbit the planet. (Scientific American $)+ The moon is just the beginning for this waterless concrete. (MIT Technology Review) Quote of the day “It jumbles my freaking nugget that people can look at

Google has just released a technical report detailing how much energy its Gemini apps use for each query. In total, the median prompt—one that falls in the middle of the range of energy demand—consumes 0.24 watt-hours of electricity, the equivalent of running a standard microwave for about one second. The company also provided average estimates for the water consumption and carbon emissions associated with a text prompt to Gemini. It’s the most transparent estimate yet from a Big Tech company with a popular AI product, and the report includes detailed information about how the company calculated its final estimate. As AI has become more widely adopted, there’s been a growing effort to understand its energy use. But public efforts attempting to directly measure the energy used by AI have been hampered by a lack of full access to the operations of a major tech company.  Earlier this year, MIT Technology Review published a comprehensive series on AI and energy, at which time none of the major AI companies would reveal their per-prompt energy usage. Google’s new publication, at last, allows for a peek behind the curtain that researchers and analysts have long hoped for. The study focuses on a broad look at energy demand, including not only the power used by the AI chips that run models but also by all the other infrastructure needed to support that hardware.  “We wanted to be quite comprehensive in all the things we included,” said Jeff Dean, Google’s chief scientist, in an exclusive interview with MIT Technology Review about the new report. That’s significant, because in this measurement, the AI chips—in this case, Google’s custom TPUs, the company’s proprietary equivalent of GPUs—account for just 58% of the total electricity demand of 0.24 watt-hours.  Another large portion of the energy is used by equipment needed to support AI-specific hardware: The host machine’s CPU and memory account for another 25% of the total energy used. There’s also backup equipment needed in case something fails—these idle machines account for 10% of the total. The final 8% is from overhead associated with running a data center, including cooling and power conversion.  This sort of report shows the value of industry input to energy and AI research, says Mosharaf Chowdhury, a professor at the University of Michigan and one of the heads of the ML.Energy leaderboard, which tracks energy consumption of AI models.  Estimates like Google’s are generally something that only companies can produce, because they run at a larger scale than researchers are able to and have access to behind-the-scenes information. “I think this will be a keystone piece in the AI energy field,” says Jae-Won Chung, a PhD candidate at the University of Michigan and another leader of the ML.Energy effort. “It’s the most comprehensive analysis so far.” Google’s figure, however, is not representative of all queries submitted to Gemini: The company handles a huge variety of requests, and this estimate is calculated from a median energy demand, one that falls in the middle of the range of possible queries. So some Gemini prompts use much more energy than this: Dean gives the example of feeding dozens of books into Gemini and asking it to produce a detailed synopsis of their content. “That’s the kind of thing that will probably take more energy than the median prompt,” Dean says. Using a reasoning model could also have a higher associated energy demand because these models take more steps before producing an answer. This report was also strictly limited to text prompts, so it doesn’t represent what’s needed to generate an image or a video. (Other analyses, including one in MIT Technology Review’s Power Hungry series earlier this year, show that these tasks can require much more energy.) The report also finds that the total energy used to field a Gemini query has fallen dramatically over time. The median Gemini prompt used 33 times more energy in May 2024 than it did in May 2025, according to Google. The company points to advancements in its models and other software optimizations for the improvements.   Google also estimates the greenhouse gas emissions associated with the median prompt, which they put at 0.03 grams of carbon dioxide. To get to this number, the company multiplied the total energy used to respond to a prompt by the average emissions per unit of electricity. Rather than using an emissions estimate based on the US grid average, or the average of the grids where Google operates, the company instead uses a market-based estimate, which takes into account electricity purchases that the company makes from clean energy projects. The company has signed agreements to buy over 22 gigawatts of power from sources including solar, wind, geothermal, and advanced nuclear projects since 2010. Because of those purchases, Google’s emissions per unit of electricity on paper are roughly one-third of those on the average grid where it operates. AI data centers also consume water for cooling, and Google estimates that each prompt consumes 0.26 milliliters of water, or about five drops.  The goal of this work was to provide users a window into the energy use of their interactions with AI, Dean says.  “People are using [AI tools] for all kinds of things, and they shouldn’t have major concerns about the energy usage or the water usage of Gemini models, because in our actual measurements, what we were able to show was that it’s actually equivalent to things you do without even thinking about it on a daily basis,” he says, “like watching a few seconds of TV or consuming five drops of water.” The publication greatly expands what’s known about AI’s resource usage. It follows recent increasing pressure on companies to release more information about the energy toll of the technology. “I’m really happy that they put this out,” says Sasha Luccioni, an AI and climate researcher at Hugging Face. “People want to know what the cost is.” This estimate and the supporting report contain more public information than has been available before, and it’s

I remember using a princess toothbrush when I was little. The handle was purple, teal, and sparkly. Like most of the other pieces of plastic that have ever been made, it’s probably still out there somewhere, languishing in a landfill. (I just hope it’s not in the ocean.) I’ve been thinking about that toothbrush again this week after UN talks about a plastic treaty broke down on Friday. Nations had gotten together to try and write a binding treaty to address plastic waste, but negotiators left without a deal. So, in March 2022, the UN Environment Assembly set out to develop an international treaty to address plastic pollution. Pretty much everyone should agree that a bunch of plastic waste floating in the ocean is a bad thing. But as we’ve learned over the past few years, as these talks developed, opinions diverge on what to do about it and how any interventions should happen. One phrase that’s become quite contentious is the “full life cycle” of plastic. Basically, some groups are hoping to go beyond efforts to address just the end of the plastic life cycle (collecting and recycling it) by pushing for limits on plastic production. There was even talk at the Assembly of a ban on single-use plastic. Petroleum-producing nations strongly opposed production limits in the talks. Representatives from Saudi Arabia and Kuwait told the Guardian that they considered limits to plastic production outside the scope of talks. The US reportedly also slowed down talks and proposed to strike a treaty article that references the full life cycle of plastics. Petrostates have a vested interest because oil, natural gas, and coal are all burned for energy used to make plastic, and they’re also used as raw materials. This stat surprised me: 12% of global oil demand and over 8% of natural gas demand is for plastic production.   That translates into a lot of greenhouse gas emissions. One report from Lawrence Berkeley National Lab found that plastics production accounted for 2.24 billion metric tons of carbon dioxide emissions in 2019—that’s roughly 5% of the global total.   And looking into the future, emissions from plastics are only set to grow. Another estimate, from the Organisation for Economic Co-operation and Development, projects that emissions from plastics could swell from about 2 billion metric tons to 4 billion metric tons by 2060. This chart is what really strikes me and makes the conclusion of the plastic treaty talks such a disappointment. Recycling is a great tool, and new methods could make it possible to recycle more plastics and make it easier to do so. (I’m particularly interested in efforts to recycle a mix of plastics, cutting down on the slow and costly sorting process.) But just addressing plastic at its end of life won’t be enough to address the climate impacts of the material. Most emissions from plastic come from making it. So we need new ways to make plastic, using different ingredients and fuels to take oil and gas out of the equation. And we need to be smarter about the volume of plastic we produce.   One positive note here: The plastic treaty isn’t dead, just on hold for the moment. Officials say that there’s going to be an effort to revive the talks. Less than 10% of plastic that’s ever been produced has been recycled. Whether it’s a water bottle, a polyester shirt you wore a few times, or a princess toothbrush from when you were a kid, it’s still out there somewhere in a landfill or in the environment. Maybe you already knew that. But also consider this: The greenhouse gases emitted to make the plastic are still in the atmosphere, too, contributing to climate change.  This article is from The Spark, MIT Technology Review’s weekly climate newsletter. To receive it in your inbox every Wednesday, sign up here. Read More

The latest addition to China’s Tiangong space station is an AI chatbot with expertise in navigation and tactical planning. Named Wukong AI—after the protagonist of the “Monkey King” legend in Chinese mythology, Sun Wukong—the chatbot was introduced on the space station in mid-July, and has already completed its first mission: supporting three taikonauts during a spacewalk. Information about Wukong AI remains limited. Chinese authorities have said that they developed it from a domestic open-source AI model; according to Xinhua, China’s state-run news agency, engineers designed it to meet the requirements of manned space missions, and focused its knowledge-base on aerospace flight data. “This system can provide rapid and effective information support for complex operations and fault handling by crew members, improving work efficiency, in-orbit psychological support, and coordination between space and ground teams,” Zou Pengfei of the taikonaut training center, told Xinhua. Technicians connected the AI to Tiangong on July 15. It began providing support a month later, this being the first time that China’s space station has used a large language model (LLM) during in-orbit missions. Wukong AI assisted the crew on a six-and-a-half-hour mission, which involved taikonauts installing space debris protection devices during a spacewalk and performing a routine inspection of the station. The taikonauts claim that their new assistant “offers very comprehensive content.” Chinese media describe Wukong AI as a classic question-and-answer system divided into two modules: one installed on the station, and one on Earth. The ground module performs in-depth analysis, while the module accompanying the crew solves immediate challenges. The combination of the two creates an advanced assistant capable of adapting to each mission. Wukong is neither the first AI system in space nor the first on a station. The International Space Station already has Astrobee, a robot that assists astronauts with routine tasks, and CIMON, a conversational psychological support system. The particularity of Wukong AI is that it combines the functions of an intelligent assistant—like those used on Earth—with a total focus on space navigation. The Tiangong station is the core of China’s strategy to consolidate its position as a space power over the next 30 years. The station currently serves as a microgravity laboratory for experiments that would be impossible on Earth. In the future, China plans to expand it and turn it into an intermediate logistics and training platform between the moon and the Earth’s surface. And the reason for the AI’s name? Sun Wukong is a mythical person who appears in the classic novel Journey to the West. In China, he symbolizes cunning, adaptability, endurance, and the pursuit of knowledge. This story originally appeared on WIRED en Español and has been translated from Spanish. Read More

And that’s precisely what happens. A quick blowtorch to the back frame loosens the bolts, and Pala is able to swap the original part for a state-of-the-art titanium rear triangle in just a few minutes. Instead of the original hefty old Sturmy-Archer three-speed gearing, Brompton has developed its own new 12-speed system, which has a mini three-speed cassette on the outside, and three in the internal hub. “Because the dimensions are the same, we’re going to be able to strip the bike back completely,” enthuses Pala. “We’re then going to rebuild it around the main frame and fork.” Now, while WIRED appreciates this is going to be a Brompton stowed safely on the Ship of Theseus, it remains an impressive, decade-spanning example of good design, and the importance of considered, rather than reactionary upgrades. Before… Photograph: Brompton … after. Photograph: WIRED Staff This sort of advanced-level repairability is par for the course for Brompton, and if you have your own tired old folding bike, every nut, bolt, bracket, and accessory is available to order. If bike maintenance isn’t something you’re comfortable with, there’s a detailed list of all Brompton stores and authorized dealers on the company’s website, with around 150 across the USA. In the UK, especially around London, there’s plenty of scope for repairs, and Brompton offers a complete service at one of its dealerships for £295 (less than $400.) It’s this familiarity of design and attention to detail that has transformed the bicycle company into something of a global luxury brand. It’s a bike for people who wouldn’t necessarily call themselves cyclists, with a uniform in some countries that’s more Lacroix than lycra. “We’re very globally spread out,” says Will Carleysmith, Brompton’s chief design and engineering officer. “The UK is our most commuter-focused audience, but it represents just 16 percent of our business—the rest of what we make goes overseas, with China taking 40 percent of our sales, interestingly with a 50-50 male/female split.” In Asia, the Brompton is viewed quite differently than in the UK, where it’s typically seen as a practical tool for urban commuting. “It’s a super social, highly desirable tool that’s much more about self-expression,” claims Carleysmith. Collaborations are helping to underline this “style” narrative, too, with the likes of Barbour, Palace Skateboarding, Liberty London, Tour de France, LINE Friends, and art collaborations with Crew Nation and cultural luminaries including Radiohead, Phoebe Bridgers, and LCD Soundsystem. But like any good 50th birthday, there have been both happy and sad tears. In 2022, Brompton sold its 1 millionth bike. During Covid lockdown, demand increased five-fold, but as a result of supply chain and shipping bottlenecks, rising costs, and heavy investment in new designs, pretax profits plunged by 99 percent to just $6,335 (£4,602, or roughly the cost of a single Brompton T-Line One Speed) for the year ending March 31, 2024. It’s not the only cycling brand faced with post-Covid cash flow issues, but rather than being stuck with excess stock, its financial woes have been thanks to a global drop in demand and heavy investment, first with the Brompton Electric range and the bigger, 20-inch all-terrain G Line, which WIRED tested at launch. Read More

“It’s just very sad, and it’s kind of pointless,” Rader says. “And I think they’re going to look back at it in a couple of years, maybe less, and go, ‘Oh my gosh, what did we do?’” No one I spoke to for this piece thinks NASA is literally going away. For one thing, Congress is pushing back on the changes, though the administration seems determined to ram them through one way or another. Instead, what they imagine is a kind of rump agency. “The sense that I got was, it was a very real possibility that NASA could be reduced to something just kind of in name only,” Rader says. “Almost maybe a version of the FAA (the Federal Aviation Administration), but for space.” What’s being undercut isn’t just NASA’s technical ability to carry out missions, although that would be bad enough. It is America’s—and the world’s—capacity to wonder, to believe, to know. “It’s almost like a diminution of our own vision and ambition to say we’re literally, I mean, again, not figuratively, literally, closing our eyes to the cosmos and turning inwards,” says Casey Dreier, the space policy chief at the nonprofit Planetary Society. “It’s like witnessing a death of an ideal.” That death is already underway. Around 4,000 NASA staffers are scheduled to leave the agency this year, either through what the Trump administration calls “deferred resignation”—a kind of delayed, voluntary layoff—or what NASA is branding “normal attrition,” which includes people like Rader who are leaving of their own accord. That represents about a quarter of the agency’s total staff and includes more than 2,000 senior leaders, according to a report in Politico. (In a statement, Cheryl Warner, NASA’s news chief, said safety “remains a top priority for our agency as we balance the need to become a more streamlined and more efficient organization and work to ensure we remain fully capable of pursuing a Golden Era of exploration and innovation, including to the moon and Mars.”) The administration, meanwhile, has proposed a 2026 NASA budget that would slash overall agency spending by 24 percent and science spending specifically by almost half. “This is the largest single-year cut as a percentage ever proposed to NASA,” Dreier says. “It would bring NASA’s overall resources, adjusted for inflation, down to a level not seen since before the first humans went into space in 1961.” The Trump proposal projects a frozen NASA budget until at least 2030 even as the administration touts a new “golden age of innovation and exploration.” To cap it off, NASA has been without a full-time administrator—the agency’s top official—since January. Sean Duffy, the transportation secretary and a former champion lumberjack and Real World cast member, has been doing double duty in the role on an interim basis since July. Much has been written about what the proposed budget cuts and job losses will do to NASA. To begin with, they would mean the end of 41 planned or current missions, according to the Planetary Society. Those include an audacious, and long-underway plan to gather pristine soil samples on Mars and return them to earth, a probe exploring the solar system beyond Pluto, and a lander set to catch and study a giant asteroid that will barely miss the earth in 2029. They would also force NASA to essentially get out of the business of tracking climate change. Read More

I asked my editors if I could go work at a tech startup. It was an unusual request. But I wanted to learn to vibe-code. My need to know felt urgent. I wanted to survive the future. The pitch process was surprisingly easy: First my editors said yes, and then the tech startup I lobbed my wild idea to, Notion, agreed to let me embed with them. Why? It’s hard to say. Possibly because Notion’s own workforce has fully embraced vibe coding—“vibe” here being a euphemism for “AI-assisted.” Some tech companies have estimated that around 30 to 40 percent of their code is now written by AI. Notion is a 1,000-person, venture-backed San Francisco startup with a $10 billion valuation. It makes the ultimate to-do and note-taking app, consisting of so many templates and tables and ways to format tasks that figuring out how to use Notion is a task in itself. On YouTube, productivity gurus attempt to make sense of Notion using the well-worn vernacular of personal optimization. One such video is titled “How to Get Started in Notion Without Losing Your Mind.” It has 3.4 million views. I was scheduled to start at Notion as a vibe-coding engineer on a Thursday in mid-July. The night before, I found myself panic-watching these YouTube videos. Surely I would need to be a power user of the app if Notion was allowing me—an English major!—to fiddle with its code base. In an earlier onboarding call, a new coworker had encouraged me to download the AI coding platform Cursor and play around with it. I did. No real code emerged from this homework. My desk on my first day at Notion. Photograph: Lauren Goode Fortunately, I would be pair-programming at Notion, which meant that I’d be working alongside experienced (human) coders. Upon my arrival, Sarah Sachs, an AI engineering lead at Notion, set me up at a desk. A company tote bag and notebook awaited me. Sachs informed me that the following day, I would be presenting my work to the staff at a weekly demo meeting. Was I good with that? I said yes. We were all committed to the bit. Sitting a few feet away was Simon Last, one of Notion’s three cofounders. He is gangly and shy, an engineer who has relinquished management responsibilities to focus on being a “super IC”—an individual contributor. He stood to shake my hand, and I awkwardly thanked him for letting me vibe-code. Simon returned to his laptop, where he was monitoring an AI as it coded for him. Later, he would tell me that using AI coding apps was like managing a bunch of interns. Since 2022, the Notion app has had an AI assistant to help users draft their notes. Now the company is refashioning this as an “agent,” a type of AI that will work autonomously in the background on your behalf while you tackle other tasks. To pull this off, human engineers need to write lots of code. They open up Cursor and select which of several AI models they’d like to tap into. Most engineers I chatted with during my visit preferred Claude, or they used the Claude Code app directly. After choosing their fighter, the engineers ask their AI to draft code to build a new thing or fix a feature. The human programmer then debugs and tests the output as needed—though the AIs help with this too—before moving the code to production. At its foundational core, generative AI is enormously expensive. The theoretical savings come in the currency of time, which is to say, if AI helped Notion’s cofounder and CEO Ivan Zhao finish his tasks earlier than expected, he could mosey down to the jazz club on the ground floor of his Market Street office building and bliss out for a while. Ivan likes jazz music. In reality, he fills the time by working more. The fantasy of the four-day workweek will remain just that. My workweek at Notion was just two days, the ultimate code sprint. (In exchange for full access to their lair, I agreed to identify rank-and-file engineers by first name only.) My first assignment was to fix the way a chart called a mermaid diagram appears in the Notion app. Two engineers, Quinn and Modi, told me that these diagrams exist as SVG files in Notion and, despite being called scalable vector graphics, can’t be scaled up or zoomed into like a JPEG file. As a result, the text within mermaid diagrams on Notion is often unreadable. Quinn slid his laptop toward me. He had the Cursor app open and at the ready, running Claude. For funsies, he scrolled through part of Notion’s code base. “So, the Notion code base? Has a lot of files. You probably, even as an engineer, wouldn’t even know where to go,” he said, politely referring to me as an engineer. “But we’re going to ignore all that. We’re just going to ask the AI on the sidebar to do that.” His vibe-coding strategy, Quinn explained, was often to ask the AI: Hey, why is this thing the way it is? The question forces the AI to do a bit of its own research first, and the answer helps inform the prompt that we, the human engineers, would write. After “thinking,” Cursor informed us, via streaming lines of text, that Notion’s mermaid diagrams are static images that, among other things, lack click handlers and aren’t integrated with a full-screen infrastructure. Sure. Using Claude’s notes, I wrote up the request and pasted some notes from the engineering team into Cursor, like this: Ticket: Add Full Screen / Zoom to mermaid diagrams. Clicking on the diagram should zoom it in full screen. Notes from slack: “mermaid diagrams should be zoom / fullscreenable like uploaded images. they’re just svgs right, so we can probably svg -> dataurl -> image component if we want to zoom” We waited. Time is inverted in the land of vibes. Projects that used to

A new pocket-sized Starlink alternative promises secure military communications — safe from interference by billionaire CEOs. The system, named the RU1, was unveiled today by Swedish startup TERASi. It’s billed as the world’s smallest and lightest mm-Wave radio, a form of communications that offers blazing-fast speeds and huge bandwidth. James Campion, the CEO and co-founder of TERASi, describes the portable device as “the GoPro of backhaul radios.” “RU1 can be deployed in minutes to keep units connected in fast-changing environments,” Campion told TNW. The devices, he continued, can be installed on tripods or drones. Multiple RU1s can then link into a resilient mesh, providing bandwidth for mission-critical applications such as live drone video, autonomous fleet control, and sensor data fusion. It also can’t be remotely disabled or controlled by external actors — a safeguard notably absent from SpaceX’s Starlink. That vulnerability has been strikingly exposed in Ukraine. Military control conflicts Just days after Russia’s full-scale invasion began in February 2022, Elon Musk, the CEO of SpaceX, agreed to supply Ukraine with Starlink. The satellite internet service quickly became indispensable, keeping Ukraine’s military and civilian systems online despite relentless Russian attacks. Yet it has also been restricted at crucial moments. In the autumn of 2022, Musk ordered cut coverage during a Ukrainian counteroffensive in Kherson. The move disrupted surveillance drones, artillery targeting, and troop coordination, according to a Reuters investigation.  Later that year, Musk refused a request to activate Starlink near Crimea for a naval drone strike. He was also allegedly asked by Vladimir Putin to limit coverage over Taiwan — as a favour to Xi Jinping. These episodes underscored the dangers of a private operator maintaining control of military communications. TERASi says the RU1 removes that risk. “The RU1 gives users complete control over their communications by creating a secure, high-speed network that they own and operate themselves, without input from third-party providers like Starlink that can be switched off or restricted remotely, as the 2022 incident in Ukraine showed all too clearly,” said Campion. Battlefield performance TERASi, a spinout from Stockholm’s KTH Royal Institute of Technology, also distinguishes the RU1 from Starlink in performance. The device uses highly focused antennas with very narrow, “laser-like” beams that are extremely difficult to jam or intercept. The beams reduce interception risks by creating small ground footprints of less than 3km. Starlink, by contrast, covers areas of around 1,000 km using lower-frequency radio waves, which Campion argues makes it “much more interceptable.” On performance, TERASi claims RU1 supports data rates of up to 10 Gbps — 50 times faster than Starlink. Future versions promise 20 Gbps, providing a true wireless alternative to fibre. Latency, meanwhile, is below 5 milliseconds, which is over five times quicker than Starlink, according to Campion. “This is crucial for rapid response in dynamic scenarios such as drone detection,” he said. The tech isn’t only for military operations. TERASi envisions it providing uninterrupted, high-speed communications in various hostile or remote environments that struggle with traditional infrastructure. In disaster relief, it could instantly restore gigabit links for first responders without waiting on satellites or fibre repairs. In industry, it could enable temporary high-capacity networks for remote construction, mining, or energy sites. For militaries, the RU1 is already available for evaluation by defence units. TERASi said it’s currently being integrated into systems with tactical communications providers and drone makers.  Campion believes the device offers different strengths from Starlink. “Satellite communication services like Starlink offer wide area coverage that is useful for connecting static, low data rate sensors and devices to a global network,” he said.  “RU1 gives users control over their data and the freedom to build sovereign networks on-the-fly, changing the frontline paradigm from waiting on infrastructure to creating it instantly, from depending on external actors to self-sufficiency.” Even if it can’t match Starlink’s global scale, the RU1’s mix of speed, security, simplicity, and sovereignty could prove compelling — especially when a single CEO can cut the alternative at will. Read More