Author: TechGeeks

Billed as the world’s first AI-powered, high-precision, and versatile 3D scanner, the POP 4 brings real-time AI segmentation and 3D Gaussian Splatting to a hybrid blue-laser-and-infrared platform. As the newest and most advanced member of the POP series, the POP 4 is built to handle everything from intricate geometries to larger workpieces, including black and shiny surfaces that previous POP models couldn’t capture directly — no scanning spray or markers required.

The Revopoint POP 4 Kickstarter campaign is now live, with early-backer pricing and bundle details available here.

For 3D makers and artists, the POP 4 fits naturally into existing workflows. Users can scan a broken bracket to reprint it, capture an existing part for modification, or create a reference shape for a new design. The exported mesh goes straight into a slicer without requiring extensive CAD cleanup, making the POP 4 feel less like a standalone metrology tool and more like a seamless extension of how makers already work.

Real-time AI object segmentation

It is not uncommon for 3D scanning specialists to spend more time on manual trimming than on the scan itself. The POP 4 provides a solid solution with AI-powered object segmentation that identifies the target object in real time and filters out unwanted data such as desks or background elements during capture. Even the operator’s own fingers are excluded from the model when holding the part directly while scanning.

3D Gaussian Splatting anchored to accurate geometry

Most 3DGS captures today come from phone video or DSLR footage — visually impressive, but with geometry underneath that is often unreliable. The POP 4 flips that approach. Splats are built on top of a structured-light scan with up to 0.03 mm accuracy, making the visual result spatially measurable as well.

The outcome is photorealism with translucency, reflections, and view-dependent highlights, without sacrificing dimensional reliability. For rapid prototyping, realistic AR scene capture, or replacing days of manual modeling, it offers a different starting point from current 3DGS pipelines.

A multi-mode scanning system for indoor and outdoor work

The POP 4 integrates blue multi-line laser, near-infrared full-field structured light, and VCSEL structured light into a unified platform. Its five scanning modes are designed to cover most indoor and outdoor scenarios.

Full-Field HD is tuned for detailed point cloud capture at up to 5M points/s (depending on the PC-hardware), while VCSEL Rapid prioritizes fast capture at up to 30 fps and remains stable in direct sunlight up to 100,000 lux. Hybrid HD combines both modes for accurate surface modeling with stable tracking. The 30 cross-laser lines mode is designed for shiny and dark surfaces at up to 2M points/s, while the single-line mode is built for deep holes, crevices, and grilles.

Precision, speed, and scale

In multi-line laser mode, the POP 4 reaches volumetric accuracy of 0.03 mm + 0.05 mm × L (m), with a fused point distance of up to 0.05 mm and single-frame accuracy of 0.03 mm. Scan speeds can reach up to 105 fps, depending on system configuration.

Its working distance ranges from 200 mm to 800 mm depending on the selected mode, supporting close-range blue-laser scanning at 200 mm and larger-area VCSEL scans at distances up to 800 mm.

Outdoor operation is supported in lighting conditions up to 100,000 lux. Detailed blue-laser scans can operate at the full 100,000 lux using the Outdoor Blue Filter included with the KS Special Edition and Software Bundle Rewards. Without the filter, blue-laser scanning supports lighting conditions up to 50,000 lux.

Wireless operation and integrated workflow

The POP 4 is designed for wireless scanning via mobile phones or tablets, powered by the included 5500 mAh battery grip (Premium Edition) that delivers roughly four hours of mobile use.

Data captured in Revo Scan can be transferred with one click into Revo Measure for measurement and inspection tasks or into Revo Design for reverse engineering workflows. Together, the hardware and software workflow positions the POP 4 as more than just a scanning device, but a connected 3D creation ecosystem built for design, prototyping, and reverse engineering workflows.

Pricing and availability

The Revopoint POP 4 launches on Kickstarter on May 7 with introductory pricing offering up to 30% off for early backers. With an MSRP of $919, the POP 4 positions itself as a capability-focused scanner built around hybrid laser-and-infrared optics, real-time AI segmentation, and 3DGS export. Kickstarter pricing tiers and bundle details are available on the campaign page, where backers can also secure early access to the POP 4.

Back the POP 4 on Kickstarter now

Samsung’s foldables have consistently led the market in polish and software, but one complaint has followed the Galaxy Z Fold series for years: battery life. Now, a new leak suggests Samsung may finally be addressing that issue with the upcoming Galaxy Z Fold 8, alongside camera improvements and a thinner design.

According to a report from Greek publication TechManiacs, the Galaxy Z Fold 8 could pack a larger 5,000mAh battery – a notable jump from the 4,400mAh cell used in the Galaxy Z Fold 7. If accurate, this would mark one of the most significant hardware upgrades to Samsung’s book-style foldable lineup in years.

Samsung may finally be catching up on foldable hardware

Battery capacity has increasingly become a weak point for Samsung foldables, especially as Chinese rivals like OnePlus, Honor, Vivo, and Xiaomi continue pushing larger batteries into thinner foldable designs. While Samsung has traditionally prioritised weight, software optimisation, and durability, the company has also been criticised for relying on relatively small batteries compared to competitors.

The new report suggests Samsung is now trying to balance both. The Galaxy Z Fold 8 is tipped to feature the Snapdragon 8 Elite Gen 5 chipset while becoming slightly thinner and lighter at 4.1mm unfolded and around 210 grams. That would make it about 5 grams lighter than its predecessor despite the larger battery.

Camera upgrades are also reportedly on the way. Samsung is expected to retain the 200MP primary sensor and 10MP selfie camera while upgrading the ultra-wide camera to 50MP. There is currently no information about the telephoto sensor, though the Fold 7 used a 10MP setup.

A new “wide” foldable could also be coming

The report also references another foldable device currently referred to as the “Galaxy Z Fold 8 Wide.” Samsung has not officially acknowledged such a product yet, but the leak suggests it could feature a different approach to foldable design.

According to the report, the device may use a 4:3 aspect ratio inner display alongside dual 50MP rear cameras and a 4,800mAh battery. It is also said to weigh around 200 grams.

If Samsung does launch a wider foldable, it could signal the company experimenting with different form factors as competition in the foldable space intensifies.

Why this matters

Samsung remains one of the most important players in the foldable market, but the company has recently faced pressure from brands offering thinner designs, larger batteries, and faster charging.

A move to a 5,000mAh battery may sound basic compared to regular flagship phones, but for Samsung foldables, it would represent a long-overdue improvement. Combined with newer silicon-carbon battery technologies becoming more common across the industry, users are increasingly expecting foldables to stop compromising on endurance.

At the same time, Samsung still holds a major advantage in software support, ecosystem integration, and global availability – areas where many competitors continue to lag behind.

What happens next

As with most early leaks, these details should still be treated cautiously. TechManiacs has correctly leaked some Samsung hardware details in the past, but it has also missed on certain specifications, including the Galaxy Z Fold 7’s thickness.

Samsung is expected to unveil the Galaxy Z Fold 8 later this year alongside the next Galaxy Z Flip lineup. If these leaks prove accurate, the Fold 8 may end up being less about radical redesigns and more about finally fixing the practical compromises users have been complaining about for years.

Owning a pool often means enjoying backyard get-togethers, relaxing weekends, and the simple pleasure of a swim on a hot day. But in reality, it also comes with a fair amount of upkeep. In my experience, keeping a pool clean can quickly turn what should feel like downtime into another ongoing chore.

That’s where the shift in smart home technology becomes interesting. As it extends beyond living rooms and kitchens into outdoor spaces, a new generation of AI-powered systems is changing how pool care is managed — moving it from manual effort to real-time, automated operation. The iGarden Pool Cleaner M1-AI Series reflects this transition toward smarter outdoor living, combining adaptive cleaning and advanced vision into a low-intervention system designed around everyday convenience.

How the M1-AI Series Is Evolving Modern Pool Care

Earlier, robotic pool cleaners prioritized power over logic, often getting stuck in tangled cables and requiring manual intervention. The M1-AI Series overcomes these challenges by putting intelligence first.

Moving away from a purely power-driven approach, this robotic pool cleaner uses Bionic AI Dual-Camera technology to build a 3D understanding of its underwater environment. Through AI Target Mode, it can identify debris and obstacles in real time and adjust its cleaning path dynamically, rather than repeatedly covering the same areas. It is further supported by 3D “S” path planning, which adapts to different pool shapes and sizes for more complete and efficient coverage.

Another noticeable improvement is how focused and efficient the cleaning feels, with the system capable of removing up to 99% of pool floor debris in as little as 20 minutes. But beyond speed, the real win is the freedom. Instead of turning pool maintenance into a recurring task on your to-do list, the iGarden M1-AI Series is engineered to handle routine cleaning quietly in the background while homeowners spend more time enjoying their pool.

What Makes It Feel Truly Hands-Free

While the cleaning speed of the iGarden M1-AI is impressive, what makes it especially compelling is how little attention it demands once it is in the pool.

For many pool owners, the frustration isn’t just the physical cleaning, but the management required. The traditional robotic cleaners often required constant intervention, like adjusting timers, manually recharging units, or restarting cycles when debris was missed. Over time, what was supposed to make maintenance easier became another routine to worry about.

By pairing a long-lasting battery with intelligent wake-and-sleep cycles, this robotic pool cleaner operates in the background without the need for constant supervision. With up to 30 days of maintenance on a single charge, you can simply leave it in the water and let it handle the heavy lifting on its own schedule.

Such autonomy makes the iGarden M1-AI feel like a natural extension of the smart home devices many of us already rely on indoors. Much like a robot vacuum quietly handling floors or a smart thermostat managing everyday adjustments in the background, it takes care of the small but constant tasks so you do not have to. Whether it is preparing for a spontaneous weekend gathering or keeping the pool swim-ready, the hands-free convenience is genuinely practical.

But that capability extends to how the cleaner responds to dirt underwater. Through AI Adaptive Suction, the iGarden M1-AI automatically adjusts suction power based on the type of debris it encounters. It continuously evaluates what’s ahead, running on low power for fine dust and seamlessly switching to higher suction when it detects heavier leaves or grit.

Together, these adjustments deliver a more responsive cleaning performance, turning pool care from a repetitive chore into exactly what smart outdoor living should feel like — consistent, effortless, and fully automated while you focus on your poolside retreat.

Smart Energy Use for Everyday Pool Care

As smart home technology becomes more seamless, convenience alone no longer feels impressive. What stands out now is technology that saves time, reduces effort, and blends naturally into daily life. Pool maintenance, especially during peak summer months, can quickly start feeling like a second job.

The iGarden Pool Cleaner M1-AI Series is built to reduce that burden. Instead of operating at a fixed power level throughout the entire cleaning cycle, the system uses AI-Inverter 2.0 technology to balance performance and energy use in real time. The robotic cleaner stays powerful enough for demanding cleaning tasks while operating efficiently enough to reduce unnecessary energy consumption.

20-Minute Clean. 30-Day Hands-Free.

That promise sits at the center of the M1-AI Series experience, combining fast cleaning performance with low-maintenance operation designed to run more independently over time.

Supporting this is the OmniLogic AI platform, which acts as the cleaner’s decision-making system. Rather than following repetitive routes, the platform continuously analyzes pool conditions to determine more efficient cleaning paths. Powered by a 4K-grade ISP and a 6 TOPS NPU, it enables faster recognition, smarter navigation, and more adaptive cleaning decisions as conditions shift underwater. The result is smoother coverage and a cleaning operation that feels consistently responsive.

That same thoughtful engineering extends to the hardware. Driven by features like InfinityDrive for extended runtime and HyperBoost suction, this pool cleaner maintains a reliable performance through the messier parts of the season. Overall, ensuring that it can effortlessly handle everything from leaves after a windy afternoon to the inevitable clutter following a lively weekend by the water. 

Why the M1-AI Series is the New Pool Care Standard

Traditionally, the yardstick for a good pool cleaner was simply how hard it could scrub or how fast it could move. But the standard is starting to change. Today, the real luxury is not just having a clean pool, but not having to constantly think about the cleaning in the first place.

That is where the iGarden Pool Cleaner M1-AI Series earns its keep in modern households. Currently available through Amazon as well as its official site for U.S. buyers at just $999, it is a premium upgrade for homeowners looking to bring smarter automation into their outdoor spaces. By handling the stress of scheduling, navigation, and routine maintenance more independently, it transforms pool care into a quieter background process rather than something you plan your weekend around.

At its core, the real value lies in the freedom it gives back. If you’re someone looking to reclaim nearly a month of your summer from manual pool maintenance, the M1-AI Series delivers a level of self-sufficiency that makes a smart backyard feel genuinely smart and your time feel truly yours again.

Robot vacuums have reached a point where expectations are no longer limited to basic cleaning. Users now expect strong suction, reliable carpet performance, and minimal maintenance, but getting all of that typically means stepping into premium pricing. The Narwal Freo Z10 Turbo is built to challenge that trade-off by bringing flagship-level cleaning technologies into a more accessible category without compromising on real-world performance.

As Narwal’s first major mid-range release of 2026, the Freo Z10 Turbo is positioned to bridge the gap between affordability and high-end capability. Priced $599 after a $300 launch discount, it combines 25,000 Pa suction, CarpetFocus technology, and DualFlow Tangle-Free System into a single platform designed to handle mixed surfaces, pet hair, and everyday mess without requiring constant intervention.

Power that holds up across real-world cleaning scenarios

Suction power only matters if it delivers consistent results across different surfaces, and that is where the Freo Z10 Turbo stands out. With up to 25,000 Pa of suction, it sits among the highest-performing robot vacuums in its category, handling everything from fine dust on hard floors to embedded debris in carpets.

In everyday use, this reduces the need for repeat cleaning cycles. Fine particles are removed without scattering, while carpets benefit from enough lift to pull out dust and hair that would otherwise require multiple passes. This level of consistency allows the device to function as a primary cleaning solution rather than something that still depends on manual follow-up.

Deeper cleaning for carpets and hard-to-reach buildup

What looks clean on the surface often still holds onto fine dust, hair, and debris beneath it, particularly in softer materials where particles settle below the top layer and are not removed through a standard pass. Addressing that requires a more targeted approach to how cleaning is applied, especially on carpets where embedded buildup is harder to reach.

The Freo Z10 Turbo brings Narwal’s flagship CarpetFocus Technology into a more accessible segment, adapting automatically when carpet is detected. The system raises the mop to keep surfaces dry while lowering a brush cover to create a sealed high-pressure airflow zone, minimising air leakage and keeping suction concentrated where it is most effective.

This is reinforced by Carpet Max Mode, which applies a dual-pass zigzag pattern to work across the same area from multiple directions. By approaching the surface from both directions rather than relying on a single pass, the system improves how thoroughly embedded dust and hair are lifted from within carpet fibers. Taken together, this results in a level of cleaning that goes beyond surface-level pickup, allowing carpets to maintain a more consistent baseline over time without requiring repeated cleaning cycles.

A tangle-free system that holds up in everyday use

Hair buildup can quietly reduce performance over time, wrapping around brushes and lowering suction efficiency. The Freo Z10 Turbo is designed to avoid that cycle through its DualFlow Tangle-Free System.

By combining a dynamic side brush with a zero-tangling roller brush, the system guides hair directly into the dustbin instead of allowing it to collect around the mechanism. This keeps airflow consistent and reduces the need for manual cleaning.

In homes with pets or long hair, this makes a noticeable difference. The system continues to perform consistently over repeated cycles, making it better suited for daily use rather than occasional cleaning.

Coverage that extends beyond open floor areas

Even high-performing robot vacuums tend to miss edges and corners, where dust gradually builds up. The Freo Z10 Turbo addresses this with its EdgeReach mop system, which extends outward to clean along baseboards and into tighter spaces.

Combined with consistent 12N downward pressure that mimics manual scrubbing, the system is able to handle more stubborn residue rather than simply passing over it. Over time, this reduces the need for manual touch-ups in high-use areas like kitchens, hallways, and entryways.

Automation that keeps cleaning off your to-do list

Automation only matters when it removes the need to step in, and the Freo Z10 Turbo is built to operate with minimal input. Its navigation system combines tri-laser structured light with LDS mapping to move through the home with precision, detecting and avoiding obstacles in real time without relying on cameras.

The all-in-one base station handles dust collection, AI adaptive hot water mop washing, and drying, enabling up to 120 days of maintenance-free operation under typical use.

In multi-room homes, this shifts the role of the device from something you manage occasionally to something that runs consistently in the background. That focus on reducing ongoing effort reflects Narwal’s approach to building systems that prioritise reliability over time.

A best-in-class option, now at a more compelling entry point

The Narwal Freo Z10 Turbo is priced at $899.99, but is currently available at a launch price of $599.99, a $300 reduction that brings it into a far more accessible range for what it delivers. To top it off, you’ll also secure a complimentary $199 accessory bundle as part of this limited-time launch promotion.

At this price, the equation shifts noticeably. Features like 25,000 Pa suction, CarpetFocus technology, a DualFlow Tangle-Free System, and full-coverage mopping are typically spread across higher-end models, not brought together this tightly at a mid-range price point.

What stands out here is how complete the overall package feels. From handling carpets and edges to managing pet hair and reducing ongoing maintenance, the Freo Z10 Turbo delivers consistent performance across the areas that tend to matter most in daily use.

With the launch offer running from May 18 to May 31, this becomes a limited window where that balance of performance and price is at its strongest. For users who have been considering an upgrade, this is one of those moments where waiting is unlikely to offer the same combination of capability and value at this level.

You can learn more about the Freo Z10 Turbo and take advantage of the limited-time launch offer here.

AI can now answer questions so quickly that the search itself can feel optional. That convenience worries the Royal Observatory Greenwich, which has warned that instant AI answers can weaken the curiosity, scrutiny, and source-checking behind real knowledge.

The risk hides inside the usefulness. Chatbots can help people test ideas, move faster, and find new angles, but a finished response can also cut users off from the messy trail that makes learning stick. When that happens, information arrives without the struggle that turns it into judgment.

How much thinking should AI do for us

The Royal Observatory’s argument carries weight because it comes from an institution built on patient observation, not quick summaries. Paddy Rodgers, director of Royal Museums Greenwich, points to the habits that scientific discovery depends on, asking better questions, weighing evidence, and following leads that don’t look useful at first.

Astronomy’s own history backs him up. Early observers gathered vast records about the heavens, and later generations found uses for that data the original researchers couldn’t have predicted. A machine optimized for efficiency might have skipped those detours because they lacked immediate value.

What happens when intelligence becomes a utility

Sam Altman has described AI moving toward a metered service, with intelligence sold more like electricity or water and priced through usage. His framing is a business model, but it sharpens the cultural worry around AI as a replacement for mental effort.

If intelligence becomes something people buy on demand, reasoning can start to feel like a service call rather than a skill to practice. The danger grows when a polished answer gets treated as verified knowledge, especially when users can’t see what the system skipped, flattened, or failed to check.

What should people watch next

The better habit is to make AI work against your own certainty. Ask it to challenge an idea, expose missing evidence, and test a conclusion before you accept the response as finished.

That turns the Royal Observatory’s warning into a practical rule. Use AI to widen the search, not end it. Check what it leaves out, trace claims back to sources, and keep the final act of judgment in human hands.

Scientists have pushed wireless speed into territory that current mobile networks can’t touch. A Tokushima University team demonstrated a 112Gbps wireless connection in the 560GHz band, using soliton microcombs to generate a more stable terahertz signal for future 6G systems.

The near-term prize isn’t a faster handset. It’s the hidden infrastructure that carries traffic between network sites, where backhaul capacity can decide whether future 6G speeds feel real or get trapped behind crowded network pipes. That makes this a useful 6G speed breakthrough to watch, even if consumers won’t see it on a spec sheet anytime soon.

Why does this record carry weight

The 560GHz band gives the 112Gbps result its edge. The team sent a single-channel wireless signal well beyond the range where conventional electronic hardware starts running into weaker output power and higher signal noise.

That frequency range sits in the terahertz zone, which researchers are exploring as a way to open wider data lanes for 6G. Earlier communication systems at these frequencies have often stayed in the range of a few to several dozen gigabits per second. This test crossed the 100Gbps class beyond 420GHz, which pushes the work into a more serious category.

How did the signal stay clean

At these frequencies, raw speed depends on control as much as bandwidth. Phase noise and limited output power make wireless transmission harder to keep stable, especially when a system is trying to move more data through one channel without the signal falling apart.

When do real networks get closer

No one should read this as a phone upgrade arriving soon. The researchers still need to cut phase noise further, support higher-order modulation, improve terahertz output power, and extend transmission distance with better antenna design.

The first useful home for the technology will probably be mobile backhaul or photonic-wireless network links. That’s less visible than a new 6G phone, but it’s more important to the network itself. Before 6G can deliver massive speeds to everyday devices, the infrastructure behind those devices needs a faster way to move data around.

Google has always been ahead of the curve when it comes to protecting Pixel users from spam calls, and it looks like the company isn’t done yet. According to a recent teardown of the Google Phone app by Android Authority, Google is working on a new phone number spoofing detection feature.

What is phone number spoofing?

Phone number spoofing, also known as caller ID spoofing, is when a scammer tricks your phone into displaying a familiar or saved contact’s number, even though the call is actually coming from a completely different number. 

As users are more likely to pick up a call if it looks like it’s coming from family members, friends, or authorized personnel, like a doctor or a bank representative, phone number spoofing is on the rise in the scam chart. It has become a surprisingly common tactic and one that has caught a lot of people off guard.

So what is Google doing about it?

Android Authority cracked open version 222.0.913376317 of the Google Phone app and found strings of code that point to an upcoming spoofing detection system. One of the strings reads, “Someone may be pretending to call from your contact’s number,” and another suggests that users will have the option to hang up the call immediately.

It’s not entirely clear how Google plans to detect spoofed numbers, but the timing is interesting. Only a few days back, Google announced a slew of security features, including verified financial calls, OTP protection, real-time malware detection, APK scanning in Chrome, and more.

With the new call spoof detection feature and existing spam call protections, including Call Screening and spam detection, the Pixel phones have become the best anti-scam smartphones. There’s no word yet on when this feature will roll out, but it’s good to know Google is working on it.

Gemini and ChatGPT dominate the AI image generator conversation, but Ideogram has a cleaner argument for attention. It focuses on visual work people need to ship, from creator assets to layouts that have to fit a platform on the first try.

Its clearest advantage is typography. Ideogram is built for posters, banners, social posts, newsletter illustrations, and video thumbnails, with a particular strength in generating readable copy inside designs. One garbled word can wreck an otherwise usable graphic.

It also gives users practical controls, including prompt refinement, four image options per request, public galleries for inspiration, style choices, dimensions, remixing, and paid editing through Canvas. Those features help turn a rough request into something closer to publishable.

Why Ideogram keeps pulling users back

Ideogram puts text placement and format choices into the workflow from the start. For creators working on layout-heavy assets, that can cut down the repair loop that usually follows a flawed AI image result.

The service gives users four generated options each time, which adds a useful layer of selection before editing begins. Its automatic prompt refinement can expand a rough idea, while public galleries make it easier to study existing images and build from other starting points.

Why bigger tools don’t settle it

Gemini and ChatGPT still have strengths Ideogram doesn’t erase. Gemini’s Nano Banana Pro is positioned as versatile across logos, infographics, slide designs, portraits, and abstract visuals, while ChatGPT is strong for diagrams, and image edits guided through conversation.

Ideogram wins a more specific fight. It fits jobs where creator assets often fail over small details, especially copy in the design, reusable styles, flexible aspect ratios, and fast revision. For public-facing graphics, those details can outweigh brand familiarity.

Where Ideogram still makes you wait

Ideogram isn’t a clean win for everyone. The free plan includes restricted daily generations, slower rendering, public image creation, and lower-quality JPEG downloads. Paid plans add more images, faster output, extra dimensions, negative prompts, and Canvas editing.

The smartest approach is to treat Ideogram as a specialist. Flux, Adobe Firefly, Gemini, and ChatGPT all have their own strengths, but Ideogram deserves a test run when the job depends on readable design copy and repeatable formats.

Start with the free version, but don’t judge it from one request. Its value shows up after a few iterations, style changes, and format tests.

Valve’s new Steam Controller has had a pretty good start. Early reactions have been positive, and the $99 controller sold out quickly after launch.

That demand also brought scalpers, who started listing the controller at inflated prices. Valve has since introduced a reservation queue to give real buyers a better shot at future stock. Still, one complaint kept coming up. For many players, the Steam Controller was simply too locked into Steam.

What was holding the Steam Controller back?

For players who mostly game through Steam, the setup works well. Steam Input handles the controller’s extra features and gives users plenty of control over how it behaves. Still, many players do not keep all their games inside Steam. For those users, the controller was harder to recommend because it did not work as smoothly across other launchers and non-Steam games.

That is now starting to change. As spotted by Phoronix, support for the new Steam Controller has been added to SDL (Simple DirectMedia Layer), the widely used cross-platform library that many games and apps rely on for controller input. It has also received a follow-up mapping update, which should help the controller behave more like a standard third-party gamepad in SDL-supported games.

How well does it work outside Steam now?

Early testing sounds promising, although it is not perfect yet. Testers in the SDL pull request said that the controller works with or without Steam running, and that touchpads, capacitive stick touch, grip sense, back buttons, gyro, accelerometer, and the QAM button are functional in some form.

That said, there are still minor touchpad issues, and running Steam in the background can cause double-input problems in some cases.

For now, it appears that the Steam Controller will have to rely on SDL to play third-party games. Valve developer Pierre-Loup has already clarified that adding standard Windows XInput support would essentially make it behave like an Xbox controller, which could limit its unique inputs, require a separate mode-switching setup, and add extra cost for users.

Apple is reportedly preparing one of the biggest Siri redesigns in years with iOS 27, but even after multiple delays, the company may still label the upgraded assistant as a beta product. According to reports from Mark Gurman of Bloomberg, internal test versions of iOS 27 already refer to the revamped Siri as a beta experience and include an option allowing users to leave the Siri beta entirely.

The move would be unusually familiar for longtime Apple users. When Apple originally introduced Siri in 2011, the assistant itself launched under a beta label before Apple quietly removed the branding in 2013. Despite that, Siri has continued to face criticism for lagging behind competitors in reliability, conversational abilities, and overall intelligence.

Apple’s AI catch-up strategy is taking longer than expected

The revamped Siri was originally expected to arrive in 2024 as part of Apple’s broader AI push. However, multiple reports now suggest the project has faced delays of nearly two years.

The issue for Apple is timing. While Apple continues refining Siri, rivals like Google Gemini, ChatGPT, and other Android-based AI systems have already rolled out advanced conversational assistants with broader real-world capabilities.

That gap has increasingly made Siri feel outdated compared to competing AI products, especially as Apple continues marketing Apple Intelligence as a major part of the iPhone experience.

Why the beta label matters

If Apple officially launches the new Siri as a beta feature in iOS 27, it could serve two purposes. First, it gives Apple flexibility to continue refining the assistant publicly after launch while lowering expectations around bugs, hallucinations, or missing features. Second, it allows the company to release AI features sooner rather than waiting for a more polished final version.

The beta branding would also reflect the broader challenge Apple currently faces in AI. Unlike competitors that prioritize rapid deployment, Apple has historically focused more heavily on stability, privacy, and controlled rollouts.

Reports also suggest Apple is introducing stronger privacy controls into Siri’s AI experience, including optional auto-delete settings for conversation history.

What happens next

Apple is expected to reveal more about Siri’s redesign and its AI roadmap during WWDC next month. Developer beta versions of iOS 27 will likely be the first public look at the new Siri experience.

However, the larger question remains whether Apple’s slower, more cautious AI rollout can still compete in a market where rivals have spent the last two years aggressively pushing generative AI into mainstream consumer products.

For now, Siri’s overhaul appears less like a finished comeback and more like Apple finally arriving at the AI race – still mid-development.