American Arcadia Launches on PlayStation 5, Xbox Series X|S and Nintendo Switch on May 15 Players Can Soon Uncover the Truth Behind Arcadia on All New Platforms STOCKHOLM – April 14, 2025 – Today, Swedish indie publisher Raw Fury and developer Out of the Blue Games announced that their critically acclaimed puzzle game, American Arcadia, is coming to PlayStation 5, Xbox Series X|S and Nintendo Switch on May 15, joining its current availability on PC via Steam.  American Arcadia is a cinematic puzzle game that combines a 2.5D platformer and first-person gameplay to tell the extraordinary tale and experience of Trevor, an average man escaping from the world’s most popular reality TV show. Watch the announcement trailer here. Press Kit can be found here. American Arcadia is set in a 70’s retro-futuristic metropolis where all its citizens enjoy a life of luxury and comfort, where no one suspects that they are playing a role in American Arcadia, a television show that is broadcast live 24/7 and has been on the air for decades.  The carefree life of the citizens are tied to the popularity rate, and as a favorite of the audience, one has nothing to fear, but a drop in popularity comes at a cost: death. For an average man with an uninteresting life like Trevor Hills, that would mean serious trouble, but luckily Trevor is being helped by a mysterious voice which guides him through American Arcadia’s backstage to his freedom. But is this all real, or just another show gimmick to raise audience ratings? KEY FEATURES INCLUDE: Two Experiences in One Game — Control two different characters in two different worlds with unique gameplay, and play as Trevor to flee Arcadia in a 2.5D puzzle-platformer all while exploring and hacking in first-person to help him as Angela. Players must work together to overcome the powerful corporation trying to take them down. Escape a ‘70s Televised Utopia — Experience a gripping new adventure crafted with all the trademark polish, artistry and mystery from the studio behind the award-winning Call of the Sea. Players will discover a rich story and unforgettable characters set in the world’s most dangerous reality TV show and filled with twists, turns, and lies at every turn. Vibrant Characters & Performances — The incomparable voice talents of Yuri Lowenthal, Krizia Bajos and Cissy Jones breathe life into the world and characters of American Arcadia. American Arcadia will launch on PlayStation 5, Xbox Series X|S and Nintendo Switch on May 15, and is currently available to purchase on PC via Steam. To follow along and for more information, check out American Arcadia’s official website here and X page here. ABOUT OUT OF THE BLUE GAMES Out of the Blue Games is an indie studio based in Madrid, Spain, and the developers of Call of the Sea, American Arcadia, and the recently announced The Vigilante Diaries. ABOUT RAW FURY Raw Fury is an indie (un)publisher dedicated to connecting games with players by creating unforgettable experiences. With over 50 games published, including the Kingdom series, Blue Prince, Cassette Beasts, Norco, and Sable, Raw Fury has been consistently recognised as a top 10 publisher in Metacritic’s annual rankings. With an unwavering commitment to being the best partner for indie developers, Raw Fury continues to champion unique games and bring exceptional experiences to players worldwide. Web: rawfury.com X: x.com/rawfury Discord: discord.gg/rawfury BlueSky: rawfury.bsky.social MEDIA CONTACT fortyseven communications on behalf of Raw Fury rawfury@fortyseven.com The post American Arcadia Coming to Consoles on May 15 appeared first on Raw Fury.

Niantic lays off 68 workers following Scopely acquisition Employees were notified on March 20, 2025, layoffs effective from May 20 Image credit: Scopely News by Sophie McEvoy Staff Writer Published on April 16, 2025 Niantic has laid off 68 employees following the $3.5 billion acquisition of the firm by Savvy Games Group subsidiary Scopely. As spotted by Game Developer, Niantic submitted a Worker Adjustment and Retaining Notification (WARN) in California – a process that requires employers to provide workers with 60-days notice before a mass layoff. In the State of California's Employment Development Department's latest WARN report, it details Niantic's decision resulting in the redundancies of 68 people which will be made effective on May 20, 2025. The layoffs were first announced on March 20, following an internal email from Niantic CEO John Hanke that was subsequently published on the firm's website. "Niantic is splitting into two companies, with the acquisition of our games business and team, and the spin off of our technology platform into Niantic Spatial Inc.," Hanke wrote. "As we embark on our new journey as Niantic Spatial, we will need to operate as a startup organisation." "We carefully evaluated the structure we believe is needed for the new company, and after a lot of consideration, it became clear that some roles would not be required given our new focus." Hanke continued: "These decisions are never easy; they in no way reflect the individuals' performance, and we understand their impact on people's lives. We deeply appreciate the talented individuals who helped us get here and for their many contributions, and will support them as they transition to new opportunities."

Not to brag, but I’m somewhat of an expert in folding wagons. Huge flex, I know. But I’ve never tested one with a motor like BougeRV’s new Electric Foldable Wagon.I have a little surf shack that sits almost exactly 1km (0.6 miles) from the parking lot, where I regularly carry supplies and sports gear back and forth. I’ve used lots of folding carts, with prices starting around $40 for something that will only last a few months, $250 for something from a reputable brand, and closer to $800 for top-of-the-line merch. Nothing I’ve tried can make hauling 200kg (about 440 pounds) over steep inclines as effortless as the new BougeRV electric wagon. Yeah, it’s $399.99, and not without its quirks, but boy is it worth it if you regularly carry large heavy loads over hills.7Verge Score$400The GoodFlattens steep hillsHuge carrying capacityFolds for easy transportGood rangeThe BadTerrible displayElectronic brake and throttle useless on downhillsExpensiveCompany offers vague assurance against rain$400 at BougeRVHow we rate and review productsBougeRV — the company behind that impressive lantern I recently reviewed — makes luxury glamping gear for vanlifers and outdoorsy people. Its name sounds a bit too similar to the disparaging slang bougie, if you ask me, but maybe they’re in on the joke?Regardless, this Chinese company’s electric folding wagon has flattened the 5- and 10-percent grade hills (as measured by my Garmin bike computer) that have had me huffing and puffing for years, long before I even entered my wetsuit. At full throttle, the cart follows along under its own power — all I have to do is steer. That alone will justify its price to some. The wagon has a 180-liter capacity that can be expanded to 250 liters when popping open its tailgate, with plenty of tie-down points for your bungee cords. That expansion is surprisingly useful when carrying oversized items like kiteboards or deflated standup paddle boards, for example. It’s sturdy enough to pull a few small kids, a full-sized human sitting cross-legged, or an aging dog and his supplies.The two rear-wheel drive motors are nearly silent and produce 11Nm of torque, which is not much. As such, it still relies upon human power to get it rolling at your choice of low, medium, or high speeds. I found medium and high to best match my usual NYC-style (local, not tourist) walking pace on level surfaces. The cart can also be electronically reversed to help back a heavy load into place.With the tailgate down you can carry odd-sized gear. This load was about 100kg (220 pounds). Photo by Thomas Ricker / The VergeFolds up nice and compact to fit in the trunk of your car or van’s garage. Photo by Thomas Ricker / The VergeThe battery is small but the two motors produce so little torque that it lasts and lasts. Photo by Thomas Ricker / The VergeThis worthless display is on and is completely invisible in sunlight. Photo by Thomas Ricker / The VergeBougeRV says its electric wagons are “typically weather-resistant and can handle light rain or damp conditions, but it’s advisable to avoid prolonged exposure to heavy rain to protect the motor and battery components.” That’s not very reassuring, and the company doesn’t publish an official IP rating against water or dust ingress. I left it outside overnight in light rain and blowing sand and it worked fine the next day. Still, if I were spending $400 on a wagon with such vague assurances, I’d buy a simple cover for it and remove the definitely-not-waterproof battery when storing it outside. Odd that BougeRV doesn’t include a rain cover in the box like wagons that cost 1/10th the price.Operating the cart is a little peculiar. There are two yellow throttle levers on the handle to control the forward and reverse speed, and two red electric sliders you pull to activate the electronic brake. That redundancy lets you control the cart one-handed with the thumb and forefinger when using either your left or right hand to pull the wagon. When walking, it’s best to peg the throttle to max as there’s too much vibration to do any feathering. It takes a few minutes of practice to get comfortable with everything, so it’s a good idea to test drive it before rolling into a crowded area.The handle also features a terrible display that isn’t readable at all in sunlight. A few segments on the display of my review unit don’t light up. There’s also an on/off button and a button to change the speed. The cart turns off automatically after a few minutes of idling and emits warning beeps if you fold it up without turning it off — why not just turn it off when seeing this, BougeRV?Finally, there are — in theory — four braking modes. If you stop walking and naturally push the handle up against the cart, even with the throttle maxed out, the wagon will stop. Drop the handle or pull the red slider, and the wagon stops moving forward. You can also set a mechanical footbrake to prevent it from accidentally rolling away. Everything works as advertised on flat paved surfaces with the cart empty. But things get squirrely after you load it up and put it on a slope.For example, even when packed with a relatively light load of about 20kg (44 pounds), that little red slider that’s supposed to activate the electronic brake does nothing to stop the wagon from rolling down a moderate five percent slope. You have to set the footbrakes to prevent it from rolling after you let go.It’s so unnatural and jerky that I found it easier to shut it off when going down hillsThe throttle can also be unpredictable when rolling downhill, even in the slowest setting. And you have to apply it to move forward or else the motor will brake and add too much drag. It’s so unnatural and jerky that I found it easier to shut it off when going down hills and use the extended handle as a lever to keep the wagon off my heels. That requires stopping and using two hands — one to hold the cart, the other to push the on/off button — which is annoying, but… fine. The wheels performed well over paved surfaces, cobblestones, packed gravel, and concrete covered in a few millimeters of sand. But if that sand piled up like a snow drift, it would stop the wagon in its tracks even at full throttle. I’d prefer actual tires to cut down on vibration and for easier climbs over stones, but the hard plastic wheels likely keep the cost down and reduce maintenance.BougeRV claims a max range of 12 miles (over 19km) from the 133Wh rechargeable battery. The removable battery charges over USB-C and doubles as a power bank to charge USB-A (18W max) and USB-C (40W max) devices.1/11The battery slot doesn’t do a good job of protecting the battery from water and dust (shown here half ejected).I put 5km on the wagon so far, and the display tries to show what looks like “87” percent when I add back the missing display segments. Indeed, the meter on the battery says it’s over 75 percent full. That’s very good from such a small power source considering the size and weight of the loads I’ve been carrying. My testing over ten days covered 1.2km up and down a 5 percent grade, 0.6km up and down a 10 percent grade, and the rest on hard surfaces as flat as the horizon. I’m mostly impressed by BougeRV’s Electric Foldable Wagon and can easily work around its faults. It easily flattens steep hills that normally exhaust me, even when carrying large, heavy loads that would trouble many to pull on flat pavement. It also has plenty of range, so it can be used again and again after you collapse the 18kg (40 pound) wagon and hoist it into a car or van. That will make this bougie wagon’s $400 price tag worth it for many.Photos by Thomas Ricker / The VergeSee More:

Language models predict sequences of words based on vast datasets and are increasingly expected to reason and perform complex linguistic manipulations. Yet, despite their growing sophistication, even powerful models often falter when assigned problems that require step-by-step logic, especially those bound by explicit constraints or structured problem-solving, highlighting their current limitations in applied reasoning. The difficulty arises in generating language that strictly adheres to given conditions. Tasks might specify exact word counts, position of keywords, or thematic constraints, all of which are challenging for models prioritizing probability-based fluency. For example, models often fail to construct a coherent sentence while embedding words at particular locations or composing paragraphs under multiple concurrent requirements. The challenge isn’t just generating relevant content but generating content that rigidly fits a set of formal, predefined rules without compromising fluency. Currently, methods like chain-of-thought prompting attempt to guide models through a reasoning path, but these are limited by their serial execution and expensive inference costs. Parallel approaches such as guess-and-check or best-of-N sampling rely on generating and filtering multiple candidates. Yet, they need separate scoring mechanisms and often yield inconsistent results. These tools improve performance slightly but cannot guarantee the satisfaction of all constraints, especially when models lack an inherent understanding of those constraints. Researchers from MIT and Yale introduced a novel approach named DISCIPL, designed to enable what they term “self-steering” language models. This method defines two roles: a Planner language model, which generates a tailored inference program, and a population of Follower models that execute this program to solve the task. Unlike previous systems, the Planner creates a logic that structures the reasoning process. By separating the planning from execution, the method allows for dynamic and adaptive computation strategies tailored to each task. The inner workings of DISCIPL involve generating inference code using a language called LLAMPPL, which is a Python-based framework for probabilistic programming with language models. The Planner writes code that defines how to explore possible solutions, while Follower models run the code to search for valid outputs. These programs operate by iteratively proposing partial solutions and scoring them based on constraints. The architecture supports multiple inference techniques, including importance sampling, sequential Monte Carlo (SMC), and rejection sampling, which are scalable based on computational budgets. This structured decomposition lets the system reallocate resources to more promising candidates during execution, improving precision and efficiency. In performance evaluations, DISCIPL proved remarkably effective. On the COLLIE benchmark for constrained sentence generation, the Follower model Llama-3.2-1B alone achieved only 4% Pass@1 success. When enhanced with DISCIPL and SMC, performance rose to 87%, surpassing GPT-4o-mini in some instances. The same setup scored as high as 88% Pass@1 for paragraph-level tasks. On a set of difficult real-world tasks called PUZZLES, covering grant writing and itinerary planning, DISCIPL consistently outperformed both the Planner and Follower operating alone. The method also demonstrated high coherency, with average scores around 7.45 out of 10 when using SMC, which starkly contrasts the 9+ scores from more fluent but incorrect outputs produced by baseline methods. Overall, the work introduces a fresh direction in language modeling where models generate answers and devise how they should be computed. By letting the Planner generate code that structures reasoning and Followers execute this code in parallel, the method achieves precision, adaptability, and fluency without requiring larger models or manual engineering. The research’s results illustrate a clear path for enabling smaller language models to outperform their size through intelligent orchestration and self-guided inference. Here is the Paper. Also, don’t forget to follow us on Twitter and join our Telegram Channel and LinkedIn Group. Don’t Forget to join our 90k+ ML SubReddit. [Register Now] miniCON Virtual Conference on AGENTIC AI: FREE REGISTRATION + Certificate of Attendance + 4 Hour Short Event (May 21, 9 am- 1 pm PST) + Hands on Workshop The post MIT Researchers Introduce DISCIPL: A Self-Steering Framework Using Planner and Follower Language Models for Efficient Constrained Generation and Reasoning appeared first on MarkTechPost.

Warning: contains spoilers for Doctor Who episode “The Robot Revolution”. Oh, the infinite promise of the TARDIS. That little blue box can take you through all of time and space, everything that ever happened or ever will, every star that ever was… except when it doesn’t. That is what Belinda Chandra (no Miss) learned in the Doctor Who series 15 opener, when the Doctor tried and failed to take her back to Earth on May 24th 2025 (the airdate of this season’s penultimate episode, if you were wondering whether the finale will be a two-parter). Like many of the Doctor’s companions before her, Belinda will find her journey in the TARDIS is a literal Odyssey – a long and winding journey home interspersed with getting lost and attacked by monsters, and even when you get there, you’ll probably have a fight on your hands. But why is she having such a hard time getting back? It’s Not Just the Doctor’s Driving (This Time) The Doctor’s first companions, Ian and Barbara, took so long to get home simply because the Doctor didn’t know how to drive. Tegan Jovanka had similar trouble getting the Fifth Doctor to drop her off at Heathrow Airport for similar reasons. Even the Thirteenth Doctor’s “Fam”, Yasmin, Graham and Ryan, took a few goes between “The Ghost Monument” and “Arachnids in the UK” to get back to Earth. At the very best of times, the Doctor simply hasn’t always been the best navigator. Or as the TARDIS itself put it in “The Doctor’s Wife”, it takes the Doctor where they need to go, which is not always the same as where they want to go. But this time things are a bit more serious than the Doctor not being able to drive properly, and he’d know that if he ever actually looked at all those instruments on the TARDIS console rather than sticking his head out the door as if he was checking for rain. Because despite landing what he called “halfway” between Planet Missbelindachandraone and Earth, the empty space the TARDIS landed in was filled with debris including a London cab, the Eiffel Tower, an Egyptian Pyramid, and of course, the Statue of Liberty, because have you ever had an apocalyptic event if you didn’t destroy the Statue of Liberty? Something has happened to planet Earth, and on the day that Belinda left it. This is hardly a new problem for the Doctor either. Depending on you count it Doctor Who is 62, 20 or (argue in the comments on the canonical age of the Doctor themself) – at this point they’ve done everything. But when the Doctor tried to return to Earth after it had been whisked away by the Daleks in “The Stolen Earth”, he was still able to land at the right point in time and space where there should have been an Earth. There are bigger problems here than the Earth simply not being there anymore (although the people of Earth might beg to differ). Places the TARDIS cannot go Whether or not the planet Earth is still in place on May 24th 2025 (in Doctor Who I mean), something is blocking the TARDIS from actually getting there. As to what that could be, the answer is – a plot device that Russell T Davies invented to justify keeping Belinda on the TARDIS for the whole series, probably something we’ve never seen before with a name like the Continuum Nexus or a Contemporaneous Schism (not a Time Fracture, the Doctor fixed that one). But if we want to round up the usual suspects, we have got some clues as to things that have stopped the TARDIS getting where it wants to be before. First and foremost, there is the “Time Lock”, because while Russell T Davies will happily drop poetry like “the Could-have-been King with his army of Meanwhiles and Never-weres” one minute, he can get extremely Ronseal Quick Dry and Wood stain the next. The Time Lock is what sealed off the Last Great Time War from the rest of the spacetime continuum, barring anything from getting in or out except the Doctor, a last Dalek, the Dalek Emperor, the Cult of Skaro, Davros and his armies (rescued by Dalek Caan), the Master, and the entire planet of Gallifrey (twice). That might seem like a lot but it’s still pretty good for something containing a war across all of time and space. Join our mailing list Get the best of Den of Geek delivered right to your inbox! So, it could probably contain the single date of May 24th 2025. And if you’re me, or a lot of Doctor Who fans, you’re probably asking “What about May 23rd? Or May 25th? Why not pop back to May 17th, spend a relaxing week in the south of France, then come back refreshed for your next shift at the hospital? Or if you want to be really sure, travel forward to the year 4202 AD, book into the Time Hotel from last year’s Christmas special “Joy to the World”, get Joy’s Room on Boxing Day and sneak out while the Doctor from that episode isn’t looking (he’s still staying there for a year while he waits for another time portal to reopen), then spend five months going backpacking while you wait for pre-May 24th Belinda to get abducted by robots. (While we’re here – if something globally devastating has happened to Earth midway through 2025, why didn’t the Doctor in “Joy the World” notice while he was living at the Sandringham Hotel? Did he take a break from the hotel to go scuba diving with Donna Noble?) If all these workarounds sound familiar, that’s because fans were making the same arguments a few years ago when the Doctor revealed he could not go back to 1930s New York to retrieve Amy and Rory after the Weeping Angels got them. Immediately fans were thinking “He’s immortal, why not go back to the 1920s and wait for a bit? Why not land the TARDIS in Washington and catch a bus? Why not meet up with Amy and Rory in the 1950s and have a catch up?” While the episode itself seems like it leaves these plot threads dangling, on Blogtor Who then-showrunner Steven Moffat felt safer unleashing the more nerdy explanation, “New York would still burn. The point being, he can’t interfere. Here’s the ‘fan answer’ – this is not what you’d ever put out on BBC One, because most people watch the show and just think, ‘well there’s a gravestone so obviously he can’t visit them again’. But the ‘fan answer’ is, in normal circumstances he might have gone back and said, ‘look we’ll just put a headstone up and we’ll just write the book’. But there is so much scar tissue, and the number of paradoxes that have already been inflicted on that nexus of timelines, that it will rip apart if you try to do one more thing. He has to leave it alone. Normally he could perform some surgery, this time too much surgery has already been performed. But imagine saying that on BBC One!” Could we be looking at something similar on May 24th 2025? We already know that there are paradoxes at play. The whole of “The Robot Revolution” is essentially a bootstrap paradox. Belinda tells the robots to kidnap Alan, who they install as Robot King and who then orders them to kidnap Belinda. There is also the question of the star certificate, which somehow travels from Belinda’s bedroom wall to a point thousands of years in the past of planet Missbelindachandraone. Now that certificate has touched its future self, turning Belinda’s ex into a sperm and egg, and that paradox has got it aboard the TARDIS, where I’m sure there is plenty of potential for new paradoxes. Could that build of paradoxes be what is stopping the TARDIS from landing? These theories, the buildup of paradoxes, the time lock used to contain a war, the imminent-yet-somehow-unnoticed-by-the-Doctor destruction of Earth, all point towards the elephant in the room. The title of the final episode of this season of Doctor Who. Due for broadcast on May 31st this year, that episode is called “The Reality War”. The little video ident used to announce it shows us a road sign in Hackney, the brick wall behind it cracking apart in a decisively sci-fi timey wimey sort of a way. It is precisely the sort of title a man who likes Ronseal names like “Time Lock” would come up with if he wanted to one-up something called a “Time War”. And whatever a Reality War is, it seems like Belinda will be in the middle of it. Doctor Who continues with “Lux” on Saturday April 19 on BBC One and iPlayer in the UK, and on Disney+ around the world. Visit Site

Apr 16, 2025Ravie LakshmananMobile Security / Malware Cheap Android smartphones manufactured by Chinese companies have been observed pre-installed with trojanized apps masquerading as WhatsApp and Telegram that contain cryptocurrency clipper functionality as part of a campaign since June 2024. While using malware-laced apps to steal financial information is not a new phenomenon, the new findings from Russian antivirus vendor Doctor Web point to significant escalation where threat actors directly targeting the supply chain of various Chinese manufacturers to preload brand new devices with malicious apps. "Fraudulent applications were detected directly in the software pre-installed on the phone," the company said. "In this case, the malicious code was added to the WhatsApp messenger." A majority of the compromised devices are said to be low-end phones that mimic well-known premium models from Samsung and Huawei with names like S23 Ultra, S24 Ultra, Note 13 Pro, and P70 Ultra. At least four of the affected models are manufactured under the SHOWJI brand. The attackers are said to have used an application to spoof the technical specification displayed on the About Device page, as well as hardware and software information utilities like AIDA64 and CPU-Z, giving users a false impression that the phones are running Android 14 and have improved hardware. The malicious Android apps are created using an open-source project called LSPatch that allows the trojan, dubbed Shibai, to be injected into otherwise legitimate software. In total, about 40 different applications, like messengers and QR code scanners, are estimated to have been modified in this manner. In the artifacts analyzed by Doctor Web, the application hijacks the app update process to retrieve an APK file from a server under the attacker's control and searches for strings in chat conversations that match cryptocurrency wallet address patterns associated with Ethereum or Tron. If found, they are replaced with the adversary's addresses to reroute transactions. "In the case of an outgoing message, the compromised device displays the correct address of the victim's own wallet, while the recipient of the message is shown the address of the fraudsters' wallet," Doctor Web said. "And when an incoming message is received, the sender sees the address of their own wallet; meanwhile, on the victim's device, the incoming address is replaced with the address of the hackers' wallet." Besides changing the wallet addresses, the malware is also fitted with capabilities to harvest device information, all WhatsApp messages, and .jpg, .png, and .jpeg images from DCIM, Pictures, Alarms, Downloads, Documents, and Screenshots folders to the attacker's server. The intention behind this step is to scan the stored images for wallet recovery (aka mnemonic) phrases, allowing the threat actors to gain unauthorized access to victims' wallets and drain the assets. It's not clear who is behind the campaign, although the attackers have been found to leverage about 30 domains to distribute the malicious applications and employ more than 60 command-and-control (C2) servers to manage the operation. Further analysis of the nearly two dozen cryptocurrency wallets used by the threat actors has revealed that they have received more than $1.6 million over the last two years, indicating that the supply chain compromise has paid off in a big way. The development comes as Swiss cybersecurity company PRODAFT uncovered a new Android malware family dubbed Gorilla that's designed to collect sensitive information (e.g., device model, phone numbers, Android version, SIM card details, and installed apps), main persistent access to infected devices, and receive commands from a remote server. "Written in Kotlin, it primarily focuses on SMS interception and persistent communication with its command-and-control (C2) server," the company said in an analysis. "Unlike many advanced malware strains, Gorilla does not yet employ obfuscation techniques, indicating that it may still be under active development." In recent months, Android apps embedding the FakeApp trojan propagated via Google Play Store have also been found making use of a DNS server to retrieve a configuration that contains a URL to be loaded. These apps, since removed from the marketplace, impersonate well-known and popular games and apps and come fitted with the ability to receive external commands that can perform various malicious actions like loading unwanted websites or serving phishing windows. Found this article interesting? Follow us on Twitter  and LinkedIn to read more exclusive content we post. SHARE    