Nintendo Switch 2 preorders are currently underway (if you can manage to find a store with inventory, that is). As we approach the console's June 5 launch date, it's worth getting a bit reflective: The Switch as a product line is now more than eight years old, and Nintendo has made some significant changes with this latest device. If you have a Switch and are thinking about picking up a Switch 2, you might be wondering how these two generations of consoles compare.Now, Nintendo didn't just release one console during the Switch era: It released three. There's the original Switch, the Switch OLED, and the Switch Lite. I'm taking a look at all three and comparing their specs to the new Switch 2. While we'll need to wait for reviews to directly compare the overall experience of these different devices, we can at least get a sense for them on paper. DimensionsRight off the bat, all of Nintendo's Switches feel very different. The Switch 2 is the largest, coming in at 4.5 inches tall by 10.7 inches wide by 0.55 inches thick. That's with the Joy-Con 2s attached, mind you. Nintendo says the thickness from the tip of the joysticks to the ZL/ZR buttons is 1.2 inches. It weighs 0.88 lbs on its own, or 1.18 lbs with the Joy-Cons attached.The original Switch is smaller: It's 4 inches by 9.4 inches by 0.55 inches. (The joysticks to the ZL/ZR buttons is 1.12 inches.) It also weighs a little less: 0.66 lbs without Joy-Cons, or 0.88 with. It's interesting to note that the Switch 2's display is equal to the weight of the original Switch with Joy-Cons attached.The Switch OLED is quite similar to the original Switch in footprint: It's 4 inches by 9.5 inches by 0.55 inches, with the same Joy-Con dimensions as the OG Switch. However, it's right in the middle when it comes to weight: 0.71 lbs without Joy-Cons, and 0.93 with.The Switch Lite is the smallest of all: The handheld is 3.6 inches by 8.2 inches by 0.55 (the joysticks and ZL/ZR buttons add the same 1.12 inches, despite being built into the unit) and it weighs 0.61 lbs. DisplayThe Switch 2 has the largest screen of any Nintendo device ever. It comes with a 7.9 inch LCD, with a resolution of 1920 x 1080. It supports HDR10, and has a VRR (variable refresh rate) of up to 120Hz. The original Switch also uses LCD for the screen, but its display is much smaller: The OG has a 6.2-inch screen with a resolution of 1280 x 720. There is no HDR support, and the display supports a refresh rate of 60Hz. It's the same situation with the Switch Lite, only it has an even smaller 5.5-inch LCD (this is the most portable Switch, after all). The Switch OLED, of course, has an OLED display. That means that each of the pixels on this Switch's display can be shut off individually, which allows for much deeper blacks than an LCD can reproduce. Still, the Switch OLED's display is 7 inches, nearly a full inch smaller than the Switch 2, with no HDR support. It's also the same 720p resolution as the original Switch, so you won't see as much detail as compared to Nintendo's latest console. Video output and dockNintendo limits the Switch's resolution in handheld mode, likely for battery reasons. But when docked, you can push the console further. For example, the Switch 2 supports a maximum resolution of 4K (3840 x 2160) at 60 fps when connected to a TV. If you want to boost your game's frame rate to 120 fps (assuming the game supports it) you'll need to drop the resolution to at least 1440p, but you can also use 1080p if you like. The OG Switch and Switch OLED both support the same 1080p maximum resolution when docked. The Switch Lite doesn't support video output, since it's meant to be played exclusively in handheld mode. In order to output to a TV, you need a dock. The Switch 2 dock is quite similar to the OLED Switch dock: It's slightly larger, but comes with the same dual USB 2.0 ports, as well as a LAN port. The original Switch dock came with an extra USB 2.0 port, but no LAN port. The biggest difference here, however, is the Switch 2's dock now has a fan for active cooling, which makes sense for some of those more intense game play modes.Joy-ConsThe Switch's detachable controllers are called "Joy-Cons," and they're an iconic component of this particular Nintendo era.The new Joy-Cons, called Joy-Con 2, are 4.57-inches tall, 0.56 inches wide, and 1.2 inches thick. The left Joy-Con is 2.3 ounces, while the right is 2.4 ounces. That might have something to do with the right Joy-Cons "C-Button," which you can use for the new GameChat feature. Nintendo's official spec page for its devices appears to be wrong for the original Joy-Cons, so I've sourced it from this specific Joy-Con page: The original Joy-Cons are 4.02 inches by 1.41 inches by 1.12 inches. The left Joy-Con is 1.73 ounces, while the right is 1.84 ounces.While all Joy-Cons support HD Rumble (Nintendo's brand name for haptic feedback), the new Joy-Cons support HD Rumble 2, a newer standard. Most notably, however, the Joy-Con 2s support Mouse Mode, a feature that lets you use the Joy-Cons on a flat surface like a computer mouse.You also don't connect the Joy-Con 2s to the Switch 2 the same way as you do the originals: These connect magnetically (yes, they're strong magnets.) While you don't slide the Joy-Cons into rails like you do on the OG Switch, you do still release them via buttons on the back of the controllers. All Joy-Con models use Bluetooth 3.0, support NFC, and have the same battery stats: It takes three and a half hours to fully charge them, and you get 20 hours of play time on one charge. BatterySurprisingly, the Switch 2 does not make improvements to the Switch family's battery life—despite having the largest battery of all. The Switch 2's 5220mAh battery can run anywhere from two to 6.5 hours. (Nintendo warns that these are rough estimates and that battery life depends on the specific games you play.)Compare that to the original Switch and Switch OLED, both of which have a 4310mAh battery, and can run from 4.5 hours to nine hours. The Switch Lite (3570mAh) beats it too, with a range of three to seven hours. This isn't wholly surprising: The Switch 2 is quite a bit more powerful than the original Switches, so running a high-performing game at 120 fps is going to consume much more power than a 720p game running at 60 or 30 fps, or below. All consoles charge in about three hours when they're in sleep mode.Storage and communicationsThe Switch 2's games are likely the largest Nintendo has ever made, since the console is capable of much higher resolutions and frame rates than past generations. As such, it makes sense for Nintendo to pack more internal storage here: 256GB to be exact. The OG Switch and Switch Lite, on the other hand, only ship with 32GB of internal storage, while the Switch OLED offers 64GB. If you need more space, you can expand that storage with a microSDHC or microSDXC card on these models. Switch 2 also supports expanded memory, but specifically MicroSD Express cards, which limits the flexibility here. Switch 2 supports Wi-Fi 6, a step up from Wi-Fi 5 on the original Switches. Those Switches support Bluetooth 4.1, while Nintendo says the Switch 2 supports "Bluetooth" without any additional information. It likely supports a newer standard than that, seeing as it's launching eight years after the original, but we don't know for sure at this point.The Switch 2 also has a second USB-C port on the top of the unit. This is a great change that makes it possible to charge your Switch with the kickstand out.Audio and video chatFor the first time, a Nintendo console has camera support. The Switch 2 lets you connect a webcam for use with GameChat, so you can see your friends while playing games—and they can see you. Nintendo sells a camera of its own, but you can use some third-party webcams as well. The Switch 2 also has a built-in microphone for voice chat. That way, you can talk to your friends over GameChat with or without a camera—though I'm not sure how well they'll be able to hear you if your console and TV are across the room. Speaking of audio, the Switch 2 supports "3D Audio," which the company advertises as a more immersive audio experience. Essentially, sounds should seem like they're coming from where they originate in the game. (If a Mario Kart racer is on your tail, you should hear their engine "behind" you.) GamesWhen it comes down to it, the most important element of any console is its library of games. The Switch 2, like any new console, does have exclusives, though there aren't as many as you might think—at least not at launch. Some of the notable upcoming titles you will only be able to play on Switch 2 at this time are Mario Kart World, Donkey Kong Bananza, Drag X Drive, Hyrule Warriors Age of Imprisonment, Kirby Air Riders, and Elden Ring Tarnished Edition. There are also the Nintendo Switch 2 Edition titles—original Nintendo Switch games with graphical boosts thanks to the upgraded hardware. That includes both Switch Zeldas (Breath of the Wild and Tears of the Kingdom), Super Mario Part Jamboree + Jamboree TV, Kirby and the Forgotten Land + Star Crossed World, Metroid Prime 4 Beyond, and Pokémon Legends ZA.If you have a Nintendo Switch Online subscription, Switch 2 also supports GameCube titles. At launch, that includes The Legend of Zelda: The Wind Waker, Soulcalibur II, and F-Zero GX, but more are on the way.You can buy Switch 2 games physically or digitally, but not all physical games are going to ship with the game itself. That's because some of these titles are "game-key cards," which contain the "key" to let you download the game from the internet. Then, you can play the game—as long as you have the game-key inserted in your Switch. It's peculiar, but worth mentioning here, since the original Switch doesn't operate that way. As you might expect, the Switch 2 can play Switch 1 games, but you might not consider this truly backwards compatible. That's because the Switch 2 uses emulation to play Switch 1 games, and not all games play well via this emulation. Hopefully, Nintendo irons out the issues here, but in case they don't, you might not want to trade-in your old Switch to pay for a Switch 2 if you want to continue playing Switch 1 games.KickstandNintendo continues to advance the kickstand with each Switch model. The original features a small, skinny stand that was prone to snapping off (though easily reattached via magnets). The Switch OLED greatly improved upon this design, by making the kickstand take up the full width of the console, and support more angles. The Switch 2 has a kickstand that allows for even greater variety of play angles. It's not a huge change, but worth a nod. PriceThe Switch 2 was spared a price increase from the tariffs: The console runs for $449.99, or $499.99 if you want Mario Kart World bundled in. You probably want that deal, as the game costs $79.99 on its own. Not all Switch 2 games cost $80: Donkey Kong Bananza costs $69.99, but Nintendo is no longer the company for $60 games across the board.The original Switch retails for $299.99, while the Switch OLED goes for $349.99. The Lite originally launched for $199.99, but you can frequently find it on sale.

After a couple years of having its open-source Llama AI model be just a part of its Connect conferences, Meta is breaking things out and hosting an entirely generative AI-focused developer conference called LlamaCon on April 29. The event is entirely virtual, and you'll be able to watch along live on the Meta for Developers Facebook page. LlamaCon kicks off at 1PM ET / 10AM PT with a keynote address from Meta's Chief Product Officer Chris Cox, Vice President of AI Manohar Paluri and research scientist Angela Fan. The keynote is supposed to cover developments in the company's open-source AI community, "the latest on the Llama collection of models and tools" and offer a glimpse at yet-to-be released AI features. The keynote address will be followed by a conversation at 1:45PM ET / 10:45PM ET between Meta CEO Mark Zuckerberg and Databricks CEO Ali Ghodsi on "building AI-powered applications," followed by a chat at 7PM ET / 4PM PT about "the latest trends in AI" between Zuckerberg and Microsoft CEO Satya Nadella. It doesn't seem like either conversation will be used to break news, but Microsoft and Meta have collaborated before, so anything is possible. Meta hasn't traditionally waited for a conference to launch updates to Meta AI or the Llama model. The company introduced its new Llama 4 family of models, which excel at image understanding and document parsing, on a Saturday in early April. It's not clear what new models or products the company could have saved for LlamaCon. LlamaCon will stream live on April 29th through the Meta for Developers Facebook page and we'll be live-blogging the event right here on Engadget so you can get all the details as they happen.This article originally appeared on Engadget at https://www.engadget.com/ai/how-to-watch-llamacon-2025-metas-first-generative-ai-developer-conference-215241289.html?src=rss

Research revealed the DPRK is using AI in its malicious campaign.

Senators Adam Schiff and Elizabeth Warren are calling for an ethics investigation into President Trump's promotion of his meme coin.

Go behind the scenes of the visual effects of ‘Death of a Unicorn’, and see a detailed video breakdown.  When Elliot (Paul Rudd) and his daughter Ridley (Jenna Ortega) accidentally kill a unicorn on the way to the estate of Elliot’s boss in Alex Scharfman’s horror-comedy Death of a Unicorn, unusual things start to happen. The ‘dead’ unicorn’s blood and horn appear to have healing qualities, while Ridley’s encounter with the creature causes her to have cosmic visions. Eventually, the animal’s parents arrive to taunt–and kill–those at the estate trying to take advantage of the unicorn’s magical qualities. Behind much of this mayhem were visual effects supervisor Andrew Orloff, visual effects producer Jeremy Newmark, and Zoic Studios. CG unicorns were the centerpiece visual effects, while environments, set extensions, blood and gore and the cosmic visions also became key components of the 600 VFX shot film. Rob Price was Zoic Studios’ visual effects supervisor and Danica Tsang was the Zoic visual effects producer.  When Orloff came on board, he helped put together a number of decks, reference and mood boards for the creatures, working closely with concept artist Paul Mellender (Wētā Workshop had also done some preliminary designs). “The director, Alex, wanted it to be prehistoric,” recounts Orloff. “Like a real cryptid that the legend of a unicorn was based on, but as if that was a real creature that had been existing in our world, just hidden from human view.” “That idea that it had been around for so long took us away from the modern show horse with a horn on it into more of a predatory, apex predator kind of a feel,” continues Orloff. “We were looking at prehistoric horses, we were looking at lions and tigers and putting them all together to create a unique creature.”  To help establish the movements of the unicorns, Orloff oversaw a motion analysis exploration period with Zoic, relating to horses. “At one point, we were going to do some mocap’d horses, but we found that we couldn’t really get the motion that we wanted exactly. The body mechanics of the creature evolved in a way that we were going to have to step on the mocap so much that we decided to do it all keyframed. Patricia Binga was the animation supervisor and was amazing at putting together tests as part of the creature design process. This period also involved fur dynamics testing to see how thick the fur should be, and what fur patterns worked.” Death of Unicorn filmmakers and cast review a scene. What was further explored early on was the methodology for shooting scenes that included unicorns. Says Orloff: “We explored prosthetic pieces on real horses. We looked at just shooting real horses and manipulating them. Where we ended up with was CG unicorns, of course, but shooting with a combination of prosthetic puppets made by Filmefx, mostly for the foal. Filmefx took our designs and made a latex mold with a little bit of armature inside of them. We ended up building a full version of the foal that was on the ground and on the table. Then they built a head of the adult unicorn that had an articulated jaw and also a somewhat articulated neck. The horns were built as separate props from our 3D files.” To film scenes that involved close interaction between the unicorns and the characters, Orloff says ‘every trick in the book’ was utilized. “For example, there’s a scene where Téa Leoni’s character is impaled through the shoulder by the horn coming from behind a bookcase and then it slams her all around. That was a stunt where we had a plant-on for the back of the horn coming out of her back, and some prosthetic blood. And then we had a separate plate of the stuffy coming through. We ended up doing a CG version of it, but we kept as much as we could. We kept the plant-on, we kept some of the on-set special effects stuff of the books falling, and the dust, and we added to that with more books falling out and more blood. We also re-timed it in VFX so that the stunt looked more violent.” “Later in that same scene,” adds Orloff, “her character’s innards are being ripped out by the unicorn. There is a puppet in that shot that is pulling out some prop guts. We just left the stuff that’s on her torso that’s interacting with her, but the stuff that’s in the creature’s mouth was CG gore matched to the other gore.” A scene from the film. A principal challenge with the unicorns was their colorings. They are a mix of white also darker sheens, which, Orloff states, needed specific considerations. “We talked about the color of the unicorns extensively. I made it clear after doing white creatures before that a white creature isn’t white. It’s going to be the color of whatever the environment is. At the very beginning of the shot, we did do some experiments in rendering the fur to more of a mid-gray with a very strong ambient occlusion pass to ‘grab’ the details out. If it gets too dark, all the detail goes away and if it gets too bright, everything mushes together.” For the moment when Ridley experiences her cosmic visions (a phenomenon that Elliot also later encounters), Orloff says he discussed with the director a wealth of space film reference. “The idea was to keep it in the realm of the Stargate from 2001: A Space Odyssey, or Contact’s space travel, where it’s not really clear what it is you’re seeing because what you’re seeing is something that you can’t really fathom. However, we also knew we wanted it to have some realistic space imagery, like something out of the James Webb Space Telescope nebula imagery.” Orloff designed the cosmic scenes via ripomatics and animatics. The final imagery by Zoic Studios involved generating renders in Cinema 4D. “What we were trying to show was that they’re journeying into the source of the unicorn’s power,” continues Orloff. “We put some Easter eggs in there; one of them was when they shoot through the eye and they are jettisoned out. If you look at it, the Nebula there is actually derived from the model of the horn of the unicorn.” The post Creature feature: the art of a CG unicorn appeared first on befores & afters.

The Fast Company Impact Council is an invitation-only membership community of leaders, experts, executives, and entrepreneurs who share their insights with our audience. Members pay annual dues for access to peer learning, thought leadership opportunities, events and more. The landscape of home automation has sparked numerous discussions about security and control. According to SonicWall’s comprehensive 2025 Annual Cyber Threat Report, smart home products experienced a staggering 124% increase in cyberattacks during 2024, with smart plugs emerging as particularly vulnerable targets. These vulnerabilities have ignited growing consumer concern about the safety of cloud-dependent technologies that have dominated the smart home market.  As our homes become increasingly connected, the cloud-connected home automation we’ve been seeing for the past decade is being critically examined. A growing segment of consumers now seeks alternatives emphasizing local control, data privacy, and system resilience. Here we’ll examine the emerging paradigm of locally controlled smart home technologies, an approach that promises to return technological autonomy to homeowners.  The current landscape of smart home technology  The modern smart home ecosystem has evolved into an interconnected network predominantly reliant on cloud infrastructure. This cloud-centric approach requires constant internet connectivity, routing even the simplest commands through distant data centers thousands of miles away.  While promising convenience, this model introduces vulnerabilities:  A single network interruption can render home systems inoperable.  Each connected device becomes a potential cybersecurity entry point.  Data privacy becomes increasingly compromised as more personal information is transmitted and stored remotely.  Real-world IoT and smart home security failures  The most troubling aspect of cloud-dependent smart home technology is the threat of planned obsolescence. Manufacturers can shut down cloud services that power devices consumers have already purchased, rendering functional hardware useless.  For example, in February 2024, Amazon announced that its Echo Connect landline adapter would cease functioning, giving users just three weeks’ notice. This transition forced vulnerable users to quickly adapt to new technologies or lose access to a potentially life-saving communication tool.  The Echo Connect shutdown is far from an isolated incident. In April 2022, Insteon abruptly shut down its servers without warning, rendering thousands of connected devices instantly nonfunctional.   These incidents expose a critical flaw in cloud-dependent systems: Consumers never truly own their devices. When business interests change or financial pressures mount, companies can—and sometimes must—abandon their customers.  Beyond platform shutdowns, security researchers continue to uncover alarming vulnerabilities in cloud-connected devices. For example, in 2020, multiple families experienced privacy violations when their Ring security cameras were compromised, allowing strangers to speak directly to family members through the devices meant to protect them.  Corporate resistance to local control  Some manufacturers have actively blocked consumers from maintaining local control of purchased devices. In January 2024, appliance giant Haier issued legal threats against a developer who created Home Assistant integration plugins for the company’s smart appliances. These open-source tools allowed consumers to control their purchased devices without relying on Haier’s cloud infrastructure.  Despite overwhelming community support, with users creating hundreds of backup copies of the developer’s work, Haier’s aggressive stance exemplifies how manufacturers prioritize maintaining control over their ecosystems rather than respecting consumer ownership rights.  These failures underscore fundamental problems with cloud-dependent smart home technologies that consumers are increasingly recognizing:   Personal security is compromised when critical home systems depend on external servers.  Single points of failure make entire ecosystems vulnerable.  Personal data faces unnecessary exposure and transmission risks.  Manufacturers maintain excessive control over hardware consumers believe they own.  There is a growing community focused on local control, like Home Assistant, where users actively discuss strategies to maintain autonomy and minimize cloud dependency.   The local control revolution  Locally controlled smart home technology represents a paradigm shift in home automation, offering a fundamentally different approach to device management and data processing. At its core, this approach keeps all smart device functionalities within your home’s local network, providing unprecedented levels of privacy, reliability, and user control. The key principles of local control are simple.   Locally controlled smart home technology represents a paradigm shift, offering fundamentally different approaches to device management and data processing:  Local processing: Devices process information directly within the home network, delivering near-instantaneous response times  Network independence: Systems operate independently of internet connectivity  Privacy by design: Data processing remains within the home network, creating a barrier against external data collection  Platforms like Home Assistant and Hubitat have emerged as leaders in this revolution, offering robust, customizable ecosystems that prioritize user autonomy. These platforms empower tech-savvy users to build truly personalized smart home experiences that reflect their unique needs and values.  The path forward  The shift toward locally controlled smart home technology presents a significant opportunity for forward-thinking developers and manufacturers. As consumer demand for privacy, security, and true ownership grows, businesses that embrace this paradigm will gain competitive advantage in an increasingly crowded market.  For developers and product teams, the strategy is clear: Incorporate local processing capabilities alongside cloud features, respect user autonomy through open APIs, and build trust through transparent data practices. Companies that prioritize these values aren’t just serving a niche—they’re positioning themselves at the forefront of the next evolution in smart home technology.  The most successful smart home ecosystems won’t be those that lock users into vulnerable cloud dependencies, but those that empower customers with genuine control while delivering exceptional experiences.  Svetlin Todorov is cofounder of Shelly and CEO of Shelly U.S.A. 

This Quilted Furniture project is by Sapna Tayal, an Industrial Design student at Carnegie-Mellon University. I'd give her an "A;" Tayal has imaginatively harnessed production technology, collaborated with non-ID-department bodies and produced a highly professional level of documentation.Tayal's central question was: "How can we harness the unique capabilities of the overlooked CNC sewing machine, traditionally only used to make soft blankets, and expand its fabrication possibilities?" To get to the answer, she worked with researchers from CMU's Human Computer Interaction Institute (HCII) and Robotics Institute (RI).Flat-to-shape objects fabricated with a CNC sewing machineHow can we harness the unique capabilities of the overlooked CNC sewing machine, traditionally only used to make soft blankets, and expand its fabrication possibilities? Deployable objects allow quick activation and compact storage but often require intricate assembly processes. We establish a novel rapid, scalable method to fabricate functional objects using a large computer-controlled quilting machine (11 ft / 3.4 m wide). Rigid materials like plywood or acrylic are sandwiched between fabric layers and secured through sewing, forming soft hinges. These hinges allow smooth transitions into three-dimensional forms with minimal post-processing. Flat-to-shape steps of the side table Method OverviewLongarm Quilting Machine. Fabric is tensioned over the rollers. Leveraging the machine's strengthsModern CNC quilting machines, offer precision, reliability, and scale. However, their potential remains largely untapped beyond flat, fabric-based products like quilts. We take advantage of the machine's unique features to craft unique, surprising, fully functional objects. Computer-controlled sewing and panel insertion for the backpack design (10x speed).Human-machine co-creation Establishing an incremental fabrication process with alternating passes of sewing pockets and placing inserts.Designed Objects1. Lamp Made up of four panel types for localized functions to achieve dynamic functionality: plain acrylic at the base, cardstock-laminated panels in the inner ring, LED-integrated "active" panels in the upright section, and open spaces at the top for a graceful fabric drape as the lampshade.2. Backpack Inspired by packaging design, a tendon cord system transforms the bag from flat to 3D and doubles as the straps. Made of 3/16" plywood inserts. Gusset panels remain flexible with Velcro for selective attachment and locking.3. Side Table Transforming with a simple tug secured with a spring cord lock while in use. The cord acts on short rectangular struts, which meet in the center under the tabletop when fully deployed, stopping further movement.4. Lounge Chair A large (60" by 56" when flat) chair designed to support an adult's weight.The insert panels, made from 1/2" plywood cut on a CNC router, are designed to meet face-to-face when deployed. Neodymium magnets, glued into milled pockets, allow the chair to "snap" into place, aligning the panels, while the fabric and panels carry the primary loads.Design ProcessSewing PossibilitiesA research-through-design process to play with different sewing and gathering methods and studying their effects. Geometry and Method TuningPrototyping possible forms. Technical ProcessChair fabrication process where a table and weights are used to build objects larger than the bed size.Scaling upWe adapt the machine (11 ft x 25 in sewing area) for large scale rigid panel fabrication by replacing the typical fabric rolling method with a height-adjustable table. A weighted rod threaded through the fabric edge maintains tension and alignment during stitching.Original machine set up: Our modifications to expand fabrication scope: Path and Panel Sequence AlgorithmIn the full research paper, we formalize a repeatable design space to share our discoveries. Reach out for more information! Putting the CNC Sewing machine together! Mid-project we needed to move the machine (~13' by 5') from its original home in Robotics to the HCI lab across campus.Taking the machine apart and putting it back together was an incredible opportunity to learn about its workings and nuances of operating it! You can see more of Tayal's work here.

At Expo 2025 Osaka, among the eight Signature Pavilions redefining the architectural dialogue of the future, null² emerges as a standout vision. Designed by Tokyo- and Taipei-based studio NOIZ in collaboration with media artist and researcher Yoichi Ochiai, the pavilion is more than just a building, it’s a living experiment in hybrid space-making, where the boundaries between physical and digital, static and dynamic, dissolve into a shimmering architectural interspace. Conceived as a speculative model for future architecture, null² occupies a unique conceptual realm: it’s neither fully real nor entirely virtual, but something in between, a meta-environment that interacts with its users and surroundings. Its form is built from voxel-like modules, cubic units of 2, 4, or 8 meters, arranged to create a flowing sequence of exhibition, support, and rest spaces. These modules vary from rigid geometric blocks to more warped, organic shapes, each clad in a newly developed mirrored membrane that shifts with the wind and reflects the changing world around it. Designer: NOIZ This tensile membrane does more than just mirror its environment. Designed to be reactive, it shimmers and moves in response to environmental forces, creating a visual rhythm that gives the structure a sense of breath. In doing so, it transforms the building’s presence from a static object to a living system, one that visually and metaphorically blurs into its context. From its very conception, null² has been designed with agility in mind. In response to the tight timeline and uncertain future beyond Expo 2025, NOIZ developed a modular framework that prioritizes speed, flexibility, and reuse. The structure can be rapidly deployed, easily dismantled, and potentially relocated. The innovative membrane not only enhances the structure’s aesthetics but also offers solar reflectance and thermal insulation, features that hint at sustainable applications beyond the Expo, such as retrofitting existing buildings. Inside the pavilion, visitors are drawn into a multisensory exhibition experience built around the concept of digital twins. These avatars reflect and respond to the user’s presence, making the act of visiting itself a kind of co-creation. Within select modules, robotic arms and embedded woofers interact with the mirrored surfaces through movement and sound, producing an environment that feels both tactile and intelligent. The building, in this way, becomes a responsive interface, a medium for communication between the visitor and the space. But null² is not confined to its physical form. It extends into the digital, offering a participatory platform where remote users can engage with the pavilion through robotic avatars or augment their on-site experience using AR overlays. Visitors are encouraged to digitally reinterpret the structure, creating personalized spatial edits and contributing to an evolving archive of user-generated designs. This participatory approach redefines architectural authorship, replacing the singular vision of the designer with a collective, crowd-sourced process. The post Expo 2025’s Most Futuristic Pavilion Feels Alive and Responds to Your Presence first appeared on Yanko Design.

We like Max, but it may be held back by a few hardware concerns

Several founders of hot startups took big payouts and let Mark Zuckerberg gobble up their companies—and came to regret it.