I wrote this entire article while seated on an airplane experiencing unusually high turbulence. The software I used to spell-check and grammatically sanitize the draft was built at an airport. The language engine is running entirely on my Mac, fully offline, fixing all my typos and removing the double spaces while I mash the keyboard and sip a sugar-bomb coffee.
Also, I don’t know how to code. I didn’t write a single line of code, and yet, the Mac software I am using right now looks classier and feels snappier than Grammarly ever did. Grammarly, if you don’t know, is one of the most popular apps for spelling and grammar checking on the planet. So, how did I do it? I asked Claude. I narrated my wish, it asked my preferences, and in less than 30 minutes, I built myself a no-internet Grammarly replacement while also avoiding the “yet-another-subscription” curse.
The first version runs as its own website in a browser tab without any internet connection. The second version lives as a Chrome extension, and works just fine when the device is offline. And finally, I went ahead and created myself a full-fledged Mac app that lives as a menu bar utility. I had half a dozen people test these across Mac and Windows machines. They were pretty impressed by the speed and accuracy of the tool.
Solving my problems, or taming my vices
Over the past couple of weeks, I have dipped my toes (the whole feet, my arm, and my neck) into the app and software development world. My first experiment was a Mac app that used the motion sensors fitted inside the AirPods Pro to track my posture. Essentially, the app identifies a healthy posture, and every time I slouch or hunch forward, it sends a warning. The app worked fully offline, and all the processing happens entirely on my Mac.
I did it all without even seeing the underlying code.
For my next adventure, I thought of building something that can liberate me from the hassle of being perennially online. Since my job as an editor involves writing all day, Grammarly is nothing short of a lifeline, despite some of its shady shenanigans recently. I have often lost chunks of text in the Grammarly iPad app because the cloud sync didn’t work. On the days when I am carrying my trusty Mac, finding a stable internet connection becomes a constant headache.
All I want is to sit in peace, write a few articles, and have a reliable grammar-checking tool do its magic as I furiously speed through a laptop’s mushy keyboard. But that’s not the only issue. Staying connected to the internet (while it does its job as a Chrome extension in Google Docs) also means an endless stream of distracting apps. And yeah, it’s just one browser tab to take you from a strictly-for-work Google Search and deep into a doomscrolling bender on X, YouTube, or any other digital vice.
I am still in disbelief.
But why, you might ask? First, because I can finally build my own software tailored to my specific needs. Second, I am tired of the privacy compromises. Of course, not having to pay for yet another app is a huge incentive. But above all, it’s the realization of being in control of your daily workflow (or at least a part of it) that really inspires me, and I am not stopping anytime soon.
How was the process?
I’ll let the picture below do the talking. I fired up the Claude mobile app on an Android tablet to build a piece of software for the Mac. And in the very first attempt, my Grammarly alternative built by Claude worked perfectly. Actually, I built three. One of them runs fully offline, the other one requires an internet connection, while the third one is a proper macOS app that lives as a menu bar utility.
The latter combines the concept of note-taking and grammar correction into a single tool, complete with a one-click Apple Notes export facility. This one required me to fire up Xcode and build a dedicated app icon, too, which Claude was happy to do. But here is the broad reality. The barrier for “building” products has never been easier, and it has never been more versatile. A year ago, if you told me that I would make such a powerful macOS tool on an Android tablet, without writing a line of code, I would’ve laughed out loud in your face.
When I launched Claude, I simply narrated my requirements. Claude suggested three routes for building my own offline Grammar checker, and I picked the one that worked without an internet connection, ran with minimal issues, and above all, must be snappy. I initially wanted to use Apple’s on-device foundational models (shipped with Apple Intelligence). It worked, but it was pretty slow. Claude recommended that I go with the Harper engine by Automattic. Initially, the tool, which I named InkWell, required an internet connection.
The whole package came in at less than 10 megabytes in size. But I wanted something that can work without an internet connection. Claude suggested that it can actually integrate the Harper engine within the file package so that the tool (now called Quill) won’t require an internet connection at all. The only downside? The file size swells to 25 megabytes. The AI’s storage space concern was almost laughable. We live in an age where even basic calculator apps hog hundreds of megabytes in local storage, so 25MB was next to nothing.
The bigger AI dilemma
So, I have been using Grammarly for years now, but in the past couple of years, as the company has increasingly leaned into AI, some of the suggestions and grammar recommendations have turned out to be utterly vexing. And in a fair number of cases, I have also noticed that Grammarly does a downright shoddy job with even the most basic kind of spell checks.
Harper avoids the verbose recommendations and increasingly AI-fied language suggestions that you would get from the likes of ChatGPT and Gemini. It’s a tool that doesn’t rely on token-based linguistic predictions, but more on hard-coded language and grammar rules. This is exactly what I want.
I want an AI to catch a wrong spelling, or an incoherent tense here and there. That’s all. I don’t want (and would never take) an AI’s recommendation to change my voice. Harper, however, is not perfect, and so are the tools I built using the namesake engine. For example, when I check this sentence, “My name John. What your name? What day today,” the Harper engine flags it as perfectly fine. But these errors are sporadic.
On the positive side, the Harper engine takes only 20 milliseconds to identify mistakes and make language suggestions. “Harper combines lightning-fast performance with a privacy-first design, ensuring all processing happens locally on your device to keep user data completely private. It’s also completely open source,” says Automattic, the company which acquired Harper in 2024.
Those are meaningful perks, and the fact that it can run it within such a small package and without any internet connection is a massive sigh of relief. Of course, building a tool in three different functional flavors, and without writing any code, is the biggest takeaway here. Claude is putting the power in the hands of an avereage computer user like me, and you. Naturally, I can’t wait to embark on my next personal app project with it
Pope Leo XIV signed his inaugural encyclical on May 15, marking the 135th anniversary of Pope Leo XIII’s Rerum novarum. Titled Magnifica humanitas, the document was released on May 25 and tackles one of the era’s most pressing issues: artificial intelligence and its effects on humanity.
The central message is not an opposition to technology. The Pope stresses that technology is neither a threat nor inherently evil, but it is never neutral because it reflects the values of its creators, funders, and controllers. This is where the discussion becomes crucial.
Is AI turning into a tool for the elite?
A large portion of the encyclical addresses the concentration of AI power in the hands of a few. Pope Leo XIV cautions that this could widen the divide between those who benefit from the digital revolution and those left behind. He calls for ethical standards shaped by shared principles of social justice rather than a small group of decision‑makers.
He also condemns the use of AI in warfare, stating that no algorithm can render war morally acceptable. He argues that AI makes conflict faster, more impersonal, and lowers the threshold for resorting to violence, which does not constitute genuine progress.
The timing of the encyclical is striking. In July 2025, the U.S. Department of Defense awarded contracts to Google, OpenAI, Anthropic, and xAI to embed cutting‑edge AI into military operations, from battlefield decision‑making to classified intelligence systems. Anthropic withdrew from its agreement in early 2026 over concerns about autonomous weapons and mass surveillance, leading to a ban by the Trump administration. OpenAI quickly stepped in, signing its own Pentagon deal, which sparked a massive backlash from users and resulted in widespread app uninstallations.
By April 2026, the Pentagon had secured new classified AI contracts with OpenAI, Google, and others. It is evident that the Pope is displeased with these developments and is indirectly urging the U.S. government to cease employing AI in warfare.
What path does the Pope propose?
The heart of the encyclical is the call for technology to serve humanity, not the reverse. He advocates for stronger labor protections, transparency in communication, renewed focus on education, and legal frameworks that hold AI developers accountable. He highlights migrants, workers in hazardous conditions, and abuse victims as groups that require our empathy and protection. Ultimately, Magnifica humanitas reminds us that, regardless of technological advancement, true progress remains a human measure.
If you’ve ever tried to reserve a room online and felt uneasy about the AI chatbot offering assistance, you’re not imagining it. A recent study from Texas A&M’s College of Agriculture and Life Sciences confirms that hotel‑booking bots are genuinely unsettling users, and this anxiety is actually reducing bookings.
What makes hotel chatbots feel creepy?
Researchers surveyed 340 adults in the UK who had used chatbots for hotel reservations and identified three primary drivers of the “ick” factor: inaccuracy, deceptive behavior, and intrusiveness. Inaccuracy was by far the biggest culprit, provoking a negative reaction more than four times stronger than the other issues.
Errors include quoting the wrong room rates, botching cancellation policies, or simply ignoring user questions. This discomfort isn’t just a feeling—it slashed users’ willingness to continue the conversation by nearly 38 % and almost doubled the likelihood that they would postpone or abandon the booking altogether.
The study also highlighted the “uncanny valley” effect, where a bot’s attempts to sound human become even more unsettling when it fails to act like one. Lead researcher Babak Taheri summed it up, noting that when a human‑like system doesn’t behave accordingly, it triggers a reaction deeper than mere disappointment.
A simple fix hotels are overlooking
The good news is that the researchers identified an easy remedy that many hotels are ignoring. When a chatbot openly states that it’s an AI, users become considerably more forgiving of its mistakes. A straightforward introduction such as “Hi, I’m your AI assistant” can go a long way.
They also advise making it simpler for users to reach a real person for complex queries and investing in improving the AI so it can reliably handle basic tasks without stumbling.
This research arrives at a pivotal moment, as AI‑driven travel booking is currently the hottest trend in tech. Google has recently added AI trip planning to Search, and Uber has launched hotel bookings through Expedia within its app.
Huawei has revealed what it sees as a new path forward for advanced chips. At the 2026 IEEE International Symposium on Circuits and Systems in Shanghai, Huawei’s He Tingbo introduced the company’s Tau Scaling Law, a new semiconductor principle that Huawei says can guide chip development as traditional Moore’s Law runs into physical and economic limits.
The company says future high-end chips designed under this approach could reach transistor density equivalent to 14 angstroms, or 1.4nm, by 2031.
How Huawei is changing the chip game
1.4nm sounds pretty impressive, but the keyword here is equivalent. Huawei is not saying it has suddenly gained access to the most advanced chipmaking tools in the world, and the company has yet to provide any independent performance data. As of right now, China’s most advanced chipmaking capability is still widely viewed as being around 7nm (like the one powering Huawei’s tri-fold phone). However, the company’s plan is to chase performance through system-level efficiency rather than relying solely on smaller transistors.
Tau Scaling focuses on cutting the time it takes signals and data to move through chips and computing systems, with Huawei’s new LogicFolding architecture. The tech basically shortens critical-path wiring, reduces signal-propagation load, and improves both transistor density and circuit performance.
Which chips will test this first?
HiSilicon, Huawei’s chip subsidiary, is set to use this tech for its latest generation of Kirin chips. These are scheduled to debut in fall 2026 with the new LogicFolding tech. The company also claims it has designed and mass-produced 381 chips over the past six years based on Tau Scaling, covering areas such as smartphones and AI computing.
Aside from this, the company also plans to apply LogicFolding to the Ascend AI chips by 2030, along with large AI clusters used in data centers. While the 1.4nm is headline-grabbing, the Ascend chips carry a larger weight. With Chinese companies looking for alternatives to Nvidia hardware, which is restricted in the region, Huawei’s AI chips are becoming more important. Reuters also notes that Nvidia CEO Jensen Huang recently said the company had “largely conceded” China’s AI chip market to Huawei.
As the US export controls have limited Chinese access to advanced lithography equipment and other critical semiconductor technologies, making conventional progress toward frontier nodes much harder. TSMC currently uses 2nm technology and plans 1.4nm mass production in 2028, while Huawei is trying to reach comparable density through a different design route. So the company is clearly not waiting for Moore’s Law or easing of US restrictions to decide how far its chips can go.
China has launched a national programme that will assign every humanoid robot manufactured in the country a unique digital identity code, effectively a citizen ID, but for bipedal machines (those that can balance and walk/run on two legs).
The initiative, called the Humanoid Full Lifecycle Management Service Platform, was announced on Friday. It is led by the Humanoid Robotics and Embodied Intelligence Standardization committee, which is under China’s Ministry of Industry and Information Technology (via South China Morning Post).
What does a robot ID actually look like?
The machine codes are structured in four parts. While a two-digit national code tracks international shipments and sales, a four-digit manufacturer code keeps a record of the firm that manufactured it.
Then there’s a six-digit product model code that identifies the robot type, along with a 17-digit serial code distinguishing individual units from one another. The purpose is to follow the humanoid robots from the production through its entire working life, all the way to recycling.
The guidelines cover everyone involved in the supply chain, from manufacturers, service providers, sellers, end users, to the recycling facilities. The new system should also accelerate the deployment of humanoid robots in a regulated manner, wherein the manufacturing firm could be held accountable for any malfunctions.
Why is China doing this now?
According to January research from IDC (as mentioned in the report), the global humanoid robot market expanded 508% just last year, with around 18,000 units shipped globally, with the Chinese vendors leading that growth.
China already has over 100 humanoid manufacturers, and more than 28,000 robots across 200 models have already been assigned a digital ID before the public announcement of the Humanoid Full Lifecycle Management Service Platform.
Yu Xiuming, deputy head at the China Electronics Standardization Institute, said the system is designed to address core issues around safety, oversight, and governance. China’s humanoid industry is moving faster than the regulation framework, and the ID system is less a surveillance move and more an industrial infrastructure play, the kind of standardisation required before global scaling.
In a world where information defines influence, Verum Messenger is building a new architecture of digital communication — intelligent, secure, and ready for tomorrow. Here, technology serves not limitations, but possibilities.
Not being part of change. Leading it. Verum Messenger — the future that speaks first.
Modern gaming laptops have lately split into two extremes: massive 16‑inch and 18‑inch desktop replacements, or ultra‑compact 14‑inch units that still feel a bit cramped for serious gaming. That’s why HP’s new HyperX Omen 15 feels like a breath of fresh air, reviving the classic 15‑inch gaming laptop layout with a chassis that stays portable while still housing solid gaming hardware underneath.
HP’s compact HyperX Omen 15 packs RTX 5070 graphics with AMD and Intel options
The new HyperX Omen 15 is offered with either Intel Core Ultra 7 356H and Core Ultra 9 386H processors or AMD Ryzen 7 8745HX and Ryzen 9 8945HX chips, depending on the configuration. All current U.S. models pair these CPUs with RTX 5070 laptop graphics, up to 32 GB DDR5 memory, and PCIe Gen5 SSD storage.
HP is also leaning heavily into the compact premium‑gaming angle here. The laptop sports a 15.3‑inch 16:10 display with a 1600p IPS panel, 180 Hz refresh rate and 500 nits brightness. Buyers can opt for a 2.8K OLED panel at 120 Hz, delivering roughly 1,100 nits peak brightness with HDR.
Interestingly, this machine feels positioned right between HP’s larger Omen 16 lineup and ultra‑portable options like the Omen Transcend 14. In other words, it occupies the classic “middle child” gaming‑laptop niche that used to dominate the market before manufacturers became obsessed with the extremes.
The 15‑inch gaming laptop is suddenly cool again
While a 15.3‑inch screen may not sound dramatically larger than a modern 14‑inch gaming laptop on paper, that extra real‑estate opens the door for meaningful improvements elsewhere. A larger chassis typically translates to better thermals, more breathing room for higher GPU wattages, and potentially lower fan noise during gaming sessions—factors that often matter more than shaving a few millimeters off the design.
That said, at roughly 5.34 pounds, this isn’t exactly an ultra‑portable machine either. Yet it feels far more manageable than the oversized 16‑inch and 18‑inch gaming laptops that dominate the market today. Notably, HP appears to be limiting the new Omen 15 to North America for now, suggesting the company is targeting mainstream U.S. gamers who want strong RTX 5070 performance without lugging around a desktop‑replacement laptop.
For years, Unreal Engine has been the foundation of modern AAA titles. Now Epic is already gearing up for the next chapter, and surprisingly Rocket League is at the forefront. Honestly, that’s pretty cool – after being stuck on Unreal Engine 3 for ages, Rocket League fans finally get a long‑overdue upgrade to a modern engine.
And yes, the teaser looks impressive: cleaner visuals, a more connected ecosystem, and a glimpse of what Epic wants to present as the next era of Unreal. But while the hype train is already leaving the station, I think I’ll stay on the platform a bit longer. Right now, UE6 feels more like a vision statement than a concrete engine reveal.
### Is UE6 a gaming upgrade or an ecosystem upgrade?
So far, Epic hasn’t really detailed what Unreal Engine 6 changes for gamers. Most of the conversation around the reveal focuses on ecosystem integration, creator tools, and Epic’s broader metaverse ambitions.
Tim Sweeney has previously talked about integrating Verse, Fortnite‑style economies, and shared creator experiences directly into the future of Unreal Engine. That sounds great on paper, but what about the issues players are dealing with today?
There’s hardly any talk about optimization, CPU efficiency, shader‑compilation stutter, traversal stutter, or lowering the insane hardware requirements that modern AAA games continue to normalize. At the moment, UE6 feels more like an ecosystem update than a technological leap, and that’s where my skepticism begins. Flashy creator tools are nice, but most gamers would probably prefer smoother frame pacing and fewer stutters first.
### The UE5 honeymoon has already faded
Part of my caution stems from Unreal Engine 5 itself. When UE5 was first unveiled almost five years ago, it truly seemed revolutionary. Nanite and Lumen felt like game‑changing technologies that would redefine visual fidelity across the industry.
Fast forward to today: UE5 games look stunning, but optimization has become one of the biggest complaints. Modern PC gaming increasingly feels built around upscalers first and native rendering second. DLSS, FSR, frame generation, and AI‑assisted performance modes are now treated less as optional bonuses and more as necessities. Instead of engines becoming lighter and more efficient, gamers are expected to brute‑force performance problems with ever‑more expensive hardware.
That’s why the timing of UE6 feels odd to me. We’re only now fully transitioning to UE5, and even then many games still wrestle with shader stutter, inconsistent frame pacing, and heavy CPU overhead. So the question arises: are we actually solving those problems, or just moving on to the next shiny thing?
### Less buzzwords, more optimization, please
Don’t get me wrong—I want UE6 to succeed. Seeing Rocket League finally modernized is genuinely exciting, and part of me hopes Epic uses this generation to clean up the technical headaches that currently frustrate PC gamers. But before I start celebrating{
}Unreal Engine 6, I need to see more than cinematic trailers and ecosystem buzzwords.
Show me better optimization. Show me lower CPU overhead. Show me fewer stutters. Show me games that don’t need AI upscaling just to feel stable on decent hardware. Because gaming doesn’t really need prettier tech demos right now.
Not in 2024, but it’s coming. UE6 = UE5 + Verse + rough deployment parity into Fortnite and into standalone products + metaverse economy + standards + ?? magic TBD.
✔ A virtual card ✔ Instant transfers between users ✔ A modern secure messenger ✔ Apple Pay integration ✔ Contactless payments worldwide ✔ Fast setup without bureaucracy
❌ No European residency permit required ❌ No endless verification checks ❌ No piles of documents
Open it — and use it.
The future of finance and communication is already here. Verum — when freedom matters more than banking rules.
Launching a smartphone brand is already hard enough. Doing it while accidentally exposing customer data to the open internet somehow makes the whole thing even more chaotic. Trump Mobile is now investigating a potential data exposure that may have leaked personal information belonging to thousands of would-be customers who preordered the company’s gold-colored T1 smartphone.
Trump Mobile says customer information may have been exposed online
Trump Mobile says it is investigating a potential customer data exposure with help from independent cybersecurity experts after pre‑order users’ names, addresses, and phone numbers were reportedly left accessible online. The issue reportedly came from a vulnerability linked to a third‑party platform provider, although Trump Mobile told The Guardian it has found no evidence that financial data, Social Security numbers, passwords, or communication records were compromised.
The vulnerability was first flagged by independent researchers and later amplified online by YouTubers including Coffeezilla and penguinz0, who claimed they were able to verify their own leaked information after being contacted by a source with database access. The timing honestly could not be worse for Trump Mobile either. The T1 phone has already faced delays, criticism over its shifting “Made in America” claims, and skepticism around preorder numbers.
The T1 phone launch is off to a rough start
The bigger issue here is not just the leak itself, but the growing pile of awkward questions surrounding Trump Mobile overall. A lot of people already criticized the phone’s design for replacing two of the American flag’s 13 stripes with the company’s branding, and NBC News’ Brian Cheung also suggested the device may simply be a repackaged version of another existing smartphone, with one expert reportedly suggesting it is likely to be the HTC U24 Pro.
We have reached out to the Trump Organization for a statement regarding the alleged customer data leak incident. We are yet to hear back, but we will update the story when we receive a response.
