Connect with us
DAPA Banner
DAPA Coin
DAPA
COIN PAYMENT ASSET
PRIVACY · BLOCKDAG · HOMOMORPHIC ENCRYPTION · RUST
ElGamal Encrypted MINE DAPA
🚫 GENESIS SOLD OUT
DAPAPAY COMING

Tech

Anthropic Says a Mythos-Class AI Model Will Be Available Soon

Published

on

Anthropic’s most capable generative AI model is still in the hands of a select few organizations and cybersecurity professionals, but the most powerful Claude model you can use now is getting an upgrade.

Claude Opus 4.8, released Thursday, is a “modest but tangible improvement” over Opus 4.7, Anthropic said in a blog post

But the company also said it’s making significant progress on producing a version of its Claude Mythos Preview model that it’s willing to release to the public. Right now, it has restricted access to Mythos to a consortium of partners as part of what it calls Project Glasswing, explaining that the model’s cybersecurity capabilities are advanced enough to warrant giving cybersecurity experts and major tech companies some lead time to patch flaws found by the model.

Advertisement

“Models of this capability level require stronger cyber safeguards before they can be generally released,” Anthropic said. “We’re making swift progress on developing these safeguards and expect to be able to bring Mythos-class models to all our customers in the coming weeks.”

Mythos for everyone?

Anthropic’s decision to withhold Mythos Preview from the general public, at least for now, has been an interesting one. Was it a wise, forward-thinking move to protect the internet’s critical infrastructure from potential flaws? Was it an easy way to churn up marketing hype? Security researchers have found that the model is certainly capable of finding exploits much more quickly than human hackers, even if it isn’t necessarily pushing beyond human capabilities. Mozilla’s latest version of Firefox included more than 200 fixes identified by Mythos Preview.

Read more: AI Arms Race Accelerates With New Models from OpenAI, DeepSeek and Anthropic

But the fact that Mythos will soon be available to anyone, even with significant cybersecurity guardrails, means we’ll finally get to see if the model lives up to the hype, with all the risks that might entail.

Advertisement
AI Atlas

Darren Williams, founder and CEO of the cybersecurity firm BlackFog, told CNET in an email that big model releases are often tense moments.

“On one hand, Anthropic’s decision to stage the release, holding back until safeguards are developed, shows the right instincts,” he said. “But the more capable a model is, the higher the stakes if those safeguards fall short or if the model is eventually misused. The window between a powerful model’s release and broad adoption of defenses against it is always a vulnerable moment.”

Mythos is going to be far more expensive to run than other AI models, however, and that could limit its usefulness to hackers. Jake Williams, a cybersecurity researcher and faculty member at IANS Research, said Mythos was 30 times as expensive in tests as the previous Opus model.

“This is outside the reach of many, including any commodity threat actors,” Williams told CNET in an email. “Nation-state actors already had better technology for finding vulnerabilities. This is only changing the game for a small percentage of threat actors.”

Advertisement

What’s new about Claude Opus 4.8

As for Opus 4.8, Anthropic said it’s an improvement across benchmarks compared to Opus 4.7. Tests found that Opus 4.8 was less likely to make unsupported claims and more likely to indicate uncertainty, the company said.

A few new features are also coming to Anthropic’s AI products, including the ability to control how much “effort” a model will use to respond to a prompt on Claude.ai and in Claude Cowork. Higher effort will likely get better results as the model spends more time on a response, but it will burn through your usage limits faster. A lower setting will respond faster and hit rate limits more slowly.

Source link

Advertisement
Continue Reading
Click to comment

You must be logged in to post a comment Login

Leave a Reply

Tech

BYD unveils China’s first 4nm driving chip and expands God’s Eye

Published

on

TL;DR

BYD has unveiled the Xuanji A3, China’s first automotive-grade 4nm chip for self-driving cars, delivering 700 TOPS per chip. The company is expanding its God’s Eye driver-assistance system to mass-market EVs including the $10,300 Seagull, as eight consecutive months of falling sales and a 55% profit decline force a technology-led pivot.

BYD has unveiled the Xuanji A3, which it calls China’s first automotive-grade 4-nanometre chip for self-driving vehicles. CEO Wang Chuanfu announced the chip at an event at BYD’s Shenzhen headquarters on 28 May, saying it delivers the lowest power consumption per unit of compute in its class, drawing roughly 20% less than comparable semiconductors. The chip has already entered mass production.

Advertisement

A single Xuanji A3 delivers 700 TOPS of computing power. A cluster of three chips reaches 2,100 TOPS, enough to support Level 3 and Level 4 autonomous driving functions. The chip is the centrepiece of a new laptop-sized central computing platform that unifies three previously separate vehicle domains: the smart cockpit, the driver-assistance system, and the core electric propulsion.

The semiconductor approaches the capabilities of Huawei Technologies, which currently makes automotive chips at a 7nm geometry but has pledged to debut 1.4nm chips by 2031. The most advanced chip globally is TSMC’s 2nm N2 node. BYD’s ability to design and mass-produce its own 4nm driving chip deepens a vertical integration strategy that already spans batteries, motors, and vehicle manufacturing.

God’s Eye goes mass market

Alongside the chip, BYD announced it will expand its God’s Eye driver-assistance system across all models sold in China, including mass-market vehicles like the Seagull compact hatchback, which starts at 69,800 yuan ($10,300). The Seagull became the first car in its class to receive LiDAR when the 2026 model launched on 11 May, a technology previously reserved for premium vehicles.

The upgraded God’s Eye system, which includes city and highway navigation, traffic light recognition, and automated parking, will be available as an add-on for 12,000 yuan ($1,770). Wang described the pricing as cost-only. BYD is also offering one year of insurance that fully covers damages from accidents that occur while the latest version of God’s Eye is engaged, a direct bet that the system is reliable enough to underwrite.

Advertisement

A system with a troubled track record

The insurance pledge comes after months of complaints about God’s Eye’s performance. BYD made the system a standard feature on most of its vehicles last year, but the initial rollout used a tiered structure that limited advanced urban navigation to more expensive models while giving affordable cars only basic highway cruise control. Users reported dangerous malfunctions including unintended acceleration, erratic lane changes, and steering inputs that nearly sent vehicles into oncoming traffic.

Unlike Tesla’s optional Full Self-Driving package, BYD deployed God’s Eye across millions of vehicles, meaning every flaw appeared at scale. The company says it has more than 3.15 million vehicles equipped with advanced driver-assistance hardware on the road, generating roughly 200 million kilometres of driving data every day. BYD is using that data to train its algorithms through cloud-based world models and reinforcement learning, with iteration cycles running every three days.

Eight months of falling sales

The technology push arrives at a difficult moment for BYD’s business. Sales have declined year on year for eight consecutive months, with April 2026 volumes down roughly 16% compared with the same period a year earlier. First-quarter net profit fell 55% to 4.09 billion yuan ($599 million) as a fierce price war in China and a stronger yuan compressed margins.

Revenue dropped 12% to 150.2 billion yuan in the quarter. Operating expenses rose as BYD invested in smart driving hardware while simultaneously cutting vehicle prices to compete with rivals including Xiaomi, which has entered the EV market with high-performance models, and Huawei-backed brands that have intensified their presence. The only bright spot has been exports, which rose more than 70% year on year to a record 135,098 units in April.

Advertisement

The competitive context

Tesla finally launched its Full Self-Driving system in China after years of delays, but the technology still requires active human intervention and will be marketed under a different name due to scrutiny from Chinese transportation authorities. Tesla relies on a vision-only approach using cameras and neural networks without LiDAR or radar, a fundamentally different and cheaper strategy that BYD and other Chinese automakers have criticised as less capable.

China’s Ministry of Industry and Information Technology issued its first Level 3 autonomous driving certifications in December 2025, approving cars from Changan Auto and BAIC Motor. BYD is waiting for China to formalise legislation allowing broader consumer-facing deployment of self-driving vehicles, which the company expects as soon as 2027. Chinese EV makers are also expanding aggressively overseas, with BYD targeting 1.3 to 1.6 million international deliveries in 2026.

The Xuanji A3 chip and the God’s Eye expansion represent BYD’s attempt to shift the competitive battleground from price to technology. A company that made its name on affordable electric vehicles is now trying to prove it can build the silicon, software, and sensor systems needed to compete on intelligence. Whether the strategy works depends on whether the technology can outrun the complaints, and whether drivers in a market saturated with discounted EVs are willing to pay 12,000 yuan for a feature that Huawei-backed rivals and a fragmenting global EV market are also racing to deliver.

Advertisement

Source link

Continue Reading

Tech

Samsung HW-QS90H Review – Trusted Reviews

Published

on

Verdict

Samsung’s innovative new ‘convertible’ soundbar somehow manages to sound like it’s using an external subwoofer even when it isn’t, while still delivering all the power, detail and sound stage craft we’ve come to expect from Samsung soundbars. Regardless of whether you’re using it on a wall or tabletop

  • Outstanding sound quality in both of its convertible set up configurations

  • Excellent with music as well as movies

  • Great value for what it offers

  • Headline multi-placement feature is potentially niche

  • Occasionally over-bright detailing

  • No rear channel sound without adding extra speakers

Key Features

  • Trusted Reviews IconTrusted Reviews Icon

    Advertisement

    Review Price:
    £899

  • Perfect for walls or table tops

    Advertisement

    The QS90H can reconfigure which sound channels emerge from each of its speakers depending on whether the soundbar is wall mounted or sat on a stand under your TV

  • Great for music and movies

    Advertisement

    Despite creating a brilliantly detailed, aggressive and powerful multi-channel sound scape for movie playback, the HW-QS90H can also adapt itself unusually well to stereo music

  • Wi-Fi streaming support

    Advertisement

    The QS90H is compatible with Google Cast, AirPlay 2, Spotify Connect, and Tidal Connect, as well as being Roon Ready

Introduction

While most soundbars technically support wall mounting, even sometimes including wall brackets in their boxes, that support is typically compromised in both design and sound quality terms.

Soundbars designed to lie flat on a table, for instance, like the vast majority are, usually jut out awkwardly from walls, while their sound can be affected by how high they’re sat when wall hung versus a typical desktop placement.

Advertisement

Cue Samsung’s HW-QS90H: A soundbar that can actually reconfigure its speakers deliver which channels of sound depending on whether its sitting flat on a table or hanging flat against a wall.

Advertisement

This isn’t Samsung’s first stab at such a ‘convertible’ soundbar; that honour belongs to the 2025-launched HW-QS700F. The QS90H is Samsung’s first convertible soundbar to offer a 7.1.2 channel count – while simultaneously daring to ship without the external subwoofer the QS700F provided.

Can a one-bar sound solution really deliver so many unusual and premium features without its sound quality suffering?

Advertisement

Design

  • Convertible design speaker system
  • Enjoyably industrial hard plastic finish
  • 13 driver 10-channel design

The HW-QS90H closely tracks Samsung’s other recent premium soundbar designs. So there’s the same hard black plastic finish, the same engaging mix of a striped effect on its top edge and circular perforation effect on its sides, and the same crisp, rectangular shape – only without the double-angled left and right ends you get with Samsung’s flagship soundbar designs.

The QS90H’s measurements of 1245(w) x 68.8(h) x 125(d)mm are fairly substantial by most one-bar solution standards. The fact that it rests on its flat bottom edge regardless of whether it’s lying down or hanging on a wall, though, gives it a relatively minimal profile.

Advertisement

Samsung HW-Q950H designSamsung HW-Q950H design
Image Credit (Trusted Reviews)

You don’t have to accommodate an external subwoofer, either, and actually, once you start to listen to it its size suddenly starts to feel pretty small in the context of the epic soundstage it can create.

Looking at the QS90H in its table-top configuration, two speakers set at different angles in its left and right edges enable the soundbar to deliver separate side left/right and front side left/right channels, while the long straight front edge holds the more typical left and right channels, plus a centre channel speaker that unusually sits right at the top of the front edge.

There are also two drivers in the top edge (again, remember, we’re talking about the tabletop profile here) for delivering height/overhead effects.

Advertisement

To compensate for not having an external subwoofer, the QS90H also somehow finds space for four bass woofers within its compact design.

Samsung HW-Q950H wall mount positionSamsung HW-Q950H wall mount position
Image Credit (Trusted Reviews)

Advertisement

It’s all change if you want to wall hang the soundbar. To adapt it for wall hanging you have to stand the QS90H on its edge and then turn it through 180 degrees so that the control buttons are sitting along what is now the soundbar’s bottom edge. The flat edge that was the underside of the soundbar in its table top stance becomes the rear edge that lies against the wall, meaning the soundbar only sticks out 68.8mm rather than the 125mm if it was fastened to a wall in the same orientation it uses on a desktop.

The speakers that were the front left and right speakers become the two up-firing speakers, while the two speakers that were the up-firers in the QS90H’s desktop setup become the front left and right. The centre channel speaker’s unusual high position in a tabletop set up now makes sense, as it means it can continue as the centre speaker in the vertical orientation too.

The ability of the QS90H to deliver a slender profile in both its setup positions puts it in very rarified company indeed. There are a few dedicated wall hanging soundbars out there, but soundbars that can truly adopt their sound and design for both flat or vertical orientations are incredibly few and far between.

Advertisement
Samsung HW-Q950H onboard controlsSamsung HW-Q950H onboard controls
Image Credit (Trusted Reviews)

The on-bar controls I mentioned earlier offer voice control mic on/off buttons, as well as simple up and down volume adjustments. The only other disruption to the sleek black finish comes from a little row of LEDs designed to help you figure out roughly what volume you’ve selected and which input you’re using. These are a poor second, though, to the full LED display you get on Samsung’s HW-Q990H.

The HW-QS90H’s connections comprise a one in, one out HDMI pass-through, eARC support on the HDMI output, an optical digital audio input, and the now inevitable Bluetooth and Wi-Fi support.

Advertisement

Features

  • Dolby Atmos and DTS:X playback
  • 7.1.2 channel count, plus optional extra speaker support

I’ve pretty comprehensively covered the QS90H’s most unique feature: Its support for flat-profile desktop or wall-hanging placement. There’s plenty more to get our teeth into besides.

Starting with the fact that it uses its 13 built-in drivers to deliver an impressive 7.1.2 channels of sound. This is an impressive channel count for a single bar soundbar, delivering front left, front right, front centre, front left side, front right side, left side, right side and two up-firing channels, with a quartet of woofers squeezed in to deliver low frequency sounds. Which drivers deliver which channels is, of course, fluid, based on the soundbar’s orientation.

Samsung HW-Q950H rear sideSamsung HW-Q950H rear side
Image Credit (Trusted Reviews)

It’s worth noting at this point that a built-in gyroscope automatically informs the HW-QS90H whether it’s been used in its horizontal or vertical configuration; you don’t need to tell it manually.

Inevitably as a single-bar solution, the QS90H doesn’t deliver any surround sound effects. You can add extra speakers to it, though, using either Samsung’s SWA-9500S optional wireless rear speaker package, or up to four of Samsung’s startlingly powerful new Music Studio speakers.

Advertisement

Advertisement

As with all recent Samsung soundbars, the QS90H can also join forces with (rather than just replacing) the speakers in pretty much all of Samsung’s recent TVs thanks to Samsung’s Q Symphony feature, creating a bigger and more detailed centre channel sound.

Q Symphony can work via HDMI eARC or wirelessly, with further wireless functionality coming in the shape of wireless Dolby Atmos reception from suitably capable Samsung TVs.

Samsung HW-Q950H connectionsSamsung HW-Q950H connections
Image Credit (Trusted Reviews)

The Dolby Atmos support is joined by support for both the DTS:X format and the Eclipsa Audio open-source format developed by Samsung and Google.

You can get any sound format, even stereo music, to take advantage of the HW-QS90H’s full channel count if you choose a provided Surround Sound audio preset. Purists can rest assured that the soundbar’s Standard preset will play any format in its native form – including simple two-channel stereo.

Advertisement

The HW-QS90H also leans in to Samsung’s love of AI with an Adaptive mode, which analyses the incoming sound so that it can recognise the type of content and optimise the way the soundbar presents the audio accordingly. AI is in play with a dialogue enhancement feature that can isolate vocal tracks and give them more emphasis in response to detected increases in ambient noise levels in your room.

Advertisement

The QS90H’s Wi-Fi capabilities include integrated support for AirPlay, Google Cast, Spotify Connect and Tidal Connect, and it’s also Roon Ready. Accessing all these sound systems is made easy if you’re using the Samsung SmartThings or new Samsung Sound apps to control the soundbar rather than the sleek but basic provided remote control. The same apps also streamline initial set up, helping you get your soundbar online and update its software with minimal fuss.

Samsung HW-Q950H remote controlSamsung HW-Q950H remote control
Image Credit (Trusted Reviews)

The apps provides exclusive access to a couple of handy set up aids: An adaptive bass feature that automatically monitors the sound for potential bass-related distortions, correcting the sound profile to correct for this if required, and a Space Fit system that continually optimises the soundbar’s presentation to your room conditions.

The QS90H’s HDMI pass-through supports HDR10+ and Dolby Vision HDR formats, as well as the more basic HDR10 and HLG formats. The HDMI pass-through system does also deliver arguably the only real feature disappointment with the QS90H, though, as it turns out that it doesn’t join Samsung’s HW-Q990H flagship soundbar’s loop through in supporting 4K/120Hz signals.

Advertisement

You can only get 120Hz frame rates through the soundbar if you settle for a 1440p resolution and don’t mind losing high dynamic range support. If you want 4K and HDR, you’ll have to set your game source to 4K/60Hz. The soundbar does support pass-through of variable refresh rates with both 60 and 120Hz signals, and carries a Game Pro sound setting that emphasises channel steering to make gaming worlds feel more intense and help you detect where enemies might be attacking from.

Sound Quality

  • Equally excellent music and movie playback
  • Outstanding sound in both horizontal and vertical configurations
  • So good that it will tempt you to add rear speakers

Advertisement

Before getting into the specific ins and outs of how well the HW-QS90H delivers on its headline multi-placement ‘convertible’ feature, let’s get some general performance features out of the way.

Starting with the fact that it copes incredibly well without the external subwoofer shipped with 2025’s QS700F convertible soundbar debutante. The bass woofers crammed into the QS90H’s single-bar form sound anything but crammed in, rolling out some of the very deepest frequencies I’ve ever heard from an all-in-one soundbar option without the bass bottoming out, lagging behind the rest of the mix or succumbing to chuffing/crackling distortions or drop outs.

Unless you increase the soundbar’s volumes to levels far beyond anything the average human ear will be able to cope with.

Advertisement
Samsung HW-Q950H top downSamsung HW-Q950H top down
Image Credit (Trusted Reviews)

The QS90H’s bass spreads far and wide so smoothly, that it delivers exactly the sort of non-directional presence you want from any good movie bass system. It does all this without the bass overwhelming the rest of a movie mix.

In fact, the QS90H maintains levels of detail, power, effects placement and channel steering that precious few other single-bar soundbars can even get close to – exactly as I’ve come to expect with Samsung’s premium soundbars. The bass is just a beautifully weighted but also surprisingly nimble counterpoint to everything else that’s going on rather than any sort of ‘dead weight’ dragging the sound down.

Advertisement

Trebles never sound harsh or warbly either, and the speakers are sensitive enough to not only pick every sound detail out, no matter how subtle, but also present each of those sound effects with the correct level of prominence. There are no moments where subtle background ambient effects surge into the foreground, or birds tweeting suddenly sound on a par with key dialogue.

Dialogue, now I’ve mentioned it, is always clear and clean, regardless of how deep or shrill the talker’s voice might be, and the soundbar gives vocals a nice vertical lift so that they sound like they’re coming from the onscreen action rather than the soundbar below the screen.

Advertisement

There’s even an ‘Elevate’ option that does a good job of lifting the vertical position of dialogue higher should you be using the soundbar with a truly huge screen, or you’ve got a bigger vertical gap than usual between your screen and soundbar below. Which can happen with wall hung setups, or if you’ve got the soundbar sat on a sideboard with a wall mounted screen above it.

Samsung HW-Q950H wall mount frontSamsung HW-Q950H wall mount front
Image Credit (Trusted Reviews)

The channel steering that’s been such a consistent Samsung soundbar strength for years now is exceptionally well deliver by the HW-QS90H, creating a truly three-dimensional and immersive sound space between you and your screen. You truly feel the impact of the 7.1.2 channel count in the dynamism, fullness and expansive scale of the sound the QS90H creates.

So good is the QS90H’s staging, in fact, that it had me yearning for some surround sound speakers, so that I could find myself right at the heart of the audio action. The idea that a one-bar soundbar might be good enough to seriously tempt you to splash out on further speakers to create a true surround sound system is hardly a weakness!

Advertisement

Though there could be an argument here for stepping up to Samsung’s full surround sound HW-Q990H soundbar package instead.

Advertisement

All of the strengths explained so far are underpinned by phenomenal amounts of power by mid-range soundbar standards, and are achieved regardless of whether the QS90H is used in a vertical or horizontal stance.

Some differences in the sound do become apparent as you switch the HW-QS90H from a horizontal to a wall mounted position. On the downside, the wall-mounted sound doesn’t push forward with quite as much force. This means the wall-mounted sound stage feels a bit less three-dimensional, and hard impact sounds – while still clean and crisp – lack a little of the visceral potency you get in the QS90H’s horizontal stance.

Bass is a little less potent when the QS90H is wall mounted, too – though it’s still plenty powerful enough to leave the bass from many rival soundbars sounding either coarse or thin.

Height effects are delivered with actually slightly more clarity and emphasis in the QS90H’s wall mount mode than you get from its tabletop set up, though.

Advertisement
Samsung HW-Q950H onboard controlsSamsung HW-Q950H onboard controls
Image Credit (Trusted Reviews)

Advertisement

Dialogue sounds slightly more rounded and smooth from a wall-mounted QS90H, too, and the sound impressively radiates down as well as up from the soundbar’s relatively elevated wall-mount position, avoiding that ‘the only way is up’ effect you often get if you wall mount regular soundbars at higher positions than they’re truly designed to suit.

The HW-QS90H isn’t just a beast of a soundbar with film soundtracks. In both of its placement configurations it also handles stereo music remarkably well, subtly shifting to nimbler, less bombastic bass handling and beautifully recalibrating its multi-channel scale down to simple two-channel stereo without the results sounding crowded or over-aggressive.

Stereo separation is bold but not forensic or forced, and vocals of every type are beautifully located at the heart of the mix without ever becoming either too shrill or too soupy.

There’s a musicality here, in fact, that you’d normally only expect to hear from soundbars from established heritage hi-fi brands.

Advertisement

There are a couple of issues to mention beyond the small differences between the vertical and horizontal profiles covered earlier. One is that just occasionally, especially in the tabletop stance, voices can sound a little too prominent and bright. Such moments truly are rare, and even at their worst voices don’t actually sound brittle or completely lose their context.

Perhaps related to this, while ambient effects never gain too much weight in the mix, occasionally relatively high-pitched background sounds can – again only very occasionally – sound a touch sibilant.

Advertisement

Finally, again incredibly rarely, particularly violent and deep extended bass sounds can cause the QS90H’s bass speakers to start sounding just a little rough around the edges. Though they don’t succumb to outright distortion and breakdown.

Advertisement

Should you buy it?

You want a soundbar you can use anywhere

Thanks to its compact single-bar design and rare ability to reconfigure its speakers to different channels depending on whether it’s resting against a wall or sitting on a desktop, the QS90H can work in basically any and every room environment.

Advertisement

You want a full surround experience

While you can add speakers to the QS90H, in its default form it only creates a sense of three-dimensional sound in front of you, rather than all around you.

Advertisement

Final Thoughts

The HW-QS90H is a fantastic soundbar for its money. Far from being a gimmick, its convertible design works amazingly well, enabling it to work in pretty much any room setup.
 
Its sound is as powerful as anything I’ve ever heard from a single-bar soundbar solution, and it delivers a dynamic range that leaves most rivals sounding thin and weedy by comparison. Especially in the bass department.
 
It’s able to create a huge, detailed and engaging sound stage for movies, but also manages to rein itself in with stereo music, sounding truly musical without losing the power that’s now a Samsung soundbar trademark.
 
The number of individual households who specifically take advantage of the QS90H’s convertible design might be small, I guess. Surely the key point, though, is that the QS90H’s flexibility means that it’s an outstanding option for absolutely anyone in the market for a one-bar soundbar, regardless of where their TV might be positioned. And really powerful true wall-mounted soundbars, in particular, are as rare as hen’s teeth.

How We Test

We test every soundbar we review thoroughly over an extended period of time. We use industry-standard tests to compare features properly.

Advertisement

We’ll always tell you what we find. We never, ever accept money to review a product.

Find out more about how we test in our ethics policy.

  • Tested in real-world setups with real-world content
  • Tested for two weeks

FAQs

Can you add extra speakers to the Samsung HW-QS90H?

Yes. You can add up to four of Samsung’s Music Studio single speakers to create a surround experience, or you can add Samsung’s dedicated SWA-9500S rear speaker package.

Advertisement
How does the HW-QS90H connect to your TV?

The HW-QS90H supports HDMI’s eARC feature, where an eARC-capable TV (which most modern TVs are) can pass sound including Dolby Atmos through to the soundbar. It also lets you connect a single external source to it that it can then pass the pictures of through to the TV, and it can take sound in via an optical digital audio input. If you have a recent Samsung TV, the QS90H can even receive Dolby Atmos sound from the TV wirelessly.

Advertisement

Advertisement

Full Specs

  Samsung HW-QS90H Review
UK RRP £899
USA RRP $999
EU RRP €849
Manufacturer Samsung
Size (Dimensions) 1245 x 125 x 68.8 MM
Weight 8.9 KG
Release Date 2026
Sound Bar Channels 7.1.2
Driver (s) 13 in total, including four bass woofers
Connectivity Airplay 2, Google Cast, Spotify Connect and Tidal Connect
ARC/eARC ARC/eARC
Colours Black
Audio Formats Dolby Atmos, DTS:X, Eclipsa Audio
Rear Speaker Optional

Source link

Continue Reading

Tech

This cute AI sensor helps you care for your plants

Published

on

Senso is one of those products that tries to turn something traditionally passive – in this case, plant care – into a more interactive, data-driven experience. It does that by blending environmental sensing with AI guidance and a layer of gamification.

At its core, Senso is a modular smart plant sensor that tracks soil moisture, temperature, and light in real time. We saw the product in person at Beyond Expo 2026 and it’s designed to support a wide range of setups, offering support for over 12,000 plant types. It also has expandable accessories that let users adapt it to different pots and growing conditions.

Alongside the sensor data, its companion app uses AI-driven analysis to turn raw readings into actionable care instructions. Instead of simply telling you that soil is dry, for example, it aims to interpret what that means for a specific plant species. Then, it can suggest targeted next steps before issues become serious. As an owner of several dramatic tropical houseplants, this honestly sounds like a game changer.

There’s also a strong focus on engagement through gamification. A pixel-style companion character “awakens” when the sensor is placed in soil and visually reacts to the plant’s condition. As you water, adjust sunlight, or maintain care routines, the character evolves alongside the plant – a bit like a Tamogotchi.

Advertisement

The modular hardware design is another practical touch. The detachable sensor system intends to make cleaning, recharging, and repositioning easier. This matters if you’re moving it between different plants or environments. Moreover, the app’s multi-plant management system reinforces this flexibility. It does so by keeping everything organised in one interface rather than treating each plant as a separate setup.

Advertisement

On the software side, Senso combines continuous sensor data with AI-based visual analysis. This aims to identify plant species and flag potential issues early. The promise here is a more proactive approach to plant care – not just reacting when something goes wrong, but anticipating problems before they become critical.

Advertisement

Source link

Continue Reading

Tech

Meta Copies Snapchat’s Homework Again With ‘Plus’ Features for Instagram and Facebook

Published

on

Meta announced new upcoming subscription plans for Instagram, Facebook, and WhatsApp this week. Instagram isn’t going behind a paywall, no. Rather, users will now be able to pay $4 a month for extra features, like seeing who rewatched your story post or pinning more posts to the top of your profile. Instagram Plus, Facebook Plus ($4 a month), and WhatsApp Plus ($3 a month) will roll out globally sometime this summer.

These “Plus” plans are an attempt by Meta, led by CEO Mark Zuckerberg, to diversify how it makes money from users. Meta is also doing what Meta does best with this move: mimicking other social media platforms’ successes—specifically, Snapchat.

“Loving husband, father of four boys, VP Product @ Meta” reads Evan Spiegel’s tongue-in-cheek LinkedIn profile. Spiegel, the cofounder and CEO of Snap, has never actually worked at Meta, though his social media platform has directly inspired at least some existing features on Instagram. After Instagram launched Stories in 2016, then-CEO Kevin Systrom didn’t mince words about how his platform was sometimes iterating on Snapchat features, telling TechCrunch that “they deserve all the credit” for the format of Stories.

In 2017, Snap launched a “Maps” tool where users could opt in and see the pinpoint location of where all of their friends were based on when they last opened the app. Instagram launched a very similar “Maps” feature just last year, where users could track the location of friends who chose to share their GPS data.

Advertisement

And while IG’s recently launched “Instants” app, with its ephemeral, unfiltered snaps, is more like the once-popular BeReal than anything, disappearing photo messages are totally Snapchat’s main lane.

“As we shared earlier this year, we’re testing and scaling new subscriptions that provide deeper, more enhanced ways to use our apps and AI glasses,” says Maria Cubeta, a Meta spokesperson, over email. “So far, we’ve been testing subscription features for people to enhance how they express themselves and connect on Instagram, Facebook, and WhatsApp, like profile customization, stories insights, and super reactions.” She says this is just the start of the larger “Meta One” umbrella of subscription offerings, which will eventually include different, more expensive tiers catering toward businesses and creators as well as users who want extra access to Meta AI.

At launch, Instagram Plus will have the most extra features, from additional pins on your profile and unique bio fonts to creating siloed audiences for your Story posts and “Super Heart” reaction buttons. Facebook Plus will mainly allow you to control and customize your experience with Story posts, like rewatch insights. WhatsApp Plus will include more pinned chats, visual customizations, and premium stickers.

These subscription plans are simply history repeating itself. Snapchat dropped a $4 a month plan, called Snapchat+, back in 2022. It offered users access to exclusive features in the app and expanded over the few years, adding more options as well as AI tools. In February, Snapchat announced that this style of subscription plan helped the company achieve a “$1 billion annualized revenue run rate” in direct payments with over 25 million current subscribers around the world. (Despite being a font of feature inspiration, everything isn’t rosy at Snapchat, as the company struggles to turn a profit.)

Advertisement

So, even down to the Plus naming convention, Meta seems to be heavily inspired by Snapchat once again. I would expect nothing less from the company that renamed itself during the metaverse fad. Following trends is Meta’s modus operandi. Keep an eye out for what Snapchat does next for a sneak peek at what Meta might do in a few years.

Source link

Continue Reading

Tech

Microsoft Allegedly Leaked Dutch Civil Servants’ Data To the US

Published

on

An anonymous reader quotes a report from Cybernews: The technology giant Microsoft has been accused of leaking the data of civil servants working for the Netherlands’ regulatory agencies to the US House of Representatives. The civil servants affected by the leak work at the Authority for Consumers and Markets (ACM) and the Dutch Data Protection Authority (AP), according to the NL Times. They are involved in implementing the Digital Services Act (DSA), the European Union regulation on online services, aimed at combating illegal content and protecting user rights.

NL Times reports that Microsoft shared emails, minutes, and invitations sent by the civil servants without redacting their names in the documents. Willemijn Aerdts, Dutch State Secretary for Digital Economy and Sovereignty, said she discussed the allegations with US Ambassador to the Netherlands Joe Popolo. […] The allegations against Microsoft further strengthen concerns over Europe’s dependence on American technologies, which poses major risks to data privacy. Further reading: Netherlands Blocks US Takeover of Vital Digital Supplier

Source link

Continue Reading

Tech

Microsoft 0-day feud escalates as researcher threatens another Windows exploit dump

Published

on

The ongoing saga of Microsoft versus Nightmare Eclipse (aka Chaotic Eclipse), the disgruntled bug hunter with a deep understanding of Windows and an even deeper grudge against Microsoft, reached a fever pitch, with the researcher, who has thus far released six Windows zero-days, promising a “bone shattering” drop on July 14. 

Microsoft, for its part, finally responded to the security researcher and their weaponized Windows flaws with a blog post on (un)coordinated vulnerability disclosure about the now-public bugs: RedSun, UnDefend, BlueHammer, YellowKey, GreenPlasma, and MiniPlasma. Redmond says that none of these were reported via its official channels prior to being made public. 

Attackers began hammering three of the six – BlueHammer, RedSun, and UnDefend – soon after Nightmare published working proof-of-concept exploit code for each on now-banned GitHub (owned by Microsoft) and GitLab accounts. 

YellowKey, GreenPlasma, and MiniPlasma still don’t have fixes, and Microsoft has deemed “exploitation more likely” for YellowKey, aka CVE-2026-45585, citing a working POC.

Advertisement

“We remain firmly opposed to these actions, and any disclosure outside proper coordination that could harm our customers and the digital ecosystem,” Microsoft wrote in a Wednesday blog, and then seemingly threatened legal action against Nightmare:

“Uncoordinated disclosures that put proof-of-concept code for unpatched vulnerabilities into the hands of bad actors are never justifiable and have real-world consequences. Our security teams across the company work tirelessly tracking threat actors who look for weaknesses just like these to attack Microsoft and our customers. Our Digital Crimes Unit will continue bringing cases against these actors and those that enable their criminal activity – coordinating as needed with law enforcement around the world.”

Microsoft did not respond to The Register’s questions, including whether its legal team planned to sue Nightmare, whether the zero-day researcher is a current or former employee, and whether Microsoft axed Nightmare’s MSRC account, meaning that the bug hunter can’t disclose vulnerabilities to the Windows giant. 

Nightmare, in their latest anti-Microsoft missive, claims Microsoft did just that.

Advertisement

“When I actively asked you to communicate with me, you refused, humiliated me and made sure to insult me in front of people,” they wrote on Saturday. “You defame me in public with your CVE-2026-45585 advisory even though you literally deleted the Microsoft account I used to report bugs to you with and I got zero pennies from doing so and I still happily did like an idiot.”

Mark this date July 14th, I will make sure your bones are shattered that day

Nightmare also noted that “Microsoft still has chains in my hands,” preventing them from releasing “documents” yet, or anytime in June, and then warned: “Mark this date July 14th, I will make sure your bones are shattered that day.”

Regardless of what does or does not happen on July 14, Nightmare has already caused chaos – and real enterprise-level damage, as systems engineer Muhammad Qasim Shahzad said on LinkedIn. 

Advertisement

“One person caused more enterprise-level damage in six weeks than most APT groups cause in a year,” Shahzad wrote. “The gap between disclosure and weaponization is now measured in hours, not days. Your patching window is shrinking fast.”

Zero Day Initiative’s bug hunter-in-chief Dustin Childs, who previously spent about seven years working for Microsoft security and has decades of experience on both sides of the coordinated vulnerability disclosure (CVD) process, told The Register that Microsoft could have handled this better. And he wondered what happened between the two parties to get to this point.

“CVD is a two-way street,” he said. “The vendor has some responsibility as well, so to go out publicly stating this person violated CVD without showing any of the correspondence seems bold.”

Microsoft could also improve its communications to customers on “what the real risks from these bugs are and how they can defend themselves,” Childs added. “That clear direction seems to be missing.”

Advertisement

Microsoft’s ‘dumpster fire’

Luta Security founder and CEO Katie Moussouris, who pioneered Microsoft’s bug bounty program despite execs vowing never to pay researchers for bugs, said Redmond’s response to Nightmare sends “mixed messages.”

“It confusingly claims their program ‘ensures researchers are compensated and publicly acknowledged’ in a statement answering a researcher who says he got neither,” Moussouris told The Register. “The language choices are also not deescalating. Microsoft invoked the outdated term ‘responsible disclosure,’ which I retired years ago at Microsoft because it was subjective and judgy.”

This phrase, Moussouris added, “got in the way of coordination” when the two sides disagreed about how to best protect end users.

“The mention of the Digital Crimes Unit in a post discussing vulnerability disclosure makes the post vaguely threatening, which seems intentional, but then they wrap up the post saying they welcome reports regardless of disclosure history,” she said. “No one except the parties involved can know for sure what happened between this researcher and Microsoft. Whatever the facts, it’s hard to imagine why Microsoft would not try to deescalate, if for no other reason than avoiding the chilling effect on other researchers.”

Advertisement

Security sleuth Kevin Beaumont, in his blog on the ongoing Microsoft-Nightmare Eclipse saga, called it a “dumpster fire of [Microsoft’s] own making.” 

Beaumont also used to work at Microsoft, and he noted that the Windows company previously hired a hacker called SandboxEscaper after she published zero-day POC exploits for Microsoft products – something that Redmond’s blog now describes as criminal.

“If Microsoft’s tactic is to try to criminalise not following often arbitrary ‘responsible disclosure’ frameworks, good luck defending that in court – because there’s a whole clown car of prior decision making within Microsoft and facts which would emerge in that process,” Beaumont said.

To be clear: neither Beaumont nor the researchers that The Reg spoke to support Nightmare’s zero-day antics. Childs called the “July 14” post “troubling” and Moussouris said the date plus “incendiary language … doesn’t help organizations trying to make sense of the technical risk.” 

Advertisement

‘David and Goliath dynamic’ 

Moussouris did add that this latest missive, taken in context with the earlier blog posts, “paint[s] a picture of someone who believes they have been pushed to this extreme. It is the sound of someone who believes every legitimate channel was closed to them: GitHub account deleted, payments withheld, credit stripped, then publicly accused of violating CVD after Microsoft cut off their ability to coordinate. The researcher’s grievances are serious and specific.” 

Ultimately, “the bugs are Microsoft’s,” Moussouris said. “They wrote the code and they own the risk to customers. Often researchers who previously work with a vendor respond in the extreme only when they feel there is no other choice. The power they hold is not at all proportionate to the vendor. This is a David and Goliath dynamic we don’t like to see play out, especially since it’s users who lose when coordination negotiations fail.”

While it’s a very extreme – perhaps the most extreme – example of coordinated disclosure gone wrong, it’s not an isolated problem. Researchers have been complaining about CVD, and specifically Redmond’s bug disclosure habits, for years. 

“While some companies have improved, Microsoft has not,” Childs said. “If anything, they are seen as difficult to work with, especially if your bug is Moderate instead of Critical. I’ve had researchers tell me that they stopped looking at Microsoft altogether because they were too difficult to work with.”

Advertisement

Plus, these types of disagreements between researchers and bug bounty programs will likely increase, as AI-assisted bug reports become the norm and vulnerabilities skyrocket.

“We as an industry need to take a breath, remember there are real people involved, and that poor interactions could lead to real customer risk,” Childs said. “Real-world impact is lost far too often when disclosure goes wrong.” ®

Source link

Advertisement
Continue Reading

Tech

Microsoft Debuts A More Buttoned-Up Look For Copilot

Published

on

The AI assistant had its personality stripped in pursuit of a more consistent experience.

Copilot is getting yet another visual overhaul as Microsoft reconsiders its approach to AI across Windows and its various apps. The new changes are focused on the version of Copilot accessible in Microsoft 365, and visually streamline the AI assistant to using it more consistent across apps like Word, PowerPoint and Excel.

The most striking difference in Copilot’s new look is how little color it has. You can still get Copilot to produce full-color outputs and it will reference other apps by their colorful app icons. By default, though, the Copilot interface is now a largely black and white, text-forward affair. Part of this change was driven by a desire to make everything more readable and responsive, but Microsoft suggests it’s also reflective of an attempt to “craft intelligence that feels present but not imposing.”

Advertisement

That approach also applies to the tweaks Microsoft querying the AI assistant itself. The Microsoft 365 Copilot app and the Copilot experience in Microsoft apps feature a new “prompt surface” that changes size and reveals new functions as you type. You can input a purely text-based request to Copilot and it will answer, but if you refer to the AI assistant’s other skills, like the ability to research or visualize, the text box will unfurl menu options for selecting files or guiding Copilot’s visual responses. The app’s new side panels and menus, which collapse when not in use, are another example of this approach. Importantly, these changes also apply to how Copilot appears in apps like Word. The AI is now available in a consistent location across all Microsoft 365 apps — a side pane — and works similarly to the standalone Copilot app.

The only wrinkle to Microsoft’s Copilot redesign is that, at least for now, it’s limited to the company’s productivity software. The more consumer-friendly Copilot that was introduced in 2024 and lives in Microsoft’s mobile app is still bright, colorful and (occasionally) blobby. It’s possible this more buttoned-up look will make the jump to other versions of Copilot at some point, but that may depend on where Microsoft’s AI plans land.

The company has committed to being more thoughtful about where Copilot and AI features appear in Windows 11 and even started pulling Copilot out of certain apps. It’s also changing what AI models it uses. After being an early investor in OpenAI and a beneficiary of its GPT models, the two companies have redefined their partnership. Microsoft has now started rolling out its own in-house AI models and investing in other AI companies. A visual redesign isn’t a fix for the issues Windows users had with Copilot, but it does seem like a sign that Microsoft’s AI strategy is very much in flux.

Advertisement

Source link

Advertisement
Continue Reading

Tech

Trump Loses More Control Over AI Regulation As Illinois Passes Landmark Law

Published

on

Illinois lawmakers on Wednesday passed a landmark AI safety bill (SB 315) that would require major AI companies to publish safety plans, submit annual third-party testing reports, report serious incidents quickly, and protect whistleblowers who flag emerging risks. OpenAI and Anthropic supported the bill, which could make Illinois a testing ground for state-level AI governance as federal regulation remains stalled. Ars Technica reports: To force companies to be more transparent about rapid developments, Illinois would likely rely on “the Big Four accounting and auditing firms — Deloitte, EY, KPMG, and PwC — to audit their safety practices,” [said Scott Wisor, a policy director at a nonprofit called Secure AI Project, which supported the bill]. The required independent audits will likely frustrate Trump, who has tried and failed to stop states from implementing AI safety laws as Congress stalls on passing any legislation.

For Trump, the priority has been to promote AI industry interests, but he began considering expanding federal government safety testing after Anthropic’s Mythos was released and the AI firm limited access due to safety concerns. Whether or not governments at any level are prepared to protect society from the most catastrophic AI risks remains a major concern for critics who wonder how and when governments will intervene. After inside sources started leaking the details of Trump’s AI safety testing plans, critics warned that even the federal government may lack the necessary expertise to audit frontier AI models. And it seems the same criticism extends to independent auditors that Illinois may rely on but industry insiders suggest some AI firms may not entirely trust.

Adam Kovacevich is CEO of Chamber of Progress, a trade group that opposed SB 315 and counts Google and Apple among its members. He told Wired that Illinois’ requirements “would force companies to expose sensitive systems to untested auditors in a regulatory regime that’s all liability and no standards.” Governor J.B. Pritzker confirmed his intent to sign, proclaiming that “Illinois is leading the nation in holding Big Tech accountable.”

“I look forward to signing SB 315 and working with the legislature so that AI, when used, is used responsibly,” Pritzker said.

Advertisement

Steve Wimmer, a senior policy and technical advisor for the Transparency Coalition, said his group considers the law to be “one of the most important pieces of legislation in 2026.”

Source link

Continue Reading

Tech

Attack Of The Atomic Oxygen

Published

on

While designing anything for operation in space has its challenges, there is at least one thing that is more of a problem for objects in Earth orbit than for deep-space probes: atomic oxygen. We like oxygen because we need it to live, but it is also highly reactive as a single atom. Luckily, on Earth, most of what we breathe is O2. [Space Daily] talks about the challenges of the International Space Station dealing with the “space weather” of atomic oxygen in low Earth orbit.

Part of the problem is that even when we know better, we tend to think of the atmosphere coming to an abrupt end and space being a hard vacuum. But in reality, the atmosphere gradually dissipates, and at “only” 400 km above the Earth, the Space Station is really flying through a very thin atmosphere.

To compound the problem, this is above the ozone layer, so the Sun’s UV light rips O2 into single oxygen atoms. Over time, these free oxygen atoms can affect many parts of a spacecraft exposed to them. Engineers first noticed that materials recovered from spacecraft had more damage and changes to material properties on the pieces facing the direction of travel. NASA has spent years testing different materials by mounting trays of different material samples outside the ISS.

Carbon-based polymers take a big hit from atomic oxygen exposure. Polymide film is frequently used, but it erodes with exposure. Carbon composites also lose mass. Other materials change in other ways. For example, an optical surface may roughen with exposure.

Advertisement

The usual answer is to over-design for mission objectives or to cover certain polymers with coatings like silicon dioxide or aluminum oxide, which are not as reactive to free oxygen. For a long-duration mission like the ISS, you may have to pay special attention to the materials in use. Very low satellites also need special care, as there is more oxygen in lower orbits.

There are other effects, too, such as extreme thermal cycles, debris strikes, and other indignities that space-traveling materials must withstand. But in deep space, atomic oxygen is a rare issue. Until, at least, we go somewhere else that has a lot of oxygen.

Source link

Advertisement
Continue Reading

Tech

The $6 Billion Chinese Startup Trying to Build Hands for Every Robot

Published

on

If you could buy a humanoid robot for less than a smartphone, would you? Would you buy several robots to handle cooking, cleaning, babysitting, and even your job?

This is the pitch being made by Zhou Yong, the 40-year-old founder and chief technology officer of LinkerBot, one of China’s leading manufacturers of dexterous humanoid hands. The startup’s hardware comes complete with five fingers and at least 11 joints and is sold for as little as $600 in China. LinkerBot’s hands can play piano, thread needles, tighten screws, and assemble electronics. In three to five years, Zhou predicts, the price for one will fall to just $200. Eventually, “everyone will own ten robots on average,” Zhou said in an exclusive interview with WIRED.

Marketing spectacles like the humanoid robot marathon in Beijing have drawn attention to robots’ legs, but the real frontier in humanoids is hands. “The hands are the majority of the engineering difficulty of the entire robot,” Elon Musk said at an event last fall. Founded in 2023, LinkerBot has quickly emerged as a market leader in the space. The company says it shipped 10,000 robotic hands last year, representing 80 percent of worldwide demand. Its clients include research labs, manufacturers, and other humanoid robot makers.

The startup is also a venture capital darling: It completed six rounds of fundraising in just 13 months from investors including the Chinese government, Alibaba’s Ant Group, and HongShan Capital, Sequoia Capital’s Chinese spinoff. LinkerBot is now seeking another round of financing at a $6 billion valuation, double what the company said it was worth only a few months ago. And it’s reportedly exploring going public in Hong Kong, according to Bloomberg. (Zhou declined to comment on the rumored plans.)

Advertisement

In 2019, after selling a previous startup focused on autonomous driving, Zhou turned his attention to robotics. He says he predicted the industry would begin booming around 2025, but was still taken aback by how quickly it grew. While OpenAI was once at the forefront of developing robotic hands, in recent years Chinese startups have taken the lead as many of their American counterparts shifted their focus toward large language models and other AI software.

For robotics companies, “the valuation gap between the Chinese and US primary markets has been basically erased,” Zhou says.

Zhou says his lifelong goal is to make a real-life version of Doraemon, the Japanese anime character that has an infinite supply of magical gadgets in its pocket. (His WeChat avatar is a picture of Doraemon.) He sees building a capable, dexterous hand as an instrumental step toward achieving that dream.

Courtesy of LinkerBot

Advertisement

Selling Shovels to Miners

Successful companies, Zhou argues, focus on doing one thing well. That’s why LinkerBot zeroed in on hands, rather than trying to build the entire body of a humanoid. That also allows it to avoid directly competing with leading humanoid companies like Unitree or Tesla.

“When the humanoid robot industry size is so massive, specializing in making hands is like selling water or shovels [during the gold rush],” says Hong Shangguan, a veteran investor in China’s tech industry and a former partner at the Beijing-based fund Legend Capital.

Source link

Advertisement
Continue Reading

Trending

Copyright © 2025