A viral claim says “China has banned companies from laying off workers because of AI”. That sounds dramatic, and it is easy to see why people shared it. But it is not quite what happened.

The real story is narrower, and perhaps more useful: a Chinese court said one company could not use its own AI-driven restructuring as an automatic excuse to demote a worker, cut his pay, and then fire him when he refused.

The case was part of a recently publicized set of typical AI-related labor cases from the Hangzhou Intermediate People’s Court. It focused on an employee surnamed Zhou, who worked as a project supervisor / quality-inspection supervisor. According to Chinese coverage, Zhou earned 25,000 yuan per month (~3,600 USD) before his employer tried to move him into a lower-paid role at 15,000 yuan per month after introducing AI tools into the workflow.

That is where the case becomes very human. For workers, “AI efficiency” is not an abstract tech trend when it suddenly means a 40% pay cut. Zhou refused the new arrangement, and the company terminated him.

The court did not say AI can never change jobs. It did not say Chinese companies are banned from layoffs involving automation.

The key point was that the employer’s own decision to adopt AI did not automatically qualify as a “major change in objective circumstances” that made the original labor contract impossible to continue. In plain English, a company cannot simply say “AI did it” and walk away from ordinary labor obligations.

Reports differ on the exact compensation framing. Chinese coverage says the court supported compensation under the 2N formula, with several reports putting the amount at more than 260,000 yuan. A separate roughly $43,000 / $44,000 figure appears to refer either to Zhou’s reported annual salary of 300,000 yuan or to earlier severance/payment reporting, not necessarily the final wrongful-termination award.

The accurate takeaway is not “China banned AI layoffs.” It is more precise: AI may change workflows, but employers may still have to prove that demotions, pay cuts, and dismissals are lawful. In this case, the court said they were not.

The most significant shift in the XRING 03 is the removal of the “large core” cluster. While previous generations utilized a four-tier processing division, Xiaomi is transitioning to a three-block system consisting of a main core, a titanium core, and small cores.  

This architectural change has significantly enhanced the performance of the base layer. The small cores now reach speeds of 3.02 GHz, representing a nearly 60% increase in agility compared to the previous version. This advancement is designed to provide superior management of background tasks and simultaneous processes.  

The chipset also features substantial improvements in visual processing. The graphics unit operates at approximately 1.5 GHz, a 25% increase in raw speed intended to ensure fluid performance on the large-format foldable display during intensive tasks. Additionally, the system maintains the high-speed memory standards of the previous generation, supporting 9,600 MT/s RAM.  

Market Positioning 

The XRING 03 is engineered to provide aggressive performance for Xiaomi’s premium foldable segment. The MIX Fold 5 is expected to launch in Asian markets with an approximate price of $1,500. This pricing strategy reflects the high-end nature of the hardware and its specialized architecture, which prioritizes multitasking efficiency and high-frequency peak performance without relying on traditional intermediate processing layers. 

Hogwarts Legacy is an immersive, open-world action RPG set in the universe first established in the Harry Potter book series. Players are invited to embark on a journey through both new and familiar locations within the wizarding world. Key gameplay features include:  

• The discovery and interaction with fantastic beasts.  

• Extensive character customization.  

• The ability to craft potions and master a wide array of spells.  

• A talent system that allows players to upgrade their abilities and define their own path as a wizard or witch.  

Technical Requirements 

To ensure the game runs effectively on PC, the following minimum system requirements have been specified:  

• Operating System: Windows 10.  

• CPU: Intel Core i5-8400 or AMD Ryzen 5 2600.  

• Memory: 16 GB of RAM.  

• GPU: NVIDIA GeForce GTX 1070 or AMD RX VEGA 56.  

• DirectX: Version 12.  

• Storage: 85 GB of available space.  

Redemption Process 

Users interested in securing the title must navigate to the Hogwarts Legacy product page on the Epic Games Store. After logging into a valid account, users simply need to click the “Get” button to add the game to their permanent library. This limited-time offer provides significant value, given that the title was one of the most celebrated releases of 2023. 

The Yoga Pro 9i centers on the Intel Core Ultra 300 series, boasting the Core Ultra 9 386H chipset. To support demanding workflows, the device includes 32 GB of LPDDR5X RAM (7,467 MT/s) and a 1 TB PCIe 4.0 SSD. Graphical tasks are handled by NVIDIA’s RTX 50-series mobile GPUs, with options ranging up to the RTX 5060 (or the RTX 5070 in some markets outside the US). 

Advanced Display and Battery Technology 

A primary highlight of this generation is the display versatility. While the standard option features a 2.8K OLED panel at 120 Hz, users can opt for a 3.2K Tandem OLED screen. This panel achieves a peak HDR brightness of 1,000 nits and carries DisplayHDR True Black 1000 certification, ensuring high color accuracy for visual editors. 

Powering the device is a 92.5 Wh battery, supported by Rapid Charge Express technology. Lenovo claims this system provides three hours of runtime from a 15-minute charge, a critical feature for mobile professionals. 

Build and Connectivity 

Screenshot

Constructed from premium metal, the chassis maintains a thickness of 0.68 inches (17.2 mm) and a starting weight of 4.19 lbs (1.9 kg). Connectivity is robust, featuring: 

• Two Thunderbolt 4 ports (40 Gbps) 

• HDMI 2.1 and a UHS-II SD card reader 

• Wi-Fi 7 and Bluetooth 5.4 support 

Price and Availability 

The Yoga Pro 9i Aura Edition Gen 11 is now available in the United States. Reflecting current component market trends, the starting price is $2,799. This represents a shift from the $1,899 entry price originally estimated during CES, positioning the device firmly in the high-end enthusiast segment. 

According to CNevPost, Xiaomi’s automotive division reported a substantial growth trajectory in early 2026. In April, the company delivered over 30,000 units, a notable increase from the 21,440 vehicles sold in March. This upward trend highlights Xiaomi’s ability to scale production rapidly to meet growing consumer demand. 

Technical Specifications of the YU7 GT 

The YU7 GT is positioned as the high-performance variant within Xiaomi’s SUV lineup. According to official filings with the Chinese government, the vehicle features substantial dimensions designed to optimize both interior cabin space and cargo capacity: 

Despite its large footprint, the YU7 GT is engineered for elite performance. Preliminary data indicates the SUV will utilize a dual-motor configuration producing a combined output of 738 kW (approximately 990 hp). This powertrain is expected to enable a top speed of 300 km/h (186 mph) and an acceleration from 0 to 100 km/h (0–62.1 mph) in roughly two seconds. 

The YU7 GT is expected to be officially unveiled in May 2026. Regarding its market entry, the vehicle is projected to be priced between 450,000 and 500,000 Yuan. 

By leveraging its expertise in integrated technology and rapid manufacturing, Xiaomi is positioning the YU7 GT to compete directly with established luxury performance EVs. The success of its previous models has provided the brand with the necessary momentum to invest further in high-end electric mobility, signaling a long-term commitment to the global automotive sector. 

A drone made mostly from cardboard sounds almost laughable at first. It suggests shipping boxes, school projects, or something that should fall apart in bad weather, not a serious piece of military technology.

But that is why Japan’s cardboard drone story is getting attention. The concern is not that cardboard suddenly became a miracle aerospace material. The concern is what happens when drones become cheap enough to build, transport, and lose in large numbers.

The drone at the center of the story is the AirKamuy 150, made by Japanese defense-tech startup AirKamuy. The company describes it as Japan’s first fixed-wing UAV made mainly from cardboard, with a design focused on low cost and long-distance flight. AirKamuy has also publicly discussed possible roles including training targets, reconnaissance, transport, self-destruct use, and drone swarm applications.

Japan’s military is already paying attention. Defense Minister Shinjiro Koizumi recently met with AirKamuy, and recent reporting quoted him saying the Japan Maritime Self-Defense Force is already using the cardboard drones as targets. That is not the same thing as deploying cardboard attack drones in combat, but it does move the aircraft beyond a simple trade-show curiosity.

One cheap drone may not sound very frightening. A huge number of cheap drones is a different problem. If a drone is inexpensive, easy to produce, and good enough for a narrow mission, it can change the economics of defense.

Stopping drones can be (very) expensive. A low-cost aircraft can still force an opponent to spend radar time, ammunition, missiles, manpower, and attention. Even if many are shot down, the imbalance may favor the side that can produce and launch them faster.

There is an important caveat. Based on the strongest accessible evidence, Japan has not confirmed that it has adopted or deployed an armed cardboard “suicide drone” for operational combat use. The confirmed story is narrower: AirKamuy built a cardboard fixed-wing drone, the company has discussed one-way and swarm-related uses, and Japan’s Maritime Self-Defense Force is reportedly already using the drones as training targets.

Cardboard drones sound ridiculous. But the point is not the cardboard, it’s scale. A cardboard drone may look flimsy by itself. A huge swarm of cheap drones may pack a considerable punch.

Seoul Forest was supposed to feel like a gentle spring escape: gardens, walking paths, photo spots, and Pokémon hiding among the trees. Instead, opening day showed just how powerful nostalgia can be when a beloved childhood brand meets a real-world destination.

Pokémon Secret Forest opened May 1 as part of the 2026 Seoul International Garden Show, a massive public event running through October 27 across Seoul Forest and nearby areas. Seoul says this year’s show is its largest and longest yet, with 167 gardens, 90,000 square meters of garden space, and a 10 km route connecting Seoul Forest, Seongsu-dong, and surrounding neighborhoods. Pokémon Secret Forest is scheduled separately through June 21.

That mix was irresistible. Families, fans, collectors, and casual visitors poured into the area on Korea’s Labor Day holiday. What could have been a quiet garden visit quickly turned into a reminder that Pokémon is no longer just a game or cartoon. For many people, it is childhood, family memories, and a reason to get outside.

The happy rush also became a safety issue. Korean media reported that crowds packed Seongsu-dong and Seoul Forest, with Seoul estimating about 40,000 people around Seongsu Cafe Street by noon. Authorities received crowd-safety reports, police and firefighters were sent to the area, and organizers paused parts of Pokémon Mega Festa (city-wide event), the Pokémon GO Seoul Stamp Rally (one of the mega festa locations) in particular. No injuries were reported

Kyunghyang Shinmun (in Korean) reported that more than 50 crowd-safety reports were filed, about 100 police personnel and 30 firefighters were deployed, and the organizer stopped the event after requests from local authorities and police. It also says Pokémon Korea announced that the event was temporarily suspended for safety reasons due to heavy crowds.

It was a hugely popular public event that appears to have overwhelmed the opening-day crowd plan.

The good news is that the larger garden show continues for months, and Pokémon Secret Forest is scheduled to remain open through June 21. For visitors, the lesson is simple: this may be one of Seoul’s most charming spring attractions, but it is probably not a place to visit at peak crowd hours.

Modular Customization: The Clutch Talon

The Clutch Talon is designed for the competitive market, featuring a highly modular architecture. Users can quickly swap analog sticks, directional pads, and triggers without specialized tools. This design supports 3D-printed components, allowing players to tailor the controller’s ergonomics to their specific grip style.

Technologically, the controller utilizes Hall Effect sensors in its primary buttons to eliminate stick drift—a common hardware failure in traditional joysticks. It also includes three-stage trigger locks and programmable rear buttons. The device is compatible with PC, Xbox, and Android, offering dual wireless connectivity and up to 30 hours of battery life.

Comfort and Endurance: Cloud Stinger 3

The Cloud Stinger 3 series focuses on long-form gaming sessions, utilizing lightweight stainless steel headbands and memory foam ear cushions. The wireless variant is particularly notable for its 80-hour battery life and Bluetooth connectivity. Each model in the lineup includes an integrated microphone with a user-friendly “flip-to-mute” function.

Advanced Software Integration

The official NGENUITY application has transitioned out of its beta phase. The software now features professional-grade audio equalization and AI-driven voice noise reduction. For hardware enthusiasts, it allows for granular adjustments to button sensitivity and the recording of complex macros.

Pricing and Availability

The Cloud Stinger 3 headsets are currently available starting at $50. The Clutch Talon controller is scheduled for a mid-year release with a retail price of $160. The NGENUITY software is available for free as a download, globally.

The update expands native file handling capabilities, allowing users to open and extract formats such as .nupkg, .xar, .uu, and .cpio directly within the system. By integrating these backup and package tools into the Windows 11 core, Microsoft reduces the necessity for third-party extraction software.

Under the hood, the update incorporates elements from Project K2, an internal optimization initiative. These “bits” are designed to streamline the Windows 11 kernel, preparing the architecture for future AI-driven features while improving the general reliability of the explorer.exe process to prevent crashes.

Interface Refinements

Fix: Windows 11 now ensures the Downloads folder maintains its “Extra Large Icons” setting when accessed via browsers, correcting a bug that previously forced a reset to “List” view.

Several long-standing visual and functional bugs have been addressed:

• Dark Mode Consistency: The update resolves the “white flash” glitch that occurred when opening “This PC” or resizing panels, ensuring a consistent dark interface.
• Persistent View Settings: A fix has been implemented for folder viewing preferences. Previously, directories accessed via external applications often reverted to a default “List” view; the system now correctly memorizes and maintains user-selected layouts.

Deployment and Availability

The update package size is approximately 5.1 GB for x64 systems and 4.6 GB for ARM architectures. In addition to system stability, the build introduces optional “agents” to the taskbar for quicker interaction.

Microsoft reports no major known issues with this release, making it a safe installation for those seeking immediate performance gains. Users can acquire the update via Windows Update or through offline installers (.msu) on the Microsoft Update Catalog. For those who choose not to install the optional update now, these enhancements will be bundled into the mandatory May 2026 Patch Tuesday release.

Flying taxis always sounded like one of those futuristic ideas that looked good in articles, movies and concept videos but never quite arrived in real life. This week, that future came a little closer.

Joby Aviation, which we covered before, began a series of piloted electric air taxi demonstration flights in New York, flying between John F. Kennedy International Airport and several Manhattan heliports. The flights used real airport and city infrastructure, not a closed test site, which makes this more than just another tech demo.

To be clear, this is not a passenger service yet. You cannot open an app and book a flying taxi from JFK to Manhattan today. Joby still needs regulatory approval before it can carry paying passengers in the U.S. But the New York flights are an important signal: flying taxis are moving from science-fiction promise to real-world urban testing.

The appeal is easy to understand. Anyone who has landed at a major airport knows the frustration of the final trip into town. The flight may be over, but then comes the traffic, the tunnel, the surge pricing, or the unpredictable ride that can turn a short distance into a long delay. Joby has said a future JFK-to-Manhattan trip could take only minutes, compared with a much longer car ride in traffic. As a heavy traveler, I like this!

But the bigger question is not just whether flying taxis could help some travelers skip congestion. It is what everyone else will have to live with.

If electric air taxis become part of city life, most people may never ride in one. But they may still hear them. They may see them flying over buildings, parks, schools, neighborhoods, and homes. That changes the conversation from convenience to public acceptance.

Would people feel safe with a new category of aircraft flying regularly over dense cities? Would the aircraft be quiet enough for residents to tolerate? Would cities welcome more activity in the sky if only a small number of people can afford to use the service?

That is where this story becomes more emotional than technical. With the technology aspect soon to be solved, the simple question for everyone living under the flight paths is: who gets the benefit, and who gets the disruption?

For now, Joby’s New York campaign is still a demonstration, not the beginning of everyday flying taxi service. But it makes the future feel closer and more realistic than before.

would you be okay with them flying over your house?

The device is powered by the Qualcomm Snapdragon 8 Elite Gen 5 chipset, built on a 3nm process. OnePlus reports that the processor achieved over 4.1 million points on the AnTuTu benchmark, representing a 42% performance increase over its predecessor. To maintain thermal stability, the tablet integrates a massive 45,260 mm2 vapor chamber.

The visual experience is centered around a 13.2-inch LCD display with a 3.4K resolution (3392 x 2400) and a 7:5 aspect ratio. The panel supports an adaptive 144 Hz refresh rate, 1,000 nits of peak brightness, and Dolby Vision. Despite its robust internal hardware, the tablet maintains a slim profile of just 5.94 mm and weighs 672 g.

Memory, Battery, and Software

The OnePlus Pad 4 is available in two configurations:

• 8 GB or 12 GB of LPDDR5X RAM.
• 256 GB or 512 GB of UFS 4.1 internal storage.

A significant highlight is the 13,380 mAh battery, which supports 80W SuperVOOC fast charging. The company claims this capacity allows for up to 20 hours of continuous video playback. The device runs on OxygenOS 16.0, based on Android 16, and features various AI-driven productivity tools such as AI Writer, AI Summary, and real-time audio transcription.

The tablet features an eight-speaker spatial audio system, Wi-Fi 7, and Bluetooth 6.0. To enhance its utility as a workstation, OnePlus introduced the Stylo Pro pen and a new backlit smart keyboard with a large trackpad.

Pricing and Availability

The OnePlus Pad 4 will go on sale in India on May 5. The 8 GB/256 GB model is priced at approximately ₹59,999 (~$631), while the 12 GB/512 GB version is positioned at ₹64,999 (~$684). It is available in two colors: Dune Glow and Sage Mist.

The idea sounds like science fiction, but it is now part of a very real policy debate: should future cars be able to detect when a driver may be impaired and stop that person from driving?

It sounds like a good safety feature, but that is the uncomfortable question behind the latest update from the National Highway Traffic Safety Administration (PDF link). Congress had directed regulators to create a rule for “advanced drunk and impaired driving prevention technology” in new vehicles, but NHTSA has missed its deadline and has not finalized what that rule should look like.

On paper, the goal is easy to understand. Drunk driving still kills thousands of people every year in the United States, and few people would argue against preventing those crashes before they happen. The difficult part is how a car would know that the person behind the wheel is not fit to drive? By watching them, constantly.

That is where many drivers get nervous.

To work as intended, the car would need to monitor something while the vehicle is being used. That could mean watching the driver’s face and eyes with an in-cabin camera. It could also mean analyzing steering, braking, lane behavior, or using some other sensor to look for signs of impairment. The law itself does not say every car must have a driver-facing camera, but most real-world systems people have seen so far do involve cameras in some form.

For many consumers, that is the first red flag: is my car watching me?

The second concern may be even bigger: what happens if the car decides I should not drive?

A warning light is one thing. A system that limits speed, prevents the car from starting, or forces some kind of intervention is another. That raises obvious questions about false positives, emergencies, liability, and basic trust. What if the system is wrong? What if someone is tired, sick, or simply looking away for a moment? What if the car refuses to move when the driver has a legitimate reason to leave?

This is why the debate is not only about drunk driving. It is about how much authority future vehicles should have over the person sitting in the driver’s seat.

The next big question is data. If a car monitors impairment, where does that information go? Does it stay inside the vehicle? Can the automaker access it? Could insurers, law enforcement, or hackers eventually get involved? These details will matter enormously because safety technology can quickly become a privacy issue if people do not trust how it works.

For now, there is no final rule, no single mandated technology, and no immediate requirement that every new car use the same monitoring system. But the direction is clear: regulators want cars to become more proactive about preventing impaired driving.

The real fight is over how far that should go. What’s your take?

Canonical, the developer behind the Ubuntu Linux distribution, recently announced plans to integrate artificial intelligence features into the operating system. The initiative has sparked a significant debate within the Linux community, with some users expressing concerns regarding privacy and system bloat. Critics have drawn comparisons to Microsoft’s aggressive AI implementation in Windows 11, leading to calls for a global kill switch to disable all AI-related functionalities.

According to Jon Seager, Canonical’s VP of Engineering, the initial phase of AI integration will focus on accessibility and productivity. Planned features include enhanced speech-to-text and text-to-speech capabilities, as well as agentic AI designed for automated troubleshooting and system automation. Canonical is also encouraging its internal teams to leverage AI more heavily in the development process over the coming year.

In response to community feedback, Seager clarified that Canonical does not intend to implement a single global toggle. Instead, the company will adopt a modular, opt-in approach. These features are slated for a preview release in Ubuntu 26.10, where they will be strictly optional. For subsequent releases, Canonical plans to include a dedicated step in the initial setup wizard, allowing users to decide whether to enable AI-native features during the installation process.

Furthermore, these capabilities will be delivered via Snaps, which are Canonical’s containerized software packages. Because these features are layered on top of the base operating system rather than baked into the core stack, users maintain the ability to manually remove the specific Snaps associated with AI at any time.

Despite these assurances, a segment of the user base remains wary of AI’s presence in a distribution known for its focus on user agency. While the ability to opt out or uninstall Snaps addresses technical concerns, some enthusiasts have indicated they may migrate to Ubuntu-based alternatives like Linux Mint, Pop!_OS, or Zorin OS to avoid AI integration entirely. Canonical’s challenge moving forward will be balancing modern feature parity with the strict privacy expectations of the Linux ecosystem.

Samsung has officially commenced sales of its latest high-end display, the Odyssey G8 (G80HS). Distinguishing itself as the market’s first 6K monitor specifically tailored for gaming, this 32-inch peripheral aims to bridge the gap between extreme pixel density for creators and high-speed performance for enthusiasts.

Market Availability and Pricing

The monitor is currently available for purchase in Germany through Samsung’s official online store, priced at €1,499 (~$1,751). Shipping is scheduled to begin on June 19. While the product is expected to expand into additional European markets shortly, Samsung has not yet confirmed release dates or pricing for regions outside of Europe.

Display Performance and Dual Mode Technology

The core appeal of the G80HS lies in its 32-inch IPS panel, which delivers a native resolution of 6144 x 2456 pixels. This choice of panel technology ensures wide viewing angles and high color accuracy, essential for professional content creation.

The monitor features a unique Dual Mode capability to balance resolution and fluidity:

• 6K Mode: Operates at a native refresh rate of 165 Hz for maximum detail.

• 3K Mode: By reducing the resolution, the monitor can reach a refresh rate of 330 Hz, catering to competitive gaming scenarios where motion clarity is prioritized over raw pixel count.

Technical Specifications and Connectivity

Designed to handle demanding visual tasks, the Odyssey G8 boasts a 1 ms response time and a typical brightness of 350 nits. To ensure a tear-free experience across different hardware configurations, it is compatible with both AMD FreeSync Premium Pro and NVIDIA G-Sync.

Regarding connectivity, Samsung has equipped the model with modern standards to support its high bandwidth requirements:

• DisplayPort 2.1

• HDMI 2.1

These ports allow the monitor to integrate seamlessly into high-end PC setups, providing the necessary throughput for 6K gaming and high-bitrate video editing. The Odyssey G8 G80HS represents Samsung’s push to redefine the “ultra-premium” segment by combining professional-grade resolution with top-tier gaming features.

The European Union has officially implemented the next phase of its universal charging mandate, requiring all new laptops sold within the continent to feature a USB-C port for power delivery. Effective as of April 28, the regulation establishes USB-C as the universal standard for computing devices in the region, bringing notebooks into alignment with mobile phones, tablets, digital cameras, headphones, and portable consoles.

While the original legislation was approved in 2022, laptop manufacturers were granted an extended grace period to comply with the mandate. This additional time accounted for the technical complexities involved in the transition, such as redesigning motherboards and reconfiguring internal power management systems to handle the specific requirements of USB-C power delivery. Furthermore, the law dictates that manufacturers must ensure consistent charging speeds across all compatible devices.

USB-C should become the standard for new laptops in the EU.

The European Commission’s primary goal is to simplify consumer electronics usage by eliminating the need for proprietary cables and adapters. This shift is expected to yield significant benefits:

• Financial Savings: Estimates suggest that the move to a single charging standard will save European consumers approximately €250 million annually by reducing the need for separate accessory purchases.

• Waste Reduction: The environmental impact is equally substantial. European authorities estimate that discarded or unused chargers account for roughly 11,000 tons of electronic waste per year. Transitioning to a universal standard is intended to drastically reduce this volume of toxic waste.

Although the law is specific to the European Union, the impact is expected to be global. To streamline production chains and minimize design costs, most manufacturers are likely to adopt USB-C charging for all markets worldwide. Some companies, such as Apple, had already transitioned their MacBook lines to USB-C charging ahead of the deadline. This regulatory shift ensures that the broader industry now follows suit, standardizing the charging experience for the foreseeable future.

Google has introduced a significant update to Gemini, enabling the AI application to generate and export downloadable files directly within its interface. This feature is designed to enhance productivity by streamlining the workflow between content creation and final file generation, effectively removing the manual “copy-paste” steps previously required.

The update supports a wide range of industry-standard formats to cater to diverse professional and academic needs. Gemini can now produce:

• Document Formats: Google Docs, PDF, and Microsoft Word (.docx).

• Data & Technical Formats: Microsoft Excel (.xlsx), CSV, LaTeX, and Markdown.

• Plain Text: Standard .txt files.

This versatility allows users to transform raw prompts or disorganized notes into structured reports, study guides, or data tables ready for immediate sharing. For instance, extensive meeting notes can be converted into a formatted PDF guide complete with visual elements and formulas.

Beyond file generation, Google has expanded Gemini’s functional ecosystem with several features aimed at personalization and continuity:

• Memory Function: The system can now store user preferences through a “Memories” feature, which allows the AI to adjust its responses based on previous interactions and specific user requirements over time.

• Platform Portability: A new import tool allows users to bring conversation histories from other AI platforms into Gemini. This ensures project continuity for those migrating work from competitive services.

• Automated Formatting: The integrated workflow ensures that content is generated and formatted according to the specific requirements of the chosen file extension.

Global Availability

Google has confirmed that the file generation feature is now available to users globally. This update positions Gemini as a more comprehensive workstation tool rather than a simple chatbot, centralizing document production within a single AI-driven environment. As these features roll out, users can expect a more seamless transition from conceptual brainstorming to professional documentation.

ASRock has quietly expanded its Phantom Gaming lineup with the introduction of two new monitors, headlined by the ASRock Advanced PG27QFW2A. This 27-inch model is designed to bridge the gap between entry-level displays and premium competitive hardware, focusing on high-speed performance for eSports enthusiasts.

ASRock Advanced PG27QFW2A: High-Speed Precision

The primary feature of the PG27QFW2A is its flat Fast IPS LCD panel, which delivers a Quad HD (2560 x 1440) resolution. It stands out in the market with an exceptionally high 400 Hz refresh rate and a gray-to-gray (GtG) response time of 1 ms.

Beyond speed, the monitor offers significant color accuracy and sync technology:

• Color Gamut: Covers 99% of the DCI-P3 space and 132% of sRGB.

• Variable Refresh Rate (VRR): Supports AMD FreeSync Premium to prevent screen tearing.

• HDR Support: Certified with VESA DisplayHDR 400.

The unit features ASRock’s “AlienTechElegance” design, inspired by extraterrestrial aesthetics. Notably, it includes a built-in Wi-Fi antenna in the stand, intended to improve signal reception for desktop PCs. The ergonomic base allows for full adjustments, including height, tilt, swivel, and pivot.

Connectivity Suite:

• Video: 2x HDMI 2.1 and 1x DisplayPort 1.4.

• Data: A USB hub featuring 2x USB Type-A and 1x USB Type-B ports.

• Audio: 3.5 mm headphone jack.

ASRock Advanced PG32QFT: Large-Format Alternative

For users prioritizing screen real estate over extreme refresh rates, ASRock also introduced the PG32QFT. This 32-inch model maintains the Quad HD resolution but operates at a 180 Hz refresh rate.

While it shares the same visual identity as the 27-inch model, it features more modest specifications, such as 97% DCI-P3 color coverage and HDMI 2.0 ports.

Market Availability

ASRock has not yet confirmed the official pricing or specific release dates for either model. Global availability is expected to be announced in the coming weeks.

Compared to last year’s devices, the hardware updates are modest, with most of the changes focused on the uber-stylish design, elegant new Pantone colors and premium materials. I love art and design, and to me, the new Motorola Razr series is the most fashionable flip phone lineup on the planet.

"THE NEW MOTOROLA RAZR SERIES IS THE MOST FASHIONABLE FLIP PHONE LINEUP ON THE PLANET."

Alongside the design, the unique new software features are the most significant changes, pushing the user experience to the next level.

What’s New?

The Razr Ultra keeps last year’s Qualcomm flagship SoC, the Snapdragon 8 Elite, while the Razr Plus retains the Snapdragon 8s Gen 3. The Razr gets a processor update, switching from MediaTek Dimensity 7400X to Dimensity 7450X. Motorola made the right trade-off to keep the price down while offering a powerful performance.

For the camera hardware, only the Ultra gets a new LOFIC sensor, which Motorola claims can capture up to 6x more dynamic range and the same 50MP resolution as last year’s main camera. In my opinion, the combination of a dual 50MP rear camera system with a 50MP internal camera on the Ultra, or 32MP internal cameras on the Razr+ and Razr, still makes for a powerful flip-phone camera setup.

The key upgrades are in the colors, materials, displays, batteries, camera software, and Dolby support. Motorola is moving from pOLED to Extreme AMOLED for the internal displays across the new Razr family. AMOLED technology is known for self-emissive pixels, deep blacks, high contrast, fast response times, and vibrant colors. In high-end implementations, it can also support very high brightness and advanced HDR features. Extreme AMOLED is Motorola’s high-end AMOLED branding, first introduced at CES 2026 with the flagship Motorola Signature smartphone.

Motorola also added silicon-carbon battery technology, allowing higher capacities in a thin form factor, and introduced new Pantone colors and finishes.
The manufacturer also introduced impressive new software features that significantly improve usability, including Camcorder Rotate to Zoom, Frame Match, Google Photos Wardrobe, and Google Daily Drops.

With the Razr series 2026, Motorola pulled off a rare balance: highly fashionable flip phones that still deliver serious performance, an enjoyable user experience, and useful creative features that help people capture, personalize, and express more of their everyday lives. That is why the new Razr series earns one of our Ubergizmo Editors’ Choice awards.

New Software Features For All Razr 2026 Models

Camcorder Rotate to Zoom.

My favorite new feature is Camcorder Rotate to Zoom. It lets users zoom in and out while recording by gently rotating the wrist in camcorder mode, making video zoom feel super easy and natural. It only works for video capture when the phone is folded at 90 degrees, also known as “Camcorder mode.” I wish this feature were also available in photo mode when the device is unfolded.

Frame Match solves a typical travel-photo problem: handing your phone to someone else and getting back a poorly framed shot. Users first lock the desired frame in a reference shot, then the other person aligns the on-screen guide with the original image before taking the photo.

Google Photos Wardrobe can identify clothing items in the user’s Google Photos library, organize them, create clean item snapshots, and help mix and match outfits or virtually try them on. I saw the demo, and it worked pretty well. This solves a common problem many of us have: deciding what to wear every morning. In the demo, the feature was accessible from the Google Photos Collections area at the bottom of the screen.

Daily Drops is a personalized content feed that can surface reminders, weather, headlines, and Google Photos Memories directly on the device. Users can opt out of the feature.

AI Features

Moto AI, introduced last year, includes features such as Catch Me Up for notification summaries, Next Move for contextual suggestions, and global search across apps, contacts, calendar information, and messages.
The phones also support Google Gemini, Gemini Live, Microsoft Copilot, and Perplexity, giving users several AI options rather than locking them into a single assistant.

Razr Ultra

Design and durability

The Razr Ultra in PANTONE Orient Blue with a finely embossed Alcantara texture (left), and PANTONE Cocoa with a natural wood veneer finish

The flagship Razr Ultra comes in PANTONE Orient Blue with a finely embossed Alcantara texture, and PANTONE Cocoa with a natural wood veneer finish. Both are meant to give the phone a more fashion-oriented feel than a typical glass slab smartphone. Alcantara is a synthetic suede-like material that mimics the look and feel of luxury suede. It is widely used in high-end car interiors, fashion, and accessories.

Durability also gets attention. The Razr Ultra has a titanium-reinforced hinge and uses Corning Gorilla Glass Ceramic 3 on the external display. The phone is also IP48-rated, for protection against larger particles, and can survive submersion in up to 1.5 meters of water for 30 minutes.

Displays

The Razr Ultra has a 4.0-inch external display with a 165Hz refresh rate and up to 3000 nits of peak brightness. It can run full apps, show notifications, play media, display widgets, and support video wallpapers without opening the phone.
Inside, the Razr Ultra opens to a 7.0-inch Extreme AMOLED display with 10-bit color, Pantone-validated colors and skin tones, 165Hz refresh rate, and up to 5000 nits HDR peak brightness, an improvement over last year’s 3000 nits.

Cameras

The Ultra has the strongest camera system in the lineup, with a triple 50MP setup. The main upgrade is a new 50MP LOFIC sensor designed to capture up to 6x more dynamic range than the previous generation, which should help preserve detail in bright skies, harsh sunlight, and darker areas of a scene.
It also includes a 50MP Ultrawide + Macro Vision camera with a 122-degree field of view, plus a 50MP internal camera.

Flex View remains one of the most useful parts of the Razr camera experience, letting the phone act like its own tripod for hands-free shots, selfies, and group photos.

Video and Dolby

The Razr Ultra supports Dolby Vision Capture, giving users a more cinematic video option directly from the phone. Dolby-enhanced audio and Snapdragon Sound are also part of the entertainment package.

Razr Plus

Design and durability

The Razr Plus sits between the Ultra and the base Razr. Its 2026 color is PANTONE Mountain View, a deep green shade with a new woven-inspired jacquard finish. The phone also uses a titanium-reinforced hinge and Corning Gorilla Glass Victus for protection against everyday scratches and drops. Like the rest of the lineup, it has IP48 protection.

The Razr Plus comes in PANTONE Mountain View with a a new woven-inspired jacquard finish

Displays

The Razr Plus gets the same 4.0-inch external display size as the Ultra, with full app access, widgets, notifications, video wallpapers, and quick interactions without opening the phone.
Inside, it features a 6.9-inch Extreme AMOLED display with Dolby Vision support.

This is likely the sweet spot of the lineup: a large cover screen, a big internal display, and a premium foldable experience without the Ultra price.

Cameras

The Razr Plus also sports a main dual-camera system with two 50MP rear cameras. The main camera includes OIS, Quad Pixel technology, Instant All-Pixel PDAF, and Pantone-validated color and skin tones. The second rear camera is a 50MP Ultrawide + Macro Vision camera, and the internal camera is 32MP.

Video and Dolby

The Razr Plus supports Dolby Vision on its internal display and Dolby Atmos content on supported platforms through speakers or headphones. It also benefits from the new video-related software features shared across the lineup.

Razr

Design and durability

The base Razr is the most affordable model in the 2026 family. It comes in four new Pantone colors and four finishes: Hematite in a new woven-inspired finish, Bright White in Acetate, Violet Ice, and Sporting Green, both in a leather-inspired finish.
This model may be the most interesting for a lot of people because it brings several excellent foldable features to a lower price. It uses Corning Gorilla Glass Victus, has the same titanium-reinforced hinge approach as the rest of the family, and is the first flip phone in its category to meet MIL-STD-810H durability standards. I found it impressive how the Razr 2026 blends military-grade toughness with high-fashion style in a slim flip phone.

"THE RAZR 2026 BLENDS MILITARY-GRADE TOUGHNESS WITH HIGH-FASHION STYLE IN A SLIM FLIP PHONE."

Displays

The Razr has a 3.6-inch external display, smaller than the 4.0-inch displays on the Ultra and Plus, but still large enough to run apps, reply to messages, check notifications, and access widgets.
Inside, it opens to a 6.9-inch Extreme AMOLED display. This keeps the everyday Razr experience closer to the higher-end models than its price might suggest.

Cameras

The base Razr also gets a main dual-camera system with a 50MP primary camera and a 50MP Ultrawide + Macro Vision camera. That ultrawide camera is a major upgrade from last year’s 13MP ultrawide sensor. The internal camera is 32MP.

Video and Dolby

The Razr has stereo speakers enhanced with Dolby technology. Unlike the Ultra and Plus, it is not positioned around Dolby Vision capture or Dolby Vision display playback, but it still gets the core camera software upgrades that make the Razr family more usable for everyday shooting.

Prices And Availabilit

In the United States, all three 2026 Razr flip phones will be available for pre-order on May 14 and go on sale unlocked on May 21.
The Razr Ultra (2026) starts at $1,499.99. The Razr Plus (2026) starts at $1,099.99. The Razr (2026) starts at $799.99.
Unlocked models will be available through Motorola.com, Best Buy, and Amazon. Carrier availability varies by model, with the base Razr getting the widest carrier rollout at launch.

Could humanoid robots soon handle luggage, clean cabins, or take over other airport jobs? Japan Airlines is about to test that question at Tokyo’s Haneda Airport (official press release), where humanoid robots will be evaluated for ground-handling operations starting in May 2026.

Reports (watch.impress.co.jp) state that this is not a mass deployment, and JAL is not saying that robots are about to replace airport workers overnight. But the trial still lands in the middle of a much bigger anxiety: robots are no longer just factory machines or cute showroom demos. They are slowly moving into real workplaces, including the physical jobs that many people once assumed would be difficult to automate. This time, JAL is using what seems like a Unitree humanoid robot, possibly the G1.

That is what makes this project interesting. Airports have used robots before, but many of those experiments involved safer, more controlled roles such as giving directions, assisting passengers, cleaning floors, or moving through indoor spaces. Ground handling is different. This is the work that happens around aircraft, luggage, equipment, schedules, and safety rules. It is closer to the kind of human labor people worry about when they hear that robots are coming for jobs.

Official media image from the JAL press release

JAL says the project will look at where humanoid robots can fit into airport operations, with possible future targets including baggage loading, cabin cleaning, and even the operation of ground support equipment. They are real airport jobs, tied directly to whether flights leave on time and whether airport workers can keep up with growing demand.

For travelers, the idea is both fascinating and slightly unsettling. A robot that helps move luggage or clean cabins could eventually make airport operations smoother. Delayed cleaning, late baggage handling, or slow aircraft turnaround can ripple into delayed departures. If robots can take over even small parts of that process, passengers may feel the effects without ever seeing the machines.

For workers, the question is more serious. JAL frames the project around reducing workload and supporting operations, not replacing employees. It would be naive to pretend that labor-saving automation has no long-term job implications. The purpose of a trial like this is to find out which tasks machines can safely and repeatedly perform. Once that answer becomes clear, the workplace can, and will, change.

Japan faces long-term labor pressure, this is not new and it’s something we talked about for many years. That’s why Japanese companies have been experimenting with automation with that goal in mind, with some success, for almost two decades. The rise of AI finally makes humanoids possible and useful.

In parallel, inbound tourism has surged. More travelers usually means more bags, more cabin cleaning, more aircraft turns, and more strain on the people keeping airports moving. The Haneda trial should not be dismissed as a gimmick, but it should not be overhyped either. Airport jobs are not disappearing tomorrow. Still, JAL’s experiment shows that humanoid robots are moving closer to the real workplace. The question is how much human work they can eventually take over.

Apple has announced a new subscription model for the App Store designed to bridge the gap between flexible monthly payments and the discounted rates typically associated with annual plans. This update allows developers to offer a 12-month commitment option, where users pay a reduced monthly fee in exchange for agreeing to a full year of service.

Under this new structure, users can benefit from the lower pricing usually reserved for upfront annual payments without the need for a large initial expenditure. While the payments are processed monthly, the user is bound by a contract to complete all twelve installments. Apple specifies that while a subscription can be canceled at any time to prevent it from renewing for a subsequent year, the monthly payments must continue until the current 12-month obligation is fulfilled.

If a payment fails during this period, the user will lose access to the service. To maintain transparency, Apple’s system will track the number of completed and remaining payments within the user’s account settings. Furthermore, the company will issue automated email and push notification reminders as the renewal date for the next year-long commitment approaches.

Technical Deployment and Global Availability

Developers can begin configuring these options through App Store Connect and testing them in Xcode immediately. The feature is scheduled for a global rollout next month alongside the release of iOS 26.5 and corresponding updates for iPadOS, macOS Tahoe, tvOS, and visionOS.

Notably, Apple has stated that this specific billing option will not be available in the United States or Singapore at launch, with no timeline provided for those regions. For users on slightly older software, the feature will maintain backward compatibility with the 26.4 versions of Apple’s operating systems.

Corporate Context

This shift in App Store policy coincides with a significant leadership transition within the company. Long-time CEO Tim Cook is slated to step down after nearly 15 years, with John Ternus, the current Senior Vice President of Hardware Engineering, set to succeed him. This new subscription model represents one of the final service-layer adjustments under the current administration as the company prepares for a new era of hardware, including the anticipated iPhone Fold.

WhatsApp is reportedly developing its own proprietary cloud storage infrastructure, signaling a major shift away from its long-standing reliance on Google Drive and iCloud. Currently, users on Android and iOS are tethered to the storage ecosystems of Google and Apple, respectively. This dependency often creates logistical hurdles, particularly when free storage tiers become saturated by shared data like photos and emails, forcing users to manage competing file priorities or pay for additional space.

According to recent findings by WABetaInfo, this upcoming update aims to provide users with greater autonomy. By offering an internal storage alternative, WhatsApp will allow individuals to choose between traditional third-party services or the platform’s native infrastructure to safeguard their chat histories and media.

A primary pillar of this transition is enhanced security. Backups stored within WhatsApp’s native cloud will feature mandatory end-to-end encryption by default. To streamline the user experience, the platform is integrating modern authentication methods. Rather than relying solely on complex, manual passwords, users will be able to secure their data using biometric tools, such as facial recognition or fingerprint scanning. For those who prefer traditional methods, the application will still support standard passwords and 64-digit encryption keys.

Storage Tiers and Pricing

Initial reports suggest that WhatsApp plans to offer a free storage tier of up to 2 GB on its own cloud. However, it remains unclear whether this limit applies to all users or if it is a perk reserved for “WhatsApp Plus” subscribers. To accommodate users with larger data footprints, the company is considering a paid tier providing 50 GB of storage for an estimated cost of $0.99 per month.

This feature is currently in the development phase and does not yet have an official release date. It joins a suite of ongoing updates, including a redesigned status interface and new tools to limit commercial messaging, as the platform continues to refine its ecosystem for a global audience.

The OnePlus Ace 6 Ultra has officially debuted in China, positioning itself as a high-performance flagship specifically engineered for the gaming market. The device arrives with a robust suite of technical specifications and a dedicated ecosystem of accessories designed to enhance the mobile gaming experience.

At its core, the Ace 6 Ultra features the MediaTek Dimensity 9500 platform, supported by either 12 GB or 16 GB of LPDDR5X RAM. Storage options are expansive, reaching up to 1 TB of UFS 4.1 capacity. To sustain prolonged gaming sessions, OnePlus has integrated a massive 8,600mAh battery that supports 120W fast charging, running on Android 16 with the ColorOS 16 interface.

The visual experience is centered on a 6.78-inch OLED LPTS display manufactured by BOE. This screen offers a 1.5K resolution, a dynamic 165Hz refresh rate, and an industry-leading 4,000Hz touch sampling rate, ensuring near-instantaneous input response.

Thermal Management and Durability

To manage the heat generated by its powerful internal components, the device utilizes a 6,000mm² vapor chamber cooling system. This is paired with the Wind Chaser Gaming Kernel, a software optimization layer designed to minimize thermal throttling without compromising performance.

Externally, the smartphone features an all-metal structure with an IP69K rating, providing extreme resistance against high-pressure water jets and dust. While the device prioritizes gaming over photography, it still includes a functional camera setup consisting of a 50 MP primary sensor with optical image stabilization (OIS) and an 8 MP ultrawide lens.

Gaming Ecosystem and Availability

OnePlus also introduced the Strix Gaming Controller alongside the phone. This gamepad features additional customizable buttons, a low latency of just 1.8 ms, and compatibility with a magnetic cooling module.

In China, the OnePlus Ace 6 Ultra is priced starting at 3,499 yuan (~$485 USD) for the base 12/256 GB model, scaling up to 5,099 yuan (~$747 USD) for the 1 TB variant. Currently, there is no official information regarding a global release.

Adobe is expanding its footprint in the digital creation space with the launch of the Firefly AI Assistant in public beta. This new intelligent agent is designed to function as a deeply integrated layer across the entire Creative Cloud suite, aiming to centralize complex operations within a single, conversational interface. By doing so, Adobe intends to reduce the need for users to manually toggle between various applications to complete a project.

Cross-App Integration

The assistant acts as a coordinator for over 60 different Adobe tools. This means that tasks historically requiring specialized technical expertise in Photoshop, Premiere Pro, Illustrator, or Lightroom can now be initiated through simple text commands. The AI bridges the gap between these programs, allowing for a more fluid transfer of assets and edits without traditional workflow friction.

Automation via Creative Skills

To further streamline productivity, Adobe has introduced “Creative Skills“—pre-configured, automated workflows for common industry tasks. These include:

• Batch image editing and portrait retouching.

• Automated creation of semantic mood boards.

• Rapid generation of social media content.

Despite this high level of automation, Adobe emphasizes that the user remains in control. The system allows for manual intervention at any stage, ensuring that professional creators can still apply fine-tuned adjustments and maintain their artistic flexibility.

Personalized Learning and Efficiency

A standout feature of the Firefly AI Assistant is its ability to learn from the user. Over time, the assistant identifies individual preferences, such as specific visual styles, frequently used tools, and recurring work patterns.

This enables the AI to provide increasingly personalized results and suggestions, effectively acting as a digital co-pilot that adapts to the creator’s unique methodology.

The scary version of this story is easy to understand: an AI coding assistant deleted a company’s live data and even seemed to admit what it had done.

That sounds like a “rogue AI” moment. But the more important lesson is less dramatic and more worrying: the AI was apparently able to delete the data because the system gave it too much access in the first place.

According to PocketOS founder Jer Crane, the AI agent was supposed to be working in a test environment, not on the company’s real production system. But when it ran into a credential problem, it allegedly found another access token and used it to delete the company’s production data.

For most people, the technical details are not the point. The plain-English version is this: the AI did not break into the system like a hacker in a movie. It used keys that were already lying around.

That is why this story matters beyond the software world. Companies are now giving AI tools the ability to do real work, not just write text or summarize emails. These tools can change code, touch business systems, connect to cloud services, and in some cases affect live customer data. When the permissions are too broad, a mistake can move very quickly.

The issue is not that the AI became evil. The issue is that it was treated like a trusted operator before the safety rules were strong enough.

A human employee deleting a company’s live database would usually face several points of friction. There might be a warning, a second approval, a manager involved, or at least a moment of hesitation. An AI agent can move through a task in seconds if the system allows it. That speed is useful when the job is safe. It becomes dangerous when the tool has access to something critical.

Backups are another part of the story. Many people assume that if a company has backups, the data is safe. But backups only help if they are truly separate from the thing being deleted. In this case, Railway’s (a cloud computing provider) documentation reportedly indicated that deleting a storage volume also deleted the related backups. That means the safety net was not as independent as many people would expect.

Railway later restored the data and reportedly changed the system so similar deletes would be delayed. That is good, but it does not change the larger point. AI tools are only as safe as the systems around them.

For regular users, the concern is simple. If companies are going to let AI touch websites, apps, customer records, payment systems, or other important services, they need stronger guardrails. AI should not automatically get access to everything just because it is useful. It should get the minimum access needed for the job, and dangerous actions should require extra checks.

This is the real-world AI risk most people should care about. Not robots taking over, and not science-fiction machines making secret plans. The immediate risk is much more ordinary: companies moving fast, giving AI too much permission, and discovering too late that the safety locks were not ready.

The AI did not need to go rogue. It only needed the keys.

Leica has officially introduced a new aesthetic variant for its renowned Leica M-A analog camera, characterized by a distinct texture and extremely limited availability. Named the Leica M-A Hammertone Limited Edition, the release features a specialized finish designed to simulate hand-forged metal. This launch commemorates the 20th anniversary of the Leica Store in Ginza, Japan, and is restricted to a production run of only 100 units, positioning it as a high-value item for horological and photographic collectors.

Design and Aesthetics

The Hammertone edition is distinguished by its body painted in a high-resistance gray “hammer” lacquer, which provides a textured, industrial appearance. To complement the metallic finish, the camera is wrapped in black synthetic leather. Commemorative markings, specifically the inscription “20 Jahre,” are engraved on the body to signify the two decades of the Ginza flagship’s operation.

Technical Specifications

Technically, the device remains identical to the standard Leica M-A rangefinder. As a purely mechanical instrument, it operates without the need for batteries or electronic components. It lacks modern automation features such as autofocus, auto-exposure, and an internal light meter, requiring the photographer to manually manage all settings. Key mechanical specifications include:

• Mechanical focal adjustment with exposure times up to 1/1,000s.

• Flash synchronization capabilities at speeds up to 1/50s.

• Traditional analog rangefinder viewing system.

Market Availability

Given its commemorative nature, the Leica M-A Hammertone is an exclusive release for the Japanese market, available only through Leica retail locations within the country. While Leica has not officially disclosed the specific retail price for this limited run, industry expectations suggest it will exceed the $7,000 USD entry point of the standard model. This release coincides with broader brand developments, including the ongoing partnership between Leica and Xiaomi, which aims to integrate premium camera interfaces into mobile devices.

Internally, the smartphone is powered by the MediaTek Dimensity 7300e mobile platform. Consumers can choose between configurations offering 8 GB or 12 GB of RAM, paired with either 256 GB or 512 GB of internal storage. The device runs on Android 16, utilizing vivo’s proprietary OriginOS 6 interface. For durability, the Y600 Pro carries both IP68 and IP69 certifications, indicating high resistance to dust and water submersion, as well as protection against high-pressure, high-temperature water jets.

The visual experience is centered on a 6.83-inch OLED display featuring a 1.5K resolution ($2800 \times 1260$ pixels) and a 120 Hz refresh rate. A significant focus has been placed on user health, as the panel includes official eye-protection certification designed for individuals who engage in prolonged screen time. Regarding photography, the rear of the device features a circular camera module housing a 50 MP main sensor, while a 32 MP front-facing camera handles high-definition selfies.

The vivo Y600 Pro is currently in pre-order in China, with commercial release scheduled for May 5th. It will be available in four color options: Floating Gold, Bright Moon Black, Vast Blue, and Starry Purple.

Pricing starts at 2099 yuan (~$306) for the 8GB + 128GB model; the 8GB + 256GB variant is priced at 2299 yuan (~$336), while both the 8GB + 512GB and 12GB + 256GB options will be sold for 2599 yuan (~$379). The top-tier 12GB + 512GB variant will be offered for 2899 yuan (~$423). No word on dates or prices for other markets yet.

There is a striking twist in the AI rush: In some high-usage enterprise deployments, the systems meant to improve efficiency are also becoming a major cost themselves. Axios reported that some companies and teams see AI compute costs rivaling or exceeding the salary costs of the employees using the tools. That is of course not a common situation but it is still a meaningful signal about how expensive heavy AI usage can become.

The most notable example in the report came from Nvidia. According to Axios, Bryan Catanzaro, Nvidia’s vice president of applied deep learning, said that for his team, “the cost of compute is far beyond the costs of the employees.” The comment stood out because it came from a senior Nvidia executive. Nvidia is one of the companies most closely tied to the AI infrastructure boom, so an internal remark like that offers a useful reality check on the economics of large-scale AI use.

The reason is not hard to understand, even if the exact bill varies by company and workflow. Large models require substantial compute, and intensive use of coding agents or developer assistants can drive up token consumption quickly. A light chatbot workflow, such as drafting emails or summarizing notes, is one thing. A coding workflow that repeatedly analyzes code, generates revisions, and runs multiple prompt cycles and tests is another. 

Axios also pointed to a report from The Information about Uber’s internal AI tooling costs. According to that report, Uber’s 2026 budget for AI coding tools had already been exhausted early in the year, with CTO Praveen Neppalli Naga saying the budget had been “blown away already.” 

AI has often been framed as a way to amplify labor, automate repetitive work, or in some cases reduce headcount pressure. In practice, at least in some high-intensity deployments, it is also becoming a new infrastructure cost center. That does not mean the productivity upside is fictional. It means the old assumption that more automation automatically leads to lower costs is proving too simplistic, even if true in many instances.

The direction is clear enough. AI infrastructure, software, and cloud services are pushing enterprise technology budgets higher, and buyers are under more pressure to justify where that spending goes.

The more real AI becomes inside large organizations, the more it gets judged like any other enterprise tool. Does it save time, improve output, reduce bottlenecks, or create enough business value to offset its cost? Those questions are less glamorous than product demos, but they are the ones that decide whether a technology becomes durable infrastructure or just an expensive experiment.

There’s one thing most people overlook in the conversation: computing efficiency is going to skyrocket in the coming months and years. Tokens will be generated at a much cheaper price, and tasks that have reached a “good enough” level will cost dramatically less, freeing budget for higher-value workloads.

AI-generated data center, for illustration only

Utah officials are weighing a data center and energy project so large that it is difficult to describe as a normal industrial development.

The Stratos Project, also being marketed as “Wonder Valley,” is planned for rural Box Elder County and could cover as much as 40,000 acres. Public reporting and meeting materials describe a mixed-use development built around hyperscale data centers, power production, manufacturing, housing and commercial uses. That combination alone would make it significant, but the energy numbers are what have pushed the proposal into a different category.

According to reporting cited in the project materials (PDF link), Phase 1 could generate about 3 gigawatts of energy, while the full project could reach 9 gigawatts of capacity.

9 gigawatts running continuously would equal about 78.8 terawatt-hours of electricity over a year. PacifiCorp’s 2025 Integrated Resource Plan update shows Utah retail sales at about 27.7 terawatt-hours in 2026 and 28.4 terawatt-hours in 2027.

That does not mean Stratos would necessarily consume more electricity than the state uses today, because capacity, generation and actual delivered load are not interchangeable. But it does show why the proposal is being discussed in grid-scale terms rather than as a routine industrial park. (pacificorp.com)

Still, that scale is also why people are worried. Residents have already raised concerns about water use, pollution, noise, land impacts and the speed of the approval process. Those concerns are not unusual for large data center projects, especially in places where power, cooling and water questions can affect nearby communities. But in this case, the size of the proposal makes the questions more urgent.

County documents reportedly frame the project as useful for large data centers and advanced manufacturers connected to military and national-defense-related missions. But it may also make some residents more concerned about transparency, because this is not just a conventional commercial development moving through a familiar local planning process.

Google has initiated a significant design overhaul for its application icons, introducing a new visual language defined by gradient effects and modernized shapes. Following an official design update confirmed in May of last year, the Mountain View-based company is now expanding this aesthetic shift to its broader ecosystem, with a specific focus on the Google Workspace suite.

Strategic Shift in Visual Identity

Google Apps icons design before and after

Reports from 9to5Google highlight a new wave of updates encompassing 13 distinct icons. The primary objective appears to be twofold: implementing sophisticated gradient textures and streamlining the overall look to feel more contemporary.

One of the most notable aspects of this redesign is the departure from a rigid adherence to Google’s four-color palette across every service. In previous iterations, apps like Drive, Calendar, and Meet utilized a standardized combination of blue, red, yellow, and green. The new versions move toward a more personalized, specialized approach for each service. For instance, the Google Drive icon has notably removed its red element, while Google Meet, Chat, and Calendar have shifted toward a more “monochromatic” or simplified color scheme to reduce visual repetition and help users distinguish between tools more quickly.

Impacted Applications

The productivity suite—Google Drive, Docs, Sheets, and Slides—received the most substantial updates. Beyond standard refinements, the Google Keep icon underwent a significant transformation, moving away from its traditional lightbulb-on-paper look toward a cleaner design. Furthermore, Google Forms and Google Sites have received entirely new, dedicated icons; previously, these services did not have unique standalone branding within the ecosystem.

While these changes aim to provide a more cohesive and professional feel for enterprise users, the new aesthetic is expected to divide opinions among those accustomed to the classic look. This rollout marks the latest phase in Google’s long-term strategy to refresh its interface as it competes with other cloud-based productivity suites. The updates are being deployed gradually across web and mobile platforms throughout early 2026.

MediaTek has expanded its mid-range portfolio with the quiet introduction of two new 5G-enabled chipsets: the Dimensity 7450 and Dimensity 7450X. These processors are designed to bring high-end features, including 200 MP camera support and advanced artificial intelligence, to more accessible smartphone segments.

Both chipsets utilize an octa-core “big.LITTLE” configuration. The cluster consists of four Cortex-A78 performance cores clocked at up to 2.6GHz and four Cortex-A55 efficiency cores. Graphics are handled by the Arm Mali-G615 MC2 GPU.

A key highlight is the integration of MediaTek’s 6th-generation NPU, which delivers a 7% improvement in efficiency for AI-driven tasks. This enhancement specifically benefits computational photography and system-level optimizations. In the imaging department, the Imagiq 950 ISP supports sensors up to 200 MP. It includes hardware-level Precise Noise Reduction (MCNR), face detection, and video HDR, alongside support for Google Ultra HDR to maintain high-fidelity light intensity data.

Gaming and Specialized Design

For mobile gaming, the chips feature MediaTek Adaptive Gaming 3.0 technology. This system monitors device thermals and operational status in real-time to balance visual quality with performance, ensuring stable frame rates while preserving battery life.

While the technical specifications are nearly identical, the distinction between the two models lies in their intended form factors. According to industry reports, the Dimensity 7450 is optimized for standard “slab” smartphones, such as the Motorola Edge and OPPO A series. Conversely, the Dimensity 7450X is specifically tailored for foldable devices, such as the Motorola razr and OPPO Find lines. This specialized variant is expected to debut in the upcoming Motorola razr 70, scheduled for release later this week.

Motorola razr 60’s successor could be one of the first devices to get MediaTek’s new SoC Dimensity 7450X

Connectivity and Support

The platform supports modern memory standards, including LPDDR5 (up to 6400 Mbps) and UFS 3.1 storage. Connectivity is robust, featuring 5G download speeds up to 3.27 Gbps, Wi-Fi 6E, and Bluetooth 5.4. Display support ranges from WFHD+ at 120Hz to Full HD+ at 144Hz, providing manufacturers with flexibility for various screen configurations.