TechRadar News

Anker's Soundcore Space One headphones were a big hit when they launched in 2023, delivering a bargain alternative to Beats headphones with built-in ANC for a very low price. And now there's a new model, the Space One Pro. The brand says that these are its most premium headphones ever, and while they're a little more expensive than their predecessors, they're smartly designed and have an impressive specification to rival the best noise cancelling headphones.

The previous Soundcore Space One headphones were $100 / £100. These new Soundcore Space One Pro are $199 / £149 (about AU$290), so they're stepping closer to the likes of the Sony ULT Wear Headphones or the Sennheiser Accentum Plus. There's also a carrying case but that costs extra: $34.99 / £29.99. However, if you're an early buyer before 26 September, you'll get the case for free. 

Here's what's new.

(Image credit: Soundcore) Soundcore Space One Pro headphones: key features

The most obvious new feature is what Soundcore calls FlexiCurve, a structure that enables you to fold the headphones into a donut-shaped form and which reduces their overall size by 50%. That's achieved through a new multi-segment headband and a three-joint rotator base. We love a good set of folding cans, and it's one of the features we look for when rating the best travel headphones.

The noise canceling here comes via six microphones including four feed-forward mics, and the drivers are 40mm in size. Battery life is a claimed 40 hours with ANC on and 60 hours with ANC switched off, and quick charging gives you up to eight hours of play time from a five-minute charge.

There are two models, which you can see in the image at the top of the page: jet black and cream white. Both models will go on sale in the US and the UK from 5 September – we don't have any information about an Australian launch yet.

We've already had a chance to extensively test these headphones, and you can read our full Anker Soundcore Space One Pro review for more, but we praised their "rich, well-weighted all-round sound" and noise cancellation that's "a cut above" more mid-range headphones. We also found that the comfort level "doesn’t get much better" than these. However, we found the build quality to be a little disappointing, and in the US especially, the launch price is dangerously close to the price you can regularly get the Sony WH-1000XM4 for these days, which remain excellent headphones still.

We gave them a high rating though, and they're definitely worth checking out if you want a pair of the best wireless headphones that aren't hideously expensive.

You might also like

• These elite headphones mix planar magnetic and dynamic drivers for hi-res, noise-cancelling audio nirvana
• Bang & Olufsen unveils 'most advanced and luxurious' headphones yet, with Dolby Atmos, double the noise cancellation, and user-replaceable parts
• Dyson OnTrac review: a valiant second attempt that’s blown away by the competition
  

Zillow is boosting its AI search tool to end the tedious filtering of locations and other elements. Now you can just ask for a place within a travel distance of a location and it will handle the rest for you.

The upgraded AI tools, available on the Zillow mobile app, rely on casual language rather than drop-downs and other menus. The idea is that you won't have to specify affordability, proximity to schools, commute time, and related matters. You can just explain what you want as though you are talking to a human. 

For instance, Zillow described how you could ask for "homes 30 minutes from Millennium Park," “3-bedroom houses near Roosevelt High School,” or "apartments near Denver Union Station" for properties within a certain distance to your office, a good school, or a park you like. The point is to make it more intuitive to find your ideal home. It's a linguistic upgrade to Zillow's current AI features, which currently perform tasks such as estimating a home's value and sharing AI-powered Showcase listings that tour within a house.

"From streamlining the home search to personalizing the user experience, Zillow applies AI in practical ways to help people get home," said Zillow senior vice president of AI Josh Weisberg. "Search is one of the bedrocks of our platform, and we're always improving it to make it easier for users to find homes that meet their unique needs."

AI Realty

As AI becomes increasingly integrated into platforms like Zillow, the home search process could become less painful, even if the idea of home-buying being easy seems more like a hallucination than anything an AI has produced. 

Zillow's AI will also use the conversations and requests people submit for further training. Zillow claims that as more people use the platform, the AI will be better at pinpointing the places people might actually want to buy or rent.

There's no telling if real estate agents will consider Zillow's AI upgrade a win because it could shorten the home search process or a problem because it might undercut the relationship they're trying to build with potential home buyers.

"The rise of AI in the real estate sector signals a broader shift in how people will search for homes in the future," Weisberg said. AI could ultimately make the home search process more personalized and efficient. But don't expect it to do anything about your down payment.

You might also like...

• OpenAI unveils SearchGPT and prepares to take on Google
• Perplexity's AI search could eliminate the need for follow-ups and beat ChatGPT at its own game
• OpenAI may be working on AI that can perform research without human help – which should go fine

Volkswagen has upgraded the voice assistant in its cars to be a lot more talkative than before, thanks to the power of ChatGPT. OpenAI’s conversational companion has been integrated across VW vehicles as promised back at CES, including the newest Jettas and electric cars. 

ChatGPT is more of an upgrade than a complete reworking of VW’s infotainment center. The chatbot relies on VW’s long-time voice AI partner, Cerence, and the Cerence Chat Pro platform to enhance VW’s existing Ida voice assistant. The result goes well beyond the usual navigation and climate controls. Leveraging OpenAI’s creation, a driver can get everything from local entertainment suggestions to a spontaneous trivia quiz. 

Volkswagen’s upgraded voice assistant offers a more intuitive user experience. Activated by phrases such as “Hello IDA” or “Hello Volkswagen,” depending on the vehicle, drivers and passengers can now interact with their cars in a more casual way. If Ida can’t help you, it connects to the cloud, where Cerence Chat Pro processes the query. This cloud-based system taps into multiple databases and sources, including ChatGPT, to answer your question.

"We are proud to partner with Volkswagen Group to deliver immediate value to their drivers as they look to leverage generative AI while on the go," Cerence CEO Stefan Ortmanns said. "These new capabilities – which will continue to expand – make the IDA voice assistant more knowledgeable, capable, and valuable than ever and enable a safer and more enjoyable journey."

Voice Clone of Your Own

One important element is how the voice assistant deals with data privacy. Volkswagen emphasized that the AI does not access sensitive vehicle data and that all session data is deleted immediately after each use. Most 2025 Volkswagen models will include the subscription as a free trial when they hit the market. Others will require you to subscribe to VW’s Plus Speech with AI, part of Volkswagen’s Car-Net system to begin with, but that could change as ChatGPT becomes embedded in more VW vehicles.

Voice assistants are a common feature in modern vehicles. But VW sees ChatGPT as more than just a minor perk. A more advanced solution than traditional voice assistants could help entice those looking for an AI that can handle natural conversations while driving. Though Mercedes-Benz and other carmakers are experimenting with ChatGPT and other generative AI models for their voice assistants, VW’s early adoption might keep it ahead in the endless race to win new car buyers.

“Volkswagen has a long tradition of democratizing technologies. As a volume manufacturer, we make these technologies accessible to large numbers of people,” Volkswagen Brand Board of Management member Kai Grünitz explained in a statement this summer. ”By seamlessly integrating ChatGPT into the backend of our voice assistant, we are now offering drivers the opportunity to use this artificial intelligence on a daily basis, thus underlining how innovative our products are.”

You might also like...

• Volkswagen brings ChatGPT to its cars for AI conversations, but is that a good idea?
• Mercedes-Benz is bringing ChatGPT into cars for the first time
• ChatGPT won't let you give it instruction amnesia anymore

The Border Gateway Protocol (BGP) is flawed, and needs to be fixed. Repairing this protocol would minimize data theft, extortion, state-level espionage, as well as the disruption of security-critical transactions. This is the conclusion of a new roadmap document, published earlier this week by the White House.

The document is called “Roadmap to enhancing internet routing security”, and it discusses the problems, and potential solutions, of BGP.

The Border Gateway Protocol (BGP) is the primary routing protocol used to exchange routing information between different autonomous systems (AS) on the internet. In other words, it’s the glue that holds the entire internet together.

Espionage and data theft

It enables routers to determine the most efficient paths for data to travel across the vast expanse of interconnected networks that make up the internet. BGP is crucial for maintaining a stable and scalable internet by allowing networks to share reachability information and make routing decisions based on a variety of policies.

But the protocol was designed back in 1989, and security was more of an afterthought. As a result, BGP has been abused multiple times throughout the years in some high-profile attacks. For example, in 2008, a Pakistani ISP wanted to block access to YouTube within Pakistan but accidentally announced a more specific BGP route that led to YouTube’s global traffic being redirected through Pakistan. This caused a worldwide outage of YouTube for several hours.

Two years later, China Telecom advertised incorrect BGP routes that caused a significant amount of global internet traffic, including that of U.S. government and military sites, to be routed through China for about 18 minutes. China claimed it was an incident, while some researchers in the west thought it was a deliberate attempt at cyber-espionage.

In 2018, attackers hijacked BGP routes for Amazon’s Route 53 DNS service to redirect traffic intended for MyEtherWallet, a popular cryptocurrency wallet service, to a malicious server. The attackers then stole users' cryptocurrency by tricking them into entering their credentials on the fake site.

The solution is an authentication scheme called Resource Public Key Infrastructure (RPKI) - a security framework designed to enhance the security of the Border Gateway Protocol (BGP) by providing a way to cryptographically verify the ownership of IP address blocks and the authorization of networks to announce specific routes.

“To that end, this document serves as a roadmap to increase the adoption of technologies that address critical vulnerabilities associated with the Border Gateway Protocol (BGP) and drive improvements in Internet inter-domain routing security and resilience,” the White House’s document concludes.

“This roadmap is not a technical guide on how to implement routing, but rather points to best-available guidance and practices, details United States Government (USG) actions to promote BGP security, and makes recommendations to improve routing security throughout the Internet ecosystem.”

Via The Register

More from TechRadar Pro

• Did Russia's biggest ISP just try and steal Apple network traffic?
• Here's a list of the best firewalls around today
• These are the best endpoint security tools right now

Apple's Touch ID has had a good run, but when the iPhone SE switches to what most people think will be an OLED display with Face ID it will mark the end of the line for the more than decade-old technology.

This last vestige of not only Apple's circular fingerprint reader but of the iPhone's once-iconic home button has nowhere to go, no place to hide. No other piece of Apple hardware, mobile or otherwise, uses it. When that iPhone SE redesign comes (likely not, though, as part of the iPhone 16 Apple Event), the Apple Home button and Touch ID will begin a slow fade into memory, and then be buried with all the other long-forgotten classic technology.

When I posted one of our stories about the rumored changes coming to the next iPhone SE, one former colleague claimed they couldn't live without Touch ID. I assured them they would survive, but I understood the devotion.

When Apple introduced Touch ID on the iPhone 5s in 2013, I lauded it in my review: "Overall, Touch ID feels easy and secure. And speaking of security, your fingerprints are not stored with Apple; instead they’re locally encrypted at a hardware level. It's a smart move, and I applaud Apple for getting this right out of the gate."

Touch ID was such a big and comparatively new idea that it sparked unusual ideas and use cases. As technology reporters, we found ourselves answering questions like "Can a severed finger access a stolen iPhone 5s?"

Turns out the answer is no. The RF capacitor sensor technology would only work with a live finger. You might surmise that in addition to verifying those tiny ridges, it could pick up the blood flow or pulse underneath the skin.

Oh, but it gets better. Since most people didn't have mobile fingerprint readers in their pockets, the arrival of Touch ID inspired people to try using the sensor with other body parts. No, not those body parts (as far as I know). There was, though, a man in Japan who figured out how to register his nipple on and then use it to unlock his iPhone 5s. Why? Your guess is as good as mine.

Letting go

Touch ID eventually became as commonplace as, well, iPhones, and we didn't start to consider the loss of this effective, tactile, biometric security feature until the arrival of Apple's iPhone X and Face ID in 2017.

As is typical of Apple, changes flow through Apple product lines in an evolutionary fashion on both iPhones and iPads (the iPod Touch died with its Touch ID-free home button intact). But, as was the case with the transition from 30-pin charge ports to Lightning and now to USB-C, changes eventually visit all Apple products and classes.

Anyone who thought the iPhone SE would somehow escape the update, slipping through innovation's hands like a greased banana, was fooling themselves. Change is inevitable in all things, and a requirement for technology.

And yet, I still feel a twinge of sadness at the impending end of what was once a symbol of all iPhone technology. The home button, which originally featured a small printed square in the middle, was recognizable from a distance. It became slightly less so with Touch ID, which eventually traded movement for haptic response and the square for sparkling glass, latterly with a metal ring surrounding the Touch ID circle.

Perhaps we wouldn't have fallen in love with the iPhone and this little button if Apple hadn't done such a good job of creating it. As I wrote back in 2013, "Placing the fingerprint reader under the home button is a brilliant idea – even more so because the execution is nearly flawless."

So we have only Apple to thank and to blame for our devotion to this disappearing invention. Face ID is probably smarter, faster, and more secure, but we'll never forget the home button and Touch ID, and we may miss them long after the next iPhone SE arrives.

You might also like

• How to watch the iPhone 16 launch event live
• Apple's iPhone 16 launch event – 5 things we expect and 3 things ...
• Apple just sent out invites for the iPhone 16 event – 'It's Glowtime.' is ...

We still consider the original OnePlus Open to be the best foldable phone you can buy right now, but some new leaks suggest a successor isn't too far away – and could soon overshadow the imminent Honor Magic V3.

According to the prominent leaker Digital Chat Station on Weibo (via Android Authority), the OnePlus Open 2 will have "record-breaking thinness" alongside a host of other upgrades. Like before, it's expected to be a rebranded Oppo phone, this time the Oppo Find N5.

The main spec boost will apparently be a Snapdragon 8 Gen 4 chipset (up from the Snapdragon 8 Gen 2), with the Open 2 also including a triple-camera setup with a 50MP primary camera. It isn't yet clear how those cameras will differ from the current model, which already has a 64MP periscope telephoto.

Previous rumors from the same source have also hinted at a new 6,000mAh battery for the new foldable, which would be a huge upgrade on the 4,805 mAh cell in the current OnePlus Open. 

While that battery rumor is more speculative, these new design rumors seem more plausible. Foldable phones are currently battling to shrink their footprints closer to the size of non-foldables, with Honor preparing to launch the Magic V3 at IFA 2024 – and that phone is confirmed to measure just 4.35mm when unfolded and 9.2mm when folded. The current OnePlus Open measures a comparatively beefy 11.9mm when folded.

The only issue for OnePlus Open 2 coveters is that previous rumors have pointed to a launch in early 2025 rather than a year after the current model (which landed in October 2023).

Shallow crave

The original OnePlus Open (above) has a big camera bump, but does also pack in the best cameras we've seen on a foldable. (Image credit: Future / Philip Berne)

While this new battle for thinness among traditional phones – like the rumored iPhone Air – feels like a slightly unnecessary blast from the early 2010s, it makes much more sense for foldables, which are naturally a bit chunkier than the handsets we've become accustomed to.

This OnePlus Open 2 leak suggests the phone will be well under 10mm thick when closed. That would make it significantly thinner than the Samsung Galaxy Z Fold 6, which is a comparatively thick 12.1mm when folded (and 5.6mm thick when unfolded). The Google Pixel 9 Pro Fold, meanwhile, is 10.5mm thick when folded, or 5.1mm when you unfold it.

Still, who will take the 'thinnest foldable' crown between the Honor Magic V3 and OnePlus Open 2 remains to be seen. The Magic V3 measures only 4.35mm when unfolded and 9.2mm when folded.

More important will be the overall experience of using the phones – and our OnePlus Open review described it as "the only big foldable phone that doesn’t feel like a compromise." If the OnePlus Open 2 can fix its weaknesses (namely, battery life, features and its sizable camera bump), then it'll remain the favorite and retain the overall top spot in our foldable phones guide.

You might also like...

• OnePlus Open 2: what we want to see
• Best foldable phones 2024: the top folding smartphones right now
• OnePlus Open may soon become the 'Apex' foldable as hardware upgrade launches soon

While Beats has been busy with new Solo Buds and reentering speakers with the Pill, it seems the Apple-owned audio brand has also been focused on updating another product… one that’s been five years in the making, as of this story's publication. The iconic earhooked Powerbeats Pro are set to return, at least according to a new teaser video posted to the Beats by Dre Instagram. 

As shown off by LA Dodgers baseball superstar Shohei Ohtani, Powerbeats Pro 2 will arrive in 2025 and sport a redesigned, modern look. Not much else is known, aside from the fact that Ohtani can rock them while batting at home plate, and that the caption says they’re built "for him".

Powerbeats Pro 2 … landing in 2025

A post shared by Beats by Dre (@beatsbydre)

A photo posted by on

When 2025 rolls around, it will have been six years since Powerbeats Pro launched in 2019. 

In that time, Beats has continued to invest and roll out other earbuds, all of which sit directly in your ear – Fit Pro, Studio Buds Plus, and Solo Buds included. The appeal of the Powerbeats Pro is the earhook, and for those who’ve been waiting for an upgrade, faith is being rewarded. The earhook is in full force on Powerbeats Pro 2 with more subtle, rounded edges in what I’d described as a modern look.

From the teaser video, Ohtani’s Powerbeats Pro 2 appears to have a slightly thinner earhook, but it’s hard to say for sure. The bridge from the ear to the hook still has a slant – albeit a somewhat more pronounced one – but like the hook, it is dramatically thinner.

Aside from this quick look at the Powerbeats Pro 2 on-ear, everything else is up in the air, including what processor Beats is using inside, the driver for producing audio, and whether they’ve managed to shrink the case. It's safe to assume the Lightning port will be swapped for USB-C, though. Chances are a port won't be entirely removed, though, like the one on the Solo Buds.

The first-generation Powerbeats Pro in black. (Image credit: TechRadar)

The original Powerbeats Pro features Apple’s H1 chip, a custom drive with support for Spatial Audio, and nine hours of playback. In fact, in TechRadar’s review, the Powerbeats Pro scored a four out of five, and we called them a big step up from previous Beats earbuds and, at the time, “Apple’s most premium workout buds.” That latter has likely been outpaced by AirPods Pro or Beats Fit Pro, but for the ear-hook fanatic, Powerbeats still likely remains the top choice.

Those specs look a little dated now, and Beats could use this as an opportunity to boost them significantly. It could opt to use its in-house developed silicon, like with other earbuds, or tap one of Apple’s chips.

I think we all hope that the combination carrying and charging case is at least shrunk. But as someone who rocked Powerbeats Pro for a while and still uses them occasionally, I’m keen to try out the second generation and thankful that Beats isn’t pulling the plug on earhooks; if anything, they’re making them cooler.

(Image credit: Beats) You Might Also Like

• Beats Fit Pro review: great workout true wireless buds
• Bose Ultra Open Earbuds review: The best-sounding open-ear buds ...
• The best earbuds: top buds for all budgets

IFA 2024 has been bringing in plenty of tech and gaming news, and MSI has dropped its own bomb by confirming the specs for its upcoming PC gaming handheld, the MSI Claw 8 AI+.

It turns out that MSI will still be using an Intel CPU to power its system, namely the Intel Core Ultra 200V Lunar Lake SoC. MSI is also boosting the memory speed, battery life, and screen size compared to its predecessor. 

The maximum system memory is increasing to 32GB, and will move from LPDDR5-6400 to LPDDR5x-8533. Battery capacity will increase to 80Wh (nearly double the original system's 53Wh), and there will be two Thunderbolt 4 ports. The new model will support the M.2 2280 SSD form factor and the screen size will increase to eight inches, though it will not be upgraded to an OLED display.

The MSI Claw 8 AI+ is the official follow-up to the original Claw 8 A1M model which was infamously met with plenty of scathing reviews and poor reception upon launch. Since then, MSI pushed out multitudes of BIOS updates that greatly improved the performance. However, it still falls short of other superior portables like the Steam Deck, Asus ROG Ally X, and Lenovo Legion GO.

Will MSI succeed this time? 

While I do agree with MSI’s decision to start anew with the MSI Claw 8 AI+, as the A1M model was simply not cutting it even with all the performance updates, it’s always difficult to win back lost trust when it comes to pushing out a flawed product.

The new Intel CPU it’s equipped with sounds impressive, as the Lunar Lake architecture is potentially able to blow AMD mobile CPUs out of the water performance-wise. So theoretically, it makes sense why MSI is sticking with Intel - that and it makes the MSI Claw stand out more in the growing sea of gaming handhelds.

However, considering all the massive issues the original model had, it’s hard for me to trust that this next go around with Intel won’t backfire. Technically, we never received official confirmation as to which end of the manufacturing process went wrong, but many of the portable’s woes stemmed from the constant Intel driver issues and subsequent downloads fixing them.

Personally, I’m rooting for MSI to pull out a win here, as I always support plenty of healthy competition in any given market. Having yet another viable PC gaming handheld option to choose from is a win in my book, as it’ll motivate the other tech giants to keep innovating. It also seems like MSI is listening to buyer criticism and incorporating it into their final product, much like how nearly every improvement in the ROG Ally X can be traced directly back to community feedback.

You might also like

• Hands-on: MSI Claw preview: a bold step for MSI and Intel
• MSI Claw becomes the first gaming handheld to feature an Intel Core CPU
• Is your MSI Claw struggling with performance issues? Download the latest update

All physical multi-factor authentication (MFA) keys that work on Infeneon’s SLE78 microcontroller were said to be vulnerable to a cryptographic flaw which allows threat actors to clone the gadget and gain unabated access to restricted accounts. This includes the YubiKey 5, considered the most widely used hardware token based on the FIDO standard.

In an in-depth technical analysis, researchers from NinjaLab described how they discovered the flaw, and what it means for those using YubiKey 5. As explained, the SLE78 microcontroller implements the Elliptic Curve Digital Signature Algorithm (ECDSA) as its core cryptographic primitive. In short, ECDSA is a cryptographic algorithm used to create digital signatures, and if a hacker is able to read this signature, then they are able to undermine the security of the entire token.

And that’s exactly what NinjaLab did, by employing a technique known as “side-channel”. This is a type of security attack in which hackers exploit information gained from the physical implementation of a computer system, rather than weaknesses in the implemented algorithms. These attacks gather information by observing how a system operates, such as its timing, power consumption, electromagnetic emissions, or even sound.

YubiKey 5 not so easily exploited

With SLE78, generating a different ephemeral key takes varying amounts of time, and this is something the researchers were able to read, and from it clone their own YubiKey 5 (this is a super simplified explanation).

It is definitely a major vulnerability, but one that is not that easy to replicate in the wild. The attacker would need to know the victim’s login information first, and have physical access to the MFA token. Then, they would need to tear the token apart in order to access the hardware within, and use $11,000 worth of equipment to do the reading. The reading itself, and the process of cloning the device, only takes a few minutes.

This isn’t something your average hacker could abuse, but a nation-state - absolutely. It’s also worth mentioning that there is no patch, or fix - all YubiKey 5 devices running firmware prior to version 5.7 are permanently vulnerable.

Via Ars Technica

More from TechRadar Pro

• Uncovering the cybersecurity industry’s senseless fixation with security keys
• Here's a list of the best firewalls around today
• These are the best endpoint security tools right now

One of the best VPN services on the market has just become an early mover in delivering a VPN app compatible with next-generation Windows devices.

ExpressVPN has closely collaborated with Qualcomm (developers of the Snapdragon X Elite processor) and Microsoft to accelerate the delivery of a fully ARM-compatible Windows VPN app. The team eventually found a revolutionary solution to avoid building a native ARM application from scratch.

As of September 4, the ExpressVPN ARM app is now available in beta, with a full production launch set for the middle of October.

A hybrid solution 

While new Microsoft Copilot+ PCs powered by Snapdragon can deliver better battery life and super-fast performances, they have been worryingly lacking in VPN compatibility. This is why most VPN apps still aren't ready to support Qualcomm's high-performance ARM-based processor. This was something that the ExpressVPN team knew it needed to fix

"Snapdragon is one of the most exciting things to happen to the Windows platform in recent times, and we saw this as an opportunity to showcase a high-performance VPN solution tailored for ARM's capabilities," said Peter Membrey, Chief Engineering Officer at ExpressVPN.

The main issue with VPN compatibility on ARM Windows machines was that emulation posed serious performance limitations that could affect VPN apps' functionality. Now, ExpressVPN has found a way to fix this without the need to build a native ARM application from scratch.

As the provider explains in a blog post, the release of Microsoft’s new Prism emulator with the Windows 11 24H2 upgrade has made it possible for emulated apps to match the quality of a native build. The team then decided to harness this new opportunity by shaping its in-house built WireGuard-inspired VPN protocol, Lightway, accordingly.

"By leveraging Microsoft's enhanced emulation and Lightway's adaptable architecture, we were able to solve initial limitations and deliver this build efficiently," Membrey explains.

(Image credit: ExpressVPN)

The provider promises such a hybrid approach can offer users "the best of both worlds," namely the full security of a premium VPN service and accelerated development time.

As we mentioned earlier, just a handful of top VPN services are available for ARM Windows devices at the time of writing. Private Internet Access (PIA) launched its ARM-native app about a month ago. Windscribe and Surfshark also have ARM-native Windows VPN apps, while the likes of NordVPN, and ProtonVPN are currently working on their versions.

ExpressVPN now believes that its innovative approach to ARM compatibility marks a significant milestone in VPN technology. This could then support other providers fasten their transition to ARM.

Membrey said: "We’re proud to be one of the first major VPN providers to demonstrate our commitment to this platform, and we will continue to optimize Lightway and our software to ensure that users can enjoy the best that Snapdragon, Microsoft, and ExpressVPN have to offer."

Microsoft’s line-up of Copilot+ PCs is designed to run AI tasks in a superfast manner, along with providing welcome features like all-day battery life, but what about gaming?

Okay, so these are AI laptops, and not designed for gaming, that’s true - but what about dabbling in some PC games casually on your notebook? While it’s not unheard of for Microsoft to make questionable software design decisions, a new report gave me pause for thought, asserting that the first wave of Copilot+ PCs - powered by Arm-based Snapdragon X chips - don’t perform well even for more casual gaming in some cases.

Indeed, according to the report from The Wall Street Journal (flagged up by MacRumors), those Snapdragon X-powered Copilot+ laptops might struggle to run a large number of PC games. In extensive testing, the WSJ found that of 1,300 PC games, only half of them ran smoothly without noticeable issues or glitches.

While problems with running PC games on Arm-based Copilot+ laptops have been well-documented since the launch of these devices, the failure rate mentioned here is an eye-opening one.

(Image credit: Arm) Compatibility challenges for Arm devices

The problem, as you may be aware, is that Arm is a different architecture from traditional x86 chips that PC games are written for. So, to run those games on an Arm device, Microsoft needs to employ a software layer to translate x86 games (and apps) to get them working using a system called Prism.

Prism works as an emulation layer, constantly converting instructions designed for Windows devices with x86 chips into instructions for devices with Arm chips. Apple has a similar tool for its Macs that you’ve probably heard of, called Rosetta (we’re now on Rosetta 2, in fact), allowing Macs with M-series (Arm-based) chips to run x86 apps.

The trouble is that emulation inevitably introduces performance overheads, and the whole process can lead to compatibility problems in some cases, with Prism apparently not faring that well on this side of the equation as the WSJ report makes clear. On Arm-based Copilot+ PCs, as mentioned only half of tested games ran without nasty glitches or problems launching. Some popular games like Fortnite and League of Legends can’t be made to work on Arm-based devices at all right now (anti-cheat systems are a particular bugbear, as we’ve heard in the past).

Clearly, this is an issue that Microsoft must continue to work on. I’m stating the obvious here about apps and games, but being able to choose freely among the existing releases out there is a huge part of what makes a computing device usable for any given person.

(Image credit: Future / Lance Ulanoff) Copilot+ PC expectations and the future

For now, Microsoft has told users that if they want to play games with demanding graphics, a Copilot+ PC may not be for them. And to be fair to Microsoft, this is not one of the core aims of a Copilot+ laptop, and it was never intended to be - what is slightly worrying is the gravity of the problem as painted by this new report from the WSJ.

I hope Microsoft will address these issues with gaming on Arm silicon and continue to improve its Prism emulation tech going forward, which to be fair, is still in its early days (Rosetta is now on version 2, as noted). In the world of Copilot+ PCs, though, we are going to start seeing laptops with x86 chips inside that won’t have these compatibility issues - notebooks built on Intel’s Lunar Lake processors, and AMD’s Ryzen AI 300 silicon. This is to say this is not an issue with Copilot+ PCs as such, but with Arm chips - at least for now.

YOU MIGHT ALSO LIKE...

• Affordable Copilot+ PCs might be coming – but leak shows their 8-core Snapdragon CPU won’t cope with gaming
• CoPilot+ PCs could finally get more affordable soon – here’s what's coming
• I've been using a work laptop with a dedicated Microsoft Copilot key for months now, and it's made me realize one thing

Research from Searchlight Cyber has shown the number of ransomware groups that operated in the first half of 2024 rose to 73, up from 46 in the same period of 2023. The findings suggest law enforcement’s efforts to curb cyber criminal groups have seen some success, especially in disrupting the operations of notorious group BlackCat, which has since dissolved.

Groups were targeted by law enforcement in ‘Operation Cronos’, which facilitated the arrests of two people, took down 28 servers, obtained 1,000 decryption keys, and froze 200 crypto accounts - all linked to the infamous LockBit organization.

Although the number of groups has risen, the number of victims has fallen, which indicates a potential diversification rather than growth of ransomware groups. Other Ransomware as a Service (RaaS) groups such as RansomHub and BlackBasta have become more active, complicating the landscape for cyber security.

Persistent threats

The disruptions of cyber criminal activities should not be mistaken for the conclusion of operations. New organizations such as DarkVault and APT73 are expected to become more prolific in the near future.

Head of Threat Intelligence at Searchlight Cyber, Luke Donovan comments, “As we've seen in the first half of 2024, the ransomware landscape is not just expanding, it's fragmenting. With over 70 active ransomware groups now in operation, the ransomware landscape is becoming more complex for cybersecurity professionals to navigate.”

He adds, “The diversification we're witnessing means that smaller, lesser-known groups can emerge rapidly and execute highly targeted attacks.”

Recently, groups like Qilin have caused serious damage attacking NHS hospitals, which affected surgeries and transplants. The risks posed by these threat actors is illustrated in their willingness to attack high-impact targets in order to leverage as much ransom as possible.

More from TechRadar Pro

• Take a look at some of the best cloud firewalls
• The UK public is not confident about NHS cybersecurity at all
• Check out our pick for best malware removal software

Applications determined by Google to be not secure enough will have Workspace access disabled as of September 30.

Google’s previously announced cutoff for “less secure apps” (LSA) aims to lower the number of third-party applications that do not use secure authorization from Workspace accounts.

Only third-party applications that use OAuth will be trusted to remain connected to Workspace.

Admins beware

OAuth is an authorization protocol that allows you to use third-party applications without sharing your credentials with them, by providing an authorization token to allow interaction.

Google’s intention behind the cutoff of LSAs is to reduce the number of third-party apps that have access to your username and password to prevent attackers that compromise your credentials from having easy access to your account and networks.

As a result, CalDAV, CardDAV, IMAP, POP and Google Sync will all require OAuth to use, rather than just a password. Those using Thunderbird, the mail app for iOS or MacOS, or Outlook for Mac will simply need to re-add your account and sign in using Google with OAuth.

“Admins will need to push a Google Account using their MDM provider, which will re-add their Google accounts to iOS devices using OAuth,” Google said in the announcement.

For those using Outlook 2016 or earlier, it’s time to upgrade to Microsoft 365, as it will be impossible to log in to older versions of Outlook without OAuth. Similarly, any third-party application that does not use OAuth will no longer work, so you may need to consider an alternative app.

More from TechRadar Pro

• These are the best productivity apps
• Google Workspace will now save you from making any embarrassing file-sharing mistakes
• Take a look at the best business laptops

What if I told you that there was a way to make the quality of movies and TV shows you watch through the best streaming services better, even if your Wi-Fi isn’t the greatest? It sounds ideal, right? Well, I've got some good news and bad news…

The good news is that the next-gen video codec, Versatile Video Codec (VVC), also known as H.266, has arrived in Intel’s latest chips (as reported by FlatpanelsHD). The bad news? These chips, which are part of the Intel Core 200 series launched at IFA 2024 (also known as Lunar Lake, which are set to do battle with AMD and Qualcomm), are coming first to a range of "AI PC’s" from the likes of Lenovo, Asus, Acer, Dell, Samsung and more at the end of September 2024. 

While no official announcement has been made about when these chips will arrive in TVs, we're hopeful that it will eventually, because it would improve video encoding by making it more efficient, which would result in better quality video streaming, even for those with sub-par Wi-Fi strength. The unfortunate reality, however, is that we’re unlikely to see VVC in TVs any time soon. 

We'd love to see the VVC codec available on 4K and 8K TVs, such as the 8K Samsung QN900D pictured here.  (Image credit: Future)

Currently, TVs use the older High Efficiency Video Encoding (HEVC), or H.265, and in some cases the higher quality AOMedia Video 1 (AV1) video codec. These video codecs compress video files to make them easier to send over the internet, with the AV1 offering higher efficiency and better video quality over lower bitrates, which results in smoother streaming. But, AV1 does perform at a slower speed and have longer encoding times due to its complex algorithms. 

HEVC has been in use on TVs for years and AV1 was eventually introduced first to 8K TVs – like the best 8K TVs – in 2020 and then in 4K TVs in 2021. Some of the best TVs from the likes of Sony and Samsung also support it, while others continued to use HEVC. With that being said, just how much better could VVC be and what impact could it make?

A hope for the future of streaming 

According to the report by FlatpanelsHD, VVC has been found to offer 30-50% more compression efficiency than HEVC and 10-20% more than AV1 at 4K resolution. Meanwhile, with 8K resolution, VVC is said to offer 50-60% more efficiency than HEVC and 20-30% more than AV1. Bitmovin tested HEVC against VVC on its own Bitmovin Player app and found that VVC performed better with faster moving scenes and showed more detailing when compared with the same scene via the HEVC codec. 

This reduction in bitrate means that higher quality videos can be played back without taking up as much data, which once again means people with weaker Wi-Fi can still enjoy high-quality 4K and 8K (when it becomes available) streaming. But, it also should mean an improvement in performance, with less buffering for those with slower Wi-Fi as well. Plus, it’s great news for streaming services as the improved efficiency of VVC means less bandwidth costs, which will likely be welcomed with open arms. 

The VVC codec should increase encoding efficiency, resulting in better quality streaming on sites like Netflix.  (Image credit: Thomas Trutschel)

One of the most frustrating things about streaming is the sometimes inconsistent video quality. If you’re watching one of the best Netflix movies, for example, the last thing you want is a drop in quality, resulting in fuzzy textures and washed out colors and contrast. 

As I discovered when I tested Blu-ray vs Netflix and Disney Plus, streaming already has a disadvantage, usually topping out at a bitrate of 25Mbps during a 4K stream (Apple TV Plus can do 40Mbps), compared to a 4K Blu-ray, which runs at about 128Mbps. This means 4K streams run at about the same quality as a HD (1080p) Blu-ray, so imagine that being affected even further by a less efficient codec. The implementation of VVC won’t automatically mean 4K streaming is as high a quality as 4K Blu-ray, but it could at least make 4K, and even 8K streaming in future, a better experience. 

Worth the wait  

Although it is estimated that it could be a while before we see VVC implemented fully across TVs and streaming platforms, potentially years, VVC on the surface looks like it will be worth the wait. With streaming service prices consistently rising, with the likes of Disney Plus, ESPN Plus and Hulu increasing prices, another expected Netflix price hike and Max also raising its own prices just to name a few examples, the least we can expect is consistent performance and quality. 

With the VVC codec, it sounds like we could be heading in the right direction. General tests have found that its higher efficiency over the HEVC are significant and although it’s a closer competition between AV1 and VVC (according to reports from WinXDVD) , VVC looks more likely to be adopted by TVs in future, which is good news for streaming services looking to improve efficiency and in turn costs. 

Intel announced the new Lunar Lake series of chips at Computex 2024 in June.  (Image credit: Future / John Loeffler)

This is of course excellent news for users as well. If you’re sitting at home with one of the best OLED TVs, you’re going to want the best quality possible when streaming. So, although it may not be on the horizon just yet, the integration into Intel’s Lunar Lake chips is a good sign of wider commercial use and that hopefully mean TVs are next. Now, we play the waiting game. 

You might also like...

• Intel Lunar Lake CPUs could be revolutionary for thin-and-light laptops – but good luck buying one when they launch
• Here are my top 3 TVs I've reviewed in 2024 so far, from elite OLED to affordable mini-LED
• The best Disney Plus movies: 33 great films to stream in September 2024

The US Department of Justice (DOJ) has reportedly stepped up its antitrust investigation into Nvidia, according to Bloomberg.

Now, the DOJ has issued subpoenas to gather evidence that the AI chip giant could be engaging in business practices that enable it to monopolize the market.

The investigation addresses concerns that Nvidia could be preventing competition by making it difficult for customers to adopt rival products.

Nvidia is under the antitrust microscope

Bloomberg cites unnamed sources familiar with the matter who claim that the DOJ, which previously sent out questionnaires, is now sending legally binding requests for information.

Moreover, the DOJ also has its eyes on Nvidia’s acquisition of RunAI, an AI computing management software company. Concerns are raised that the merger could further inhibit competition.

Nvidia’s success has been largely thanks to the rapid adoption of artificial intelligence – its powerful chips have proven instrumental in setting up and maintaining high-performance AI datacenters. The company currently has a market cap of $2.02 trillion, after peaking at more than $3 trillion earlier this year and temporarily becoming the world’s most valuable company.

While companies like Intel and AMD have introduced their own products, Nvidia’s early and strong entry to the market positioned it well for future growth.

The dominating chipmaker isn’t only facing scrutiny in the US – French antitrust regulators have also launched an investigation as part of a broader look into the cloud computing market, and the EU is reportedly examining Nvidia’s business practices.

TechRadar Pro has asked Nvidia to comment on ongoing investigatory work, and we'll be sure to update this story with a response if we receive one.

More from TechRadar Pro

• These are the best AI tools and best AI writers
• Need something a little less extreme? Here are the best processors
• Nvidia was officially more valuable than Apple — for a couple of hours, at least

Cyber attacks exploiting zero-day vulnerabilities have historically been something of a worst-case scenario — a surprise attack that cannot be readily predicted. But while zero days used to be fairly rare, they are now becoming increasingly prevalent.

In two of the last three years, more of the mass compromise events Rapid7 analyzed arose from zero-day vulnerabilities than from new n-day exploits. And over the past year, major incidents involving vulnerabilities in Progress MOVEit Transfer, Barracuda ESG, Ivanti Connect Secure, and Fortra GoAnywhere MFT have underscored this trend.

As the scope of zero-day attacks grows, organizations must quickly familiarize themselves with the greatest threats and ensure they have proactive controls in place to mitigate risk.

Trends behind the rise in zero day exploits

Zero-day attacks can lead to severe consequences, including system compromises, network downtime, and substantial financial losses. However, the expertise and resources required have previously limited their use.

Discovering a new exploit requires research time and technical skill, making them the providence of more well-resourced threat groups either able to uncover the vulnerability or purchase exclusive exploits from brokers on the dark web.

But now we are seeing a distinct shift to zero days being deployed in widespread attacks — incidents where previously unknown vulnerabilities are exploited by a single skilled adversary hitting a large vulnerable target population with an orchestrated, timed attack.

Our research found that in 2023, more than half (53%) of the new widespread threat vulnerabilities were exploited as zero days, marking an increase from the previous year and a return to the high levels of zero-day exploitation seen in 2021.

Several factors are contributing to this rise. Because zero days can enable threat actors to pull off hugely profitable attacks, some threat groups are willing to pay vast sums to get their hands on new discoveries.

Rapid7 researchers have seen exploits for common enterprise tools like VPNs offered for $100K or more each on dark web forums, a powerful incentive for well-resourced cybercriminal gangs looking to increase their profits. With established threat groups taking in eight figures for large, global attack campaigns, it’s plausible that they may be able to comfortably afford the investment, bolstering a thriving underground economy.

Additionally, many of these vulnerabilities arise from simpler, easily exploitable root causes, such as command injection and improper authentication issues. These are often quicker and easier for attackers to exploit compared to more complex vulnerabilities like memory corruption flaws.

For example, vulnerabilities in Barracuda ESG and Fortra GoAnywhere MFT arose from command injection issues. Similarly, improper authentication issues have been central to many attacks on network edge devices.

Why network edge devices are at high risk

In tandem with the growing volume of CVEs and zero-day exploitation, cybercriminals have increasingly exploited network edge devices like routers, firewalls, VPNs, security gateways, and network appliances.

Edge devices present attractive targets due to their critical role in managing data flow and access. Once compromised, attackers can gain a foothold within the network, potentially allowing them to move laterally and escalate their privileges. But while they have always been a popular target, we are now seeing evidence that edge devices are increasingly being targeted en masse, with exploits that affect hundreds of organizations at a time.

Our research found that mass compromise events stemming from the exploitation of network edge devices have almost doubled since the start of 2023, with state-sponsored adversaries and ransomware groups alike racing to weaponize both new and known flaws in these technologies. High-profile ransomware groups like Cl0p, Akira, LockBit, and more have leveraged network edge device vulnerabilities in recent attacks.

Notably, 36% of widely exploited vulnerabilities occurred in network perimeter technologies. Over the last three years, more than 60% of network edge vulnerabilities have also been exploited as zero days, highlighting the value these devices offer threat groups looking to infiltrate networks in order to achieve their objectives.

Incidents involving vulnerabilities in network edge technologies, such as Citrix NetScaler ADC/Gateway, and Cisco ASA, have had significant impacts, leading to widespread compromises and service disruptions. For example, the zero-day exploitation of Barracuda Networks’s Email Security Gateway (ESG) eventually drove the company to recommend that users completely decommission some physical devices.

Proactive steps to prepare for the threat

The growing prevalence of zero-day exploits is a trend no company can ignore. Fortunately, there are multiple steps organizations can take to improve their resilience against these threats when they appear. The tried-and-true layered security strategy is key to mitigating risk. However, the growing prevalence of zero-day attacks means organizations must implement any missing controls urgently.

Frequent security assessments are important here, as they will enable security teams to build an accurate picture of what systems are most at risk. While it is not always possible to predict when a new exploit will appear, a solid understanding of the network will allow teams to understand the risks and best course of action for response.

Alongside this, regular patching and robust vulnerability management are essential. Closing off new exploit paths as soon as fixes are available will shrink the opportunity for a potential attack. Addressing other known vulnerabilities in the system will also mean fewer options are available for attackers. Patching activity should also prioritize high-value systems like network edge devices and file transfer solutions, which are prime targets for exploitation.

Finally, organizations must also be ready to act quickly when an attack does occur. Security teams can still be equipped to rapidly respond to a new attack, even if the individual exploit is initially unknown. Advanced threat detection tools, along with robust logging and monitoring capabilities, are critical for detecting indicators of compromise and attacker follow-on behavior.

The importance of MFA

Along with other proactive measures, multi-factor authentication (MFA) plays a crucial role in securing networks by adding an extra layer of protection beyond passwords. While zero-day attacks will incorporate novel exploits, many threat actors still rely on standard methods like stolen or reused credentials to execute their attacks.

Implementing MFA can contribute to security defences by reducing the risk of unauthorised access, as attackers need more than just a stolen password to breach systems. For internet-facing systems, properly implemented and enforced MFA ensures that even if credentials are compromised, additional authentication steps may be able to prevent immediate access to critical systems.

It’s not only important to have MFA implemented throughout the organization, that implementation must also be properly enforced. Sadly, 41% of incidents Rapid7 services teams responded to in 2023 were due to missing or unenforced MFA on internet-facing systems, especially on VPNs and virtual desktop infrastructure. For example, an organization may have MFA in place, but sometimes a large group of employees is placed in an MFA bypass group for convenience. So, while on paper the whole company is protected by MFA, in practice the policy is unlikely to be effective.

Remaining resilient in the zero-day surge

Zero-day vulnerabilities pose an escalating threat, and organizations must urgently adopt layered security measures to defend against the attack. There is no time to waste.

While there are multiple routes to improving resilience against zero days, making security changes can often be a painfully slow process, especially when it comes to organization-wide policies like MFA.

Further, attempting to do everything at once can often lead to limited impact — if everything is a priority, then nothing is. Security decision makers must be sure of their priorities, focusing on the issues that will have the biggest impact on resilience.

We've featured the best online cybersecurity course.

This article was produced as part of TechRadarPro's Expert Insights channel where we feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/news/submit-your-story-to-techradar-pro

The first trailer for Warner Bros. Pictures' Minecraft movie has arrived – and, as this article's title suggests, I can't decide if it's drop-dead gorgeous or the stuff of nightmares.

Officially titled A Minecraft Movie, the live-action-animation hybrid flick has also blocked out an April 4, 2025 release date in US theaters. Ordinarily, that would be the headline news for a movie adaptation of one of the world's most popular videogame franchises. But, when you watch the 75-second-long teaser trailer below, you'll see why I'm on the fence about how visually spectacular and grotesque it looks.

Let's start with the "holy heck, this looks amazing!" side of things. The cube-based aesthetic that the film's various art departments that authentically captured here is nothing short of breath-taking. I mean, just look at the landscapes that greet four of the movie's intrepid human adventurers between the eight- and nine-second marks. Take in the blocky goodness! The draw distance! The faithful recreations of the creatures that inhabit this universe! Insert the chef's kiss hand gesture emoji here.

But then we come to the goofier and even frightening portion of the trailer for one of 2025's new movies. Where the former is concerned, the blend of live-action performers and their animated world does not look good. At all. Obvious as it is to say, you can just tell that green screen technology, rather than practical effects, has been heavily utilized. As long as it wasn't used, maybe The Volume would've been a better piece of tech to employ for this film.

Kill it with fire (Image credit: Warner Bros. Pictures/Legendary Pictures)

On the latter front – I mean, just look at the pink sheep, via the image above, to see what I mean by its nightmare fuel visuals. There are other examples of... how can I put this politely... unsettling creature designs in A Minecraft Movie, too, such as the dead-eyed Pigmen, the overly expressive llama, and even the fuzzy, almost tactile take on Minecraft's iconic Creepers. I can't be the only person who's getting flashbacks to Sonic the Hedgehog's original and horrifying humanoid design for his first film, right?

Who's been cast in A Minecraft Movie? And what's its story?

Jack Black will play Steve, Minecraft's blue-shirted mascot, in A Minecraft Movie (Image credit: Warner Bros. Pictures/Legendary Pictures)

Stepping away from A Minecraft Movie's visuals for a moment, what is Warner's movie take on the videogame series – one originally developed by Mojang and now owned by Microsoft – actually about?

Here's a plot brief, courtesy of the film's official website: "Four misfits – Garrett 'The Garbage Man Garrison' (Jason Momoa), Henry (Sebastien Eugene Hansen), Natalie (Emma Myers), and Dawn (Danielle Brooks) – find themselves struggling with ordinary problems when they are suddenly pulled through a mysterious portal into the Overworld: a bizarre, cubic wonderland that thrives on imagination. 

"To get back home, they’ll have to master this world, and protect it from evil things like Piglins and Zombies, while embarking on a magical quest with an unexpected, expert crafter Steve (Jack Black). Together, their adventure will challenge all five to be bold and to reconnect with the qualities that make each of them uniquely creative…the very skills they need to thrive back in the real world."

A new kind of block-buster. #MinecraftMovie - only in theaters 2025. pic.twitter.com/lmeb1jvLfLSeptember 4, 2024

As the story synopsis confirms, former Aquaman star Momoa and The Super Mario Bros Movie's Bowser voice actor Black are part of its ensemble, as is Myers, who most readers will recognize from Wednesday, Netflix's hugely popular spin on The Addams Family. Just Mercy's Hansen and The Color Purple's Brooks are also part of the main cast, while The White Lotus' Jennifer Coolidge will play a supporting role throughout.

Jared Hess, who's best known for co-writing and co-directing 2004 cult classic Napoleon Dynamite, is A Minecraft Movie's lead filmmaker. Hubble Palmer and Chris Bowman (Masterminds, The Secret Saturdays) have penned its script.

A Minecraft Movie will arrive on April 4, 2025 in the US, and April 2, 2025 in the UK and Australia.

You might also like

• Netflix is teaming up with Microsoft and Mojang to craft a Minecraft animated show, and the timing couldn't be worse
• 'He's ripe to be toyed with': The Rings of Power's Tom Bombadil will 'unsettle' The Stranger when they meet in season 2, star teases
• FX and Hulu debut first teaser for Alien: Earth TV series, and it's the most underwhelming trailer I've seen in a while

Post-pandemic, we are witnessing the biggest workplace shift in generations. Some of the major driving forces behind this include changing working habits and an exponential rise in the use of AI. There is a significant amount of noise surrounding AI-driven business apps, but one that is making significant cut through is Microsoft Copilot.

Having recently shed its licensing restrictions to become accessible to all, Copilot is AI-driven assistance that is accessible, affordable and easy to implement in all workplaces worldwide. Yet many business owners and IT decision makers remain in the dark about the many benefits of Copilot and how it can help their business to perform better across the board.

We must work with AI, and not against it, to truly harness its potential and elevate business performance. If we don’t, competitors will, so it is in everyone’s best interest to adopt responsible AI practices into the workplace now in order to stay ahead of the curve. How can Copilot help your business performance?

Copilot has transformative potential to redefine the workplace, and its effects are wide-reaching and impactful.

Integration with existing software

As an established, leading software provider for business, Microsoft Office is already a trusted stalwart in many organizations, from small and medium-sized businesses to enterprises and multinational corporations. For this reason, Microsoft Copilot can be seamlessly integrated into existing Microsoft software, meaning there is minimal upheaval for a business during the rollout phase.

Microsoft Copilot spans across the suite of Office 365 products, whether it be Word, PowerPoint, Excel or Outlook. Copilot’s reach also expands into Teams, the popular office messaging application, which means colleagues can easily collaborate within their teams and with others across a business. Copilot can also run across devices, whether it be mobile apps, iPads or laptops, making it an all-encompassing solution.

Unlocking accessibility and efficiency

Microsoft Copilot’s newfound accessibility is reshaping how individuals and businesses operate, streamlining workflows and boosting productivity. So much of working life is taken over by administrative tasks, whether it be organizing meetings, reading and replying to long email threads, or wading through unnecessary emails.

For example, Copilot has the power to summarize key discussion points raised in a chat and highlight areas where people are aligned and where they disagree. After Teams meetings, takeaways are taken care of too, with action drafting and follow-up meetings now firmly under the jurisdiction of AI. With Copilot, we can shift our efforts to the tasks that really matter.

Fueling creativity

Whether it be a report for a client or an important presentation to the board, starting with a blank canvas can be daunting for everyone. But with Microsoft Copilot, engaging content is just a prompt away. Whether you need some starter ideas or a first draft, Copilot can help when inspiration just isn’t striking.

Not to mention, while Copilot takes care of the time-stealing administrative tasks, workers’ time is freed up to challenge the ordinary and be more creative. A significant burden is taken off employees’ plates and, in exchange, they have more time to explore new terrain, come up with ideas that will make meaningful changes within the business, and see them through.

Taking the next steps

For swathes of businesses, Copilot will be their key first step into the world of AI. But by having this foundation in place, businesses will be better prepared for future technologies. AI isn’t going anywhere, and it will only grow in capability. Preparing employees and business operations for AI now will only impact performance positively in the long-run.

For any business looking for guidance on how to begin their digital transformation journey, partnering with an expert may be the optimum way forward. This is often the fastest way to extract value from AI projects, as tech partners have expertise and resources on tap.

However, it is crucial when choosing a partner that they provide some sort of co-managed service, so tech teams are learning from the AI experts and gaining the knowledge to run the systems themselves. Employees must be effectively trained on Copilot to work in harmony with AI and reap all the rewards it offers. But once this upskilling has been implemented, the possibilities for improving business performance are truly endless.

What’s next for Copilot?

Copilot foreshadows the evolving role of AI, moving towards ‘invisible AI’ that seamlessly integrates into our workflows. While generative AI currently requires human activity to guide prompts, now and into the near future, we can expect to see a greater platform for invisible AI. Once enabled within a business, apps can work in the background on a wide range of different tasks. This transition hints at a future where AI proactively drives innovation and efficiency, revolutionizing how we work.

We've featured the best AI phone.

This article was produced as part of TechRadarPro's Expert Insights channel where we feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/news/submit-your-story-to-techradar-pro

Though it's a familiar setup, there's something very intriguing about the new Netflix show The Perfect Couple. We've got a wealthy family who seem to have it all, a dead body on the beach, and a whole host of shady characters, meaning almost anyone could be guilty. Though it's been done many times before, this kind of whodunnit can be a huge success, as we've seen in recent movies like Knives Out (which is available on Prime Video for those in the UK or Binge in Australia) and its sequel, Glass Onion (on Netflix).

This time, we've got a whole series to figure out just what's going on here. At face value, everything does indeed seem perfect. Amelia (Eve Hewson) is about to marry into the wealthy Nantucket Winbury family, as the long-term partner of Benji (Billy Howle). Enter Nicole Kidman's character, Benji's mother Greer, who is a successful novelist and lives a lavish lifestyle. The family is well known in the community, with the trailer hinting that they're donors to local causes. But while they hide behind a philanthropic nature, there is something going on with Greer, who disapproves of Amelia's arrival into her family.

The gripping new trailer (see below) for what could be one of the best Netflix shows claims that Greer makes visitors sign an NDA, which is not particularly welcoming at best, and very suspicious at worst. We also get a look at some of the other family members, Benji included, and honestly, it's not a family I'd want to marry into. No offense.

What else should we know about The Perfect Couple?

Like many Netflix thrillers, such as these three mystery thrillers with flawless ratings, this new six-part series is based on a bestselling novel of the same name by Elin Hilderbrand. With Greer being a novelist herself, the story plays on fun themes, including a real-life investigation that feels like it was plucked from one of the stories she wrote.

If you don't mind spoiling the ending for yourself before the series arrives on September 5, the book is widely available online and in bookstores. But if you'd rather watch the story play out in front of you, there's an excellent cast joining Nicole Kidman for this one. Dakota Fanning, who was in last week's recommended horror The Watchers, stars as Abby, the groom's sister-in-law.

Elsewhere, there's Liev Schreiber as Greer's husband Tag, Meghann Fahy as the bride's best friend Merritt, and Midsommar's Jack Reynor as the groom's brother, so even though the family may not be perfect, the cast certainly is!

You might also like

• 14 shows on Netflix, Prime Video, Max, and more I'm most excited to see in late 2024
• 5 new Netflix TV shows arriving in September with high scores on Rotten Tomatoes
• Netflix subscribers are calling The Accident the saddest show ever

With every passing day, the number of Artificial Intelligence (AI) and cloud apps used by enterprises in the manufacturing sector grows. While commendable, this move also resulted in new avenues that cybercriminals can explore.

This is according to a new report from Netskope Threat Labs, published earlier this week. It claims that cloud app usage has grown significantly, with organizations in the sector now interacting with 24 cloud apps each month, on average.

Of all the different apps, OneDrive seems to be the driving force (pun definitely intended), since its popularity grew from 43% to 58% year-on-year.

Higher risks

When it comes to AI, businesses are also inclined towards Microsoft products. This year, Microsoft Copilot is in the manufacturing top 10 apps. With recent updates to Windows 10, and the increasing adoption of Windows 11, which will grow even further next year when Windows 10 hits EOD, it’s safe to assume that Copilot’s share will only grow.

But with increasing adoption also lurk risks. Roughly a half of all global HTTP/HTTPS malware downloads originate from popular cloud apps, Netskope said, adding that the most popular apps around the world “are also among the top apps in terms of the number of malware downloads.” In fact, OneDrive is the top app being abused for malware delivery in manufacturing, with 22%, twice as much as the second- and third-placed Sharepoint and GitHub, who each have 10%.

Crooks are mostly deploying one of these five malware families: Downloader.Guloader; Infostealer.AgentTesla; Phishing.PhishingX; Trojan.Grandoreiro; and Trojan.RaspberryRobin. For Paolo Passeri, Cyber Intelligence Principal at Netskope, this is interesting, since hackers are mostly interested in flexibility:

“What really caught my eye in this report is the fact that threat actors are diversifying the kind of payload they are delivering to organizations in manufacturing,” he said. “Rather than focusing on specific categories of malware, they prefer to deliver flexible downloaders or remote access tools (GuLoader, AgentTesla, and RaspberryRobin), which can then distribute multiple kinds of payloads depending on the attackers' objectives. Businesses will need to implement strict policies that ensure the safe handling of sensitive data, and regularly monitor cloud traffic for malicious behavior.”

More from TechRadar Pro

• Cloud apps are coming under increasing attack
• Here's a list of the best firewalls around today
• These are the best endpoint security tools right now