News
A critical development in artificial intelligence is agentic AI, and these tools are becoming a part of everyday life faster than we realize. Unlike traditional AI models that follow predefined rules, agentic AI systems can make decisions, perform actions, and adapt to situations based on their programming and the data they process, often without human input.
In comparison, traditional AI models are unable to adapt to unexpected changes without a form of retraining, testing and validation, which requires human intervention. A traditional AI model will typically be built to fulfil a specific role, such as a classification task where it will identify if a person is likely to default on a loan payment or not.
From self-instructed house robots to focused data analysis, agentic AI promises convenience and efficiency. However, with that shift comes much deeper access to our personal data, which raises new concerns over transparency, trust, and control.
Agentic AI: The shift from support to self-directionAgentic AI behavior is goal-driven – it must determine how to achieve its primary goal and sub-goals, requiring it to prioritize tasks and solve problems independently of humans. For example, a house robot might be instructed to “keep the house clean”. The system will then act independently to assess different areas of the home and undertake tasks where appropriate, without requiring constant human intervention. As part of this process, the robot will identify individual sub goals, such as tidying the living room or vacuuming dirty floors, while making its own decisions to achieve the objectives.
Alternatively, businesses might employ agentic AI to analyze datasets over a specified period and identify trends and preferences, which can then be applied to other company strategies. For instance, creating targeted email campaigns. Automating the process brings instant access to valuable insights, increasing human capacity for time spent on more strategy-focused tasks.
These examples illustrate the power agentic systems have to identify opportunities, execute strategies, and adapt to different goals. But as the use of agentic AI increases, businesses and developers should carefully evaluate our trust in them and how they influence our human-machine relationships.
The evolving human-machine linkAgentic AI is fundamentally transforming the human role in tasks through automation, creating a more balanced human-to-machine relationship. As agentic systems utilize deep learning and complex image and object recognition, they can operate in increasingly dynamic environments and solve complex problems autonomously, without any human involvement.
Reduced human intervention not only offers greater efficiency in work, both at home and in businesses, but also frees up time to focus on more strategic initiatives. However, as trust builds towards these automation tools, reduced human oversight comes with a potential risk of over-reliance. While we benefit from agentic AI’s efficiency, we must also ensure we have the opportunity to upskill and educate ourselves.
Whilst agentic systems can operate independently, they still require value and goal alignment to maintain control over the outputs and ensure it aligns with the desired outcome, not just what was instructed of them. Otherwise, there is concern that these systems may take dangerous shortcuts or bypass other infrastructure to achieve efficiency.
The future of AI ethics and privacy demandsIn turn, this raises numerous ethical concerns surrounding agentic AI. One key debate is data privacy and the security of sensitive data – this can vary in severity depending on the industry in which the organization operates.
For example, agentic AI has already been deployed by cybersecurity companies to detect and correlate threats by analyzing network activity in real-time and then autonomously responding to potential breaches.
However, organizations who implement this must provide data to the system, raising questions over the security and privacy of their information. Without human oversight, organizations must consider if they are comfortable with agentic AI making business-altering judgement and being on the frontier of their most valuable assets.
Bias in agentic AI can occur because of human input and data. When such a system is entrusted with making moral decisions that have real-world consequences, it faces significant ethical considerations. While existing or emerging legislation provides guidance, there is still much to be done to unpack and implement these concepts within operationalized AI systems fully.
Furthermore, systems with access to highly sensitive data have raised security concerns as they can have vulnerabilities exploited. This has been highlighted by recent cyber attacks, and information within digital environments is at risk. In such complex systems, who is held accountable if things go wrong?
Key considerations for ethical AI useWhile we are guided by ethical principles and emerging legislation, it is essential that we have safeguards in place for both agentic and traditional AI systems. When automating manual tasks and analyzing datasets, it is crucial to identify and mitigate bias in both data and algorithms with consistent human oversight. Organizations that utilize agentic AI should strive for ethical practice which can be supported through training and continuous auditing. This helps ensure fairness and prevents harm while creating transparency around how automated decisions are made.
We've featured the best private browser.
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
New and sophisticated threat actors known collectively as "Scattered Spider" have emerged as one of the most concerning challenges facing organizations worldwide. These groups – including UNC3944, Oktapus, and Muddled Libra – have, yet again, fundamentally transformed the attack paradigm by prioritizing human psychology over traditional malware deployment.
A new breed of cyber adversaryWhat sets Scattered Spider groups apart from conventional cybercriminals is their masterful exploitation of human vulnerabilities rather than system weaknesses. These threat actors have perfected the art of social engineering, employing sophisticated techniques such as voice phishing (vishing), SMS phishing (smishing), and chat-based manipulation to convincingly impersonate legitimate employees and trusted contacts.
The success of these groups lies in their ability to establish initial network access through carefully orchestrated human interactions. By studying organizational structures and employee behaviors, they craft highly personalized approaches that bypass traditional security measures through the simple act of conversation. This ‘human-first’ methodology is proving remarkably effective against even well-defended networks.
The mechanics of deceptionThe true danger of Scattered Spider operations becomes apparent when examining their sophisticated approach to multi-factor authentication (MFA) bypass. Through strategic SIM-swapping attacks and carefully orchestrated helpdesk manipulation, these criminals exploit the inherent trust relationships within organizations. They convince IT support personnel to reset MFA credentials or transfer system control, often using information gathered through extensive reconnaissance to appear legitimate.
Once inside target networks, Scattered Spider groups demonstrate exceptional operational security by employing "Living off the Land" tactics. This approach involves leveraging legitimate administrative tools already present within most enterprise environments – including PowerShell, PsExec, AnyDesk, and even credential-harvesting utilities like Mimikatz. By utilizing these ‘authorized’ tools for malicious purposes, the groups effectively operate beneath the detection threshold of traditional security solutions.
This minimal malware approach presents a significant challenge for organizations relying heavily on signature-based detection systems. Traditional antivirus solutions and legacy security tools may never register suspicious activity, allowing attackers to maintain persistent access while conducting their operations with impunity.
The attack unveiledThe Scattered Spider groups’ attack methodology follows a predictable, yet highly effective pattern. The process typically begins with extensive reconnaissance and credential harvesting, where attackers gather detailed information about target organizations, their employees, and internal processes.
Following this intelligence-gathering phase, the group leverages AI-driven phishing techniques and sophisticated voice cloning technology to craft highly personalized smishing campaigns. These messages appear remarkably authentic, often incorporating specific organizational terminology, recent company events, or industry-relevant information scraped from LinkedIn and dark web data, which lends credibility to their deceptive communications.
The social engineering component represents perhaps the most insidious aspect of their operations. By impersonating legitimate employees, these criminals manipulate helpdesk staff and telephony personnel into performing security-compromising actions. They skillfully exploit the helpful nature of IT support teams, often creating artificial urgency or leveraging apparent authority to bypass standard verification procedures.
Once initial access is established, Scattered Spider groups focus on persistence through strategic deployment of remote management tools, using credential dump tools such as Mimikatz, secretdump, and ADRecon to escalate access. They also utilize internal tools and tunneling such as Ngrok, RSocx to hide movement, ultimately leading to data exfiltration or ransomware deployment, depending on the group's specific objectives for each engagement.
The warning signsSecurity teams must develop heightened awareness of specific indicators that may signal Scattered Spider activity within their environments. Sudden increases in remote-access sessions, particularly those involving tools like AnyDesk or ScreenConnect, warrant immediate investigation, especially when these sessions occur outside normal business hours or originate from unfamiliar geographic locations.
Multiple MFA reset requests channeled through phone-based support within compressed timeframes often indicate systematic attempts to manipulate helpdesk personnel. Organizations should also monitor for unrecognized SIM-swap alerts or unexpected one-time password deliveries to employees, as these frequently signal active credential compromise attempts.
Perhaps most concerning are instances where security tools experience mysterious disabling or unauthorized administrative actions appearing in system consoles. These events, particularly when coupled with anomalous lateral movement patterns involving valid employee credentials used in atypical ways, strongly suggest advanced persistent threat activity.
Industry impact and future implicationsWhile recent intelligence suggests Scattered Spider groups have begun targeting the aviation industry following successful retail sector exploits, security professionals must recognize that no industry remains immune to these sophisticated attacks. These groups’ methodology proves effective across diverse organizational structures and business models, making widespread vigilance essential.
Building defense strategiesProtecting against Scattered Spider groups requires a carefully orchestrated multi-layered approach that simultaneously addresses technical vulnerabilities and human factors. Email security solutions are needed to provide crucial frontline defense by intercepting phishing attempts, credential-harvesting campaigns, and smishing attacks before they reach employees. Advanced anti-impersonation features within these solutions help reduce CEO and CFO fraud schemes commonly leveraged in these sophisticated campaigns.
Modern endpoint security and endpoint detection and response solutions are the next layer of defense required for essential visibility, as they detect unauthorized remote access tools, identify credential dumping attempts, and flag unusual lateral movement patterns from compromised endpoints. Additionally, advanced behavioral analytics can identify "Living off the Land" techniques even when traditional malware signatures are absent, catching attackers who rely on legitimate system tools for malicious purposes.
However, technology alone cannot adequately address this rapidly evolving threat. These sophisticated attacks underscore a fundamental shift in cybersecurity: humans, not machines, have become the new perimeter that organisations must defend. Therefore, security awareness training serves as a critical human-layer defense, empowering staff to recognize deception tactics early in the attack chain.
The emergence of Scattered Spider groups represents more than just another type of cybercriminal outfit – it signals a paradigm shift requiring organizations to layer traditional technical controls with identity safeguards, behavior-based detection systems, and continuous employee education to create truly comprehensive protection against human-centric cyber threats.
We've featured the best identity theft protection.
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
When we sit down to watch our favorite TV series online, we rarely consider the streaming process – we just hit play. In reality, there’s a video codec seamlessly managing the compression and decompression of the data to deliver that effortless, high-quality viewing experience. For example, High Efficiency Video Coding (HEVC or H.265) has been one of the more widely used video standards across streaming platforms and consumer devices for the last decade.
In more recent years, Versatile Video Coding (VVC or H.266), the next generation codec launched, offering significantly improved video compression efficiency and performance over HEVC. The benefits of using VVC include streaming higher resolution content at lower bitrates, improved data efficiency and better visual quality.
However, while VVC delivers a superior performance to HEVC, widespread adoption of new codecs often takes time due to standardization process timelines, compatibility conditions and hardware requirements.
VVC is gaining traction, particularly around the development of software decoders but hardware support is still playing catch up. Several steps must now happen for VVC to achieve full industry endorsement.
The benefits of VVC and catalysts to adoption:As viewer expectations for quality increase and streaming platforms continue to develop, VVC is well positioned to meet future video compression needs. The codec has a robust technical architecture and clearly established intellectual property policies. It also delivers an impressive 50% bitrate reduction compared to HEVC, for the same perceived quality. The result? Reduced buffering and improved delivery of 4K and 8K content, without high data costs.
VVC can be used across many applications rather than one specific use case driving adoption. Technologies that would benefit most from VVC, such as 8K, VR (Virtual Reality), and volumetric video are growing in use cases, but are not yet widely adopted by consumers and businesses. While they definitely show promise, they haven't yet reached the scale of demand needed to drive immediate uptake of VVC.
The area showing most promise so far in driving VVC adoption is broadcast. VVC's superior compression efficiency makes it particularly valuable here, as bandwidth limitations mean that improved compression directly impacts cost efficiency.
In the broadcast sector, there's growing momentum from both European (DVB) and North American (ATSC) standards bodies supporting VVC, with Brazil launching experimental VVC-based transmissions in May, in advance of a subsequent commercial rollout. ATSC has recently incorporated VVC into its 3.0 specification.
Removing VVC adoption roadblocks:Despite clearly demonstrating the capabilities of VVC, one obstacle to implementation is that firmware in some devices – such as certain Smart TVs - will require updating to enable it, despite already having the necessary hardware. This can create market fragmentation where some devices are provided with updates while others remain without VVC support.
In terms of mobile platforms, VVC has been proven to run efficiently on older mobile devices in software, easing concerns about its computational demands. However, this is an area where adoption is still slightly behind.
That said, creating a viable option to enable VVC via firmware updates in consumer electronics presents an opportunity to accelerate rollout without requiring hardware refreshes. But integration in devices must be accompanied by the availability of VVC content and services.
Overcoming regional differences:Some markets have already integrated HEVC into their landscapes, meaning a new migration could take time and further investment. Yet regional discrepancies can also work as an advantage for VVC. For instance, broadcast markets such as Australia haven't yet completely migrated to Advanced Video Coding (AVC) - the codec that came before HEVC. This might mean that they bypass HEVC entirely and move directly to VVC, potentially speeding up uptake in these specific areas.
VVC also faces competition in the video codec landscape, particularly from AV1 (AOMedia Video 1). Developed by the Alliance for Open Media, AV1 has achieved significant adoption in streaming platforms and web-based applications, and the next generation codec, AV2, is expected to be finalized soon. However, AV1's performance characteristics make it less suited for complex content, meaning that, in terms of coding efficiency, AV1 and VVC aren't direct competitors in high-end applications.
The future for VVC development looks brightWe’ve recently seen VVC being demonstrated publicly. Most notably, the technology was used to power 8K trials at the Paris Olympics, with a private demonstration delivered to premier broadcasters. The practicality of VVC for streamers has been enhanced by Fraunhofer HHI’s release of the VVdeC/VVdeC software decoder (Versatile Video Encoder/ Decoder) in 2020. This software has been ported to various platforms and web browsers, so that developers wishing to integrate VVC don’t necessarily need to start from scratch.
Real-world deployment of VVC is still in its early stages. However, continued market education, industry collaboration, and integration into a variety of devices and applications will help drive implementation throughout 2025 and beyond. As a doorway to widespread VVC adoption, major streaming platforms need to begin offering VVC streams.
Even one major service adopting VVC could create a significant domino effect, where competitors follow suit and create momentum across the entire ecosystem. So the next time you sit down to stream your favorite series, you may just be pioneering the very latest video codec, in full 4k or even 8k glory.
VVC’s unique position at the intersection of broadcast, streaming, and emerging immersive media makes it a pivotal technology in the evolving digital video landscape. As content delivery networks face increasing bandwidth demands, VVC's compression advantages could become increasingly valuable, potentially reshaping how video is distributed across global networks. For businesses in these industries, it’s certainly a technology to monitor.
We've featured the best business monitor.
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
- Windows 10's extended support offer is currently rolling out
- A fair few people aren't seeing it, though - myself included
- Microsoft has said that the rollout is happening 'slowly' but that it will be available for everyone before Windows 10's End of Life in October 2025
If you're running Windows 10 and hoping to get extended support for the OS in order to avoid having to upgrade to Windows 11 come October 2025, you may be wondering exactly where the offer of that support is.
While it's officially rolling out, as of a month ago in limited form, Windows Latest reports that the extended security updates (ESU) scheme hasn't reached many people yet.
The button to kick off the enrolment wizard that allows you to sign up for the scheme - which can be done for free, as you may have seen - should appear in the Windows Update panel. However, a good few people - including Windows Latest readers - just aren't seeing this button at all, so they can't sign up.
The tech site contacted Microsoft about this and was told by the company that the enrolment wizard is indeed rolling out currently - as was officially announced recently - it's just that this is happening "slowly".
There's nothing you can do to prompt the button to appear in Windows Update; you just have to wait.
(Image credit: Shutterstock - Wachiwit)Analysis: a seemingly cautious rolloutI am still running Windows 10, and I'm planning to take the offer of extended support (for reasons I've discussed in-depth recently), but I also haven't yet seen it on my PC. There are scattered reports on various online forums from people wondering where the offer is, so it does seem like the rollout is on the cautious and slow side, as Microsoft indicates.
The good news is that Microsoft has assured Windows Latest that the enrollment wizard will arrive on all Windows 10 PCs before the End of Life of the older operating system, which is mid-October 2025. So, it's just a case of being patient, although personally I'd rather not leave signing up for the program until the last minute.
Hopefully, Microsoft will kick the ESU rollout into a higher gear with the next big update for Windows 10, which arrives on September 9, 2025. We shall see, and perhaps ironing out bugs in the wizard has been part of the issue here - there was a notable glitch whereby the sign-up process was failing for some folks, a problem Microsoft resolved the week before last.
Meanwhile, as Windows Latest points out elsewhere, Microsoft is stepping up its full-screen nags being displayed to Windows 10 users, urging them to upgrade to Windows 11, as of the most recent August update. These are pop-ups that have been seen before, and it's likely they'll be fired out again next month and when October rolls around, when support is cut off for those who haven't signed up for the ESU.
I'm going to avail myself of the free option for extended updates, which requires syncing PC settings to OneDrive, as this is something I already do anyway. For those who'd rather not do so - and remember, it is just your Windows settings, not your files and data that's synced - they can still pay $30 (or equivalent in their currency) to join the ESU (which was the only option initially).
A third route is to cash in 1,000 Microsoft Rewards points, provided you have accumulated that many.
You might also likeI adore my Nintendo Switch 2, but as many predicted, it faces a familiar challenge: battery life. It's by no means dreadful, but I definitely find it runs out a little faster compared to my Switch OLED (and the original Switch before that).
Of course, it does feature a bigger battery than its predecessors, but that's mostly to account for improved performance. Still, I've not second-guessed my generational upgrade once.
However, since I travel often and frequently forget to charge up when there's the opportunity, even a slightly diminished handheld battery life can prove irritating. There are heaps of Switch 2 cases on sites like Amazon, and an ample supply of the best power banks on offer too, but given the Switch 2's innate portability, I'm loath to pack my bag with too many loose gadgets.
That's where Belkin's Charging Case for Nintendo Switch 2 comes in. This bad boy packs a 10,000mAh battery pack complete with a display, and also comes with its own USB-C charger. I snagged one of the sage cases myself, and it's washed away my battery anxiety overnight – but there's a lot else to love.
The case is also available on Amazon in the UK here for £49.99, or from at Belkin Australia for AU$99.95.View Deal
First up, let's talk about protective measures. Now, it could be the slightly larger frame in contrast to its ever-thin profile (or the loftier price tag...), but I've been nursing my Nintendo Switch 2 like a newborn babe for fear of it being damaged, especially when I'm travelling.
Available in three chic colors (black, sand, and sage), Belkin's charging case is well-padded and made with a durable polyester outer shell and a wraparound zip. I adore the trendy sage colorway, which is slightly more green than grey and matches my largely matcha-influenced aesthetic for 2025.
Open up the case and you'll find a soft velvety interior as well as some built-in storage; namely, a mesh pocket for wires and other accessories (including a fitted Apple AirTag pocket), and a flat panel with 12 elasticated game card holders. On the other half, there's a recess to house the battery with a velcro strap to keep it secure, and the Switch 2 slots in neatly above this in a fitted bed to keep it snug.
Image 1 of 3(Image credit: Future)Image 2 of 3(Image credit: Future)Image 3 of 3(Image credit: Future)With the Switch 2 in tow, it weighs in at 2.6 lbs / 1.84kg; a little meaty, yes, but far more compact than the alternative loose portable battery. It's backed by a two-year warranty, too, so you're covered on multiple fronts.
All this contributes to far greater peace of mind for me when I'm out and about with my Switch 2, whether that's through keeping me charged and ready to play, protecting my case from damage, or offering peace of mind that, should it get lost, my Apple AirTag can easily help me relocate it.
You can grab Belkin's charge case for $69.99 / £49.99 / AU$99.95, but there's also a non-charging version available for $29.99 / £17.99 / AU$39.95. If you've not yet bought Nintendo's latest handheld, you can read TechRadar’s full review of the Nintendo Switch 2 here.
You might also likeIf you’re buying a flashy new pair of over-ear headphones, what are you looking for? Amazing sound quality, top-tier noise cancellation, and high levels of comfort are surely all crucial. But something I’m also keen to assess here at TechRadar is looks – how clean do a pair of headphones appear on the ear? What sets them apart from competitors?
Yep, if like me, you wear over-ear headphones all the time, even when taking public transport and out and about, I’m sure you’ll want them to look as good as they sound. So, I’ve compiled a list of 5 headphones that I and the wider TechRadar team have tested, all of which really are stunners.
And although this isn’t a list of the best headphones overall, I’ve made sure that each model listed supplies a strong audio performance, comes with a nice array of features, and feels pleasant in use. Interested? Then let’s get into the list.
1. The budget beauty: 1More Sonoflow Pro HQ51(Image credit: Future)The 1More Sonoflow Pro HQ51 are our pick for the best budget cans that money can buy, and their aesthetic plays a significant part in that.
Like a lot of 1More cans, they have an extremely sleek curved on-ear look, giving them a sharp, well-defined appearance. They combine that with an attractive matte black finish, red detailing on the ports and earcups, and a slim headband. Bring that all together, and you’re looking at a really attractive pair of over-ears.
They also offer surprisingly effective ANC, a highly customizable sound, and outrageously long battery life – 65 hours with ANC on and 100 hours with it off! All of that for less than $100 / £100? Oh, go on then.
Read our full 1More Sonoflow Pro HQ51 review.
Not really your thing? Then how about the awesome value Sony WH-CH720N, which appear higher in quality than their low price may suggest.
2. The magnificent mid-ranger: Sennheiser Momentum 4 Wireless(Image credit: TechRadar)Check this out(Image credit: Future)Want to know what tracks I use to test headphones here at TechRadar? These are the 7 failsafe tracks I use when reviewing audio gear
Right, a quick disclaimer, the Sennheiser Momentum 4 Wireless are mid-rangers in my book for UK buyers – they’re regularly available for well under £200. In the US, they’re a fair bit pricier, but these are simply too suave for me to omit from the list.
These five star-rated headphones combine large oval-shaped earcups, a unique fabric headband, and a far smoother appearance than their predecessors, the Sennheiser Momentum 3 Wireless. I specifically really like the variant that has copper-colored detailing around the outside of the earcups as well as the Sennheiser logo – it really makes this model pop.
Of course, you know that’s not all. They’ve got a pleasantly neutral, crisp sound signature, awesome battery life, and better noise cancelling than you might expect. Talk about the full package.
Read our full Sennheiser Momentum 4 Wireless review.
Want something a little different? Then how about the comfort kings that are the Anker Soundcore Space One Pro
3. The majestic mid-to-premium model: Marshall Monitor III ANC(Image credit: Future)Even though the Sony WH-1000XM6 are the pair of headphones I use daily, they’re not my favorite headphones you can buy looks-wise. That award would go to the Marshall Monitor III ANC, which are absolutely stunning, in my opinion.
They have a faux-leather exterior, visible wires to represent the brand’s rock ‘n’ roll roots, and gorgeous golden details. I see a lot of stylish people donning the Marshall Major IV on-ears, and I don’t mind those, but I prefer the slimmer form of the Monitor III’s.
My colleague at Future, Nikita, wears these cans everyday, so I wanted to get her opinion. She said: “There's a very good reason why Green Day's Billie Joe Armstrong is the posterboy for these headphones, and that's because they make me look and feel like a rockstar. They look like a million bucks and I love how sleek and compact they are as well. Based on looks alone, the Monitor III’s are the best-looking headphones I've owned”.
Pretty high praise, right? But there’s more! I adore the red-velvet-lined carry case, immersive sound, and seemingly endless playtime these headphones offer. They’re a true standout option in today’s market, but for all the right reasons.
Read our full Marshall Monitor III ANC review.
They’re divisive, but if you want something truly eye-catching, the Nothing Headphone (1) is an S-tier option.
4. The pretty and premium pick: Bowers & Wilkins Px7 S3(Image credit: Future / Simon Lucas)OK, we’re getting into expensive territory now, but with headphones like the Bowers & Wilkins Px7 S3, you’re getting what you pay for. I could speak for hours on end about their spacey, hypnotic sound signature, but it's looks we’re talking about here, and these have them in abundance.
The earcups are slim and well built, there’s a satisfying feeling fabric headband, and stellar color options – there’s Anthracite Black, Canvas White, or Indigo Blue. If you want cans that indicate you value sound highly, these should be a go-to.
We love the audio quality, evolving feature-set, and enhanced ANC on the Px7 S3, so if you’ve got around $450 / £400 to spend, you can’t really do much better than this.
Read our full Bowers & Wilkins Px7 S3 review.
If you’re not sold on the Bowers, I also personally like the Sonos Ace – they’re amazing for movies, very solid for music, and super stylish.
5. The luxurious looker: Dali IO-8(Image credit: Future / Simon Lucas)At the time of writing, our guide to the best over-ear headphones places the Dali IO-8 in first place for design. Yes, we’ve described them as “the best over-ear headphones for ultra-luxe looks as well as sound”.
These cans have exceptional build quality, using brushed, anodised metal on the exterior, memory foam lining the inside of the earcups, and a tasty choice of ‘iron’ black or ‘caramel’ white finishes. Their perfectly circular earcups also have a minimal, yet premium look that’s timeless.
One issue some may take with the IO-8 is that there’s leather on them. There’s no vegan alternative either, unfortunately, but the aforementioned Marshall Monitor III ANC are a nice vegan-friendly alternative.
The Dali IO-8 aren’t only beautiful, they’re also able to offer positive, detailed audio, with support for the hi-res aptX Adaptive codec also on the menu. They appear to be somewhat elusive and highly expensive in the US, but UK buyers should be able to scoop these up for less than £500.
Read our full Dali IO-8 review.
Apple fans, if you don’t own them already, the Apple AirPods Max are a seriously sexy pair of cans; they’re my alternate option in this bracket, no questions asked.
So, those are my personal picks, but I’d love to hear from you. Do you have any of these headphones? What do you make of them? Or are there some other models that you really like the look of? Let me know in the comments.
You may also like- The best noise cancelling headphones for every budget
- The best cheap headphones, all tested by our reviewers
- Want something a little more lightweight? Then head on over to our guide to the best wireless earbuds
- Google released a study outlining how much energy and water its AI uses
- Experts say the figures lack important context
- Google insists these figures represent the average user experience
A new study from Google claims its Gemini AI model only uses very minimal water and energy for each prompt - with the median usage sitting at around 5 drops (0.26 milliliters) - the equivalent electricity used for 9 seconds of TV watching (roughly 0.24 watt-hours), resulting in around 0.003 grams of CO2 emissions.
Experts have been quick to dispute the claims, however, with The Verge claiming Google omitted key data points in its study, drastically under-reporting the environmental impacts of the model.
Whilst models and data centers have become more efficient, it seems there’s more to the story than Google is letting on.
The tip of the icebergOne of the authors of a paper cited in the study, Shaolei Ren, associate professor of electrical engineering at the University of California told the publication; “They’re just hiding the critical information. This really spreads the wrong message to the world.”
AI models like Gemini are supported by data centers - huge warehouses full of servers which consume intense amounts of water and energy, straining local resources.
Governments across the globe have been sanctioning the building of these data centers, despite the destruction they could bring to local countryside - and consumers are likely to be the ones paying for the extra energy used.
One of the biggest concerns with Google's study is that it omits indirect water usage in the estimates, which form the majority of the use related to AI. Whilst the figures are technically correct, the missing context of the extreme energy use paints a misleading picture.
The study only looks at the water used by data centers to cool their servers, but left out is the electricity these data centers demand, which in turn leads to new gas and nuclear plants - which also cool their systems with water, or use steam to turn turbines.
Water isn’t the only metric Google misrepresented though, with the paper only outlining a ‘market based’ carbon emissions measure, which offsets the figure using Google’s promises to use renewable energy to support power grids. Savannah Goodman, Head of Advanced Energy Labs told TechRadar Pro,
"We hope to share environmental metrics that are representative of a typical user's behavior, and reasonably comparable over time. However, with the rapidly evolving landscape of AI model architectures and AI assistant user behavior, there are outliers either from small subsets of prompts served by models with low utilization or with high token counts."
"In order to share metrics that represent a typical user’s experience and are robust to this rapidly evolving field, we chose to measure metrics for the median prompt — which is robust to extreme values and provides a more accurate reflection of a typical prompt's energy impact."
You might also like- Adjustable channels for 3D immersive audio
- Hi-res audio via TIDAL Connect and Google Cast
- New dialog enhancement mode
Sennheiser's flagship soundbar, the Ambeo Max, is getting a major firmware update called AMBEO OS2 that should deliver a serious audio upgrade to what's already a very impressive soundbar.
This is the most important upgrade since the Sennheiser Ambeo Max launched back in 2019, and it keeps the soundbar current with more recent high-end examples of the best soundbars from the likes of Sony and KEF.
There are three key components to the upgrade. There's customizable 3D audio; extended hi-res lossless audio performance; and a voice enhancement mode to deliver better vocal clarity in movies, TV shows and sports.
(Image credit: Sennheiser)Sennheiser Ambeo OS2: what's new and how to get itOS2 delivers adjustable volume levels for the side, top and center channel speaker virtualization, enabling you to tweak the 3D audio to suit your room and your own personal preference.
The existing Dolby Atmos Music on TIDAL has been supplemented with support for audio up to 24-bit/192kHz via TIDAL Connect. And you can now stream at up to 24-bit/96kHz via Google Cast for Audio.
Last but not least, there's a new Voice Enhancement Mode that Sennheiser says "ensures dialog remains clear and intelligible even when sound effects or music dominate the mix". I wonder how that'll cope with Christopher Nolan's Interstellar.
Ambeo Max owners can get the new upgrade in two ways: via the Smart Control app on iOS or Android, or by downloading it to a USB drive from sennheiser-hearing.com.
This is obviously good news for Ambeo Max owners. But I think it's worth noting for prospective Sennheiser soundbar purchasers too, since a company delivering an update this significant after five and a bit years is a pretty impressive commitment to its customers.
You might also like- European firms are deeply locked into foreign office suites and systems
- American platforms manage the communication backbones of Europe’s largest corporations
- Reliance on external providers exposes utilities and healthcare to foreign oversight
For years, European governments and corporations leaned heavily on American technology offerings instead of nurturing local alternatives.
That choice now carries visible consequences, as sanctions and shifting trade rules brought in by the Trump administration drastically reshape the balance of power.
A recent analysis of business email domains across Europe by Proton shows a striking majority of publicly listed firms rely on American providers such as Google and Microsoft.
Data reveals the depth of relianceBehind the rhetoric of digital sovereignty, the reality is that much of Europe’s digital infrastructure rests on technology stacks that entities outside its borders control. This is not just about convenience software but also about essential systems that underpin finance, healthcare, and utilities.
Email may appear mundane, but it often serves as the gateway to office software, online collaboration platforms, and cloud-based storage.
When a company commits to a provider for email, it usually adopts the full suite, embedding foreign technology deep into its operations.
This trend is not limited to smaller economies but also includes the continent’s largest players, where dependence cuts across industries from energy and telecommunications to pharmaceuticals.
In countries like Iceland, Norway, Finland, and Sweden, over 90% of publicly listed companies rely on American services for email and related infrastructure.
However, the shocker is probably Ireland, which is at loggerheads with the US on several policies, but 93% of its businesses depend on American tech.
The UK, although mostly an ally of the US, has an alarming 88% of businesses relying on US tech, while other European heavyweights like Spain, Portugal, and Switzerland recorded 74%, 72%, and 68% of businesses relying on US tech, respectively.
Even France, which often champions its own autonomy, sees two out of three (66%) companies tied to US providers.
Eastern European countries like Bulgaria (16%) and Romania (39%) are the least dependent on American tech, and Russia is not even on the list of nations dependent on the US.
National security concerns emerge when utilities, transport systems, and healthcare facilities communicate through networks governed by foreign jurisdictions, but perhaps not when the network belongs to the US.
The reliance stretches far beyond convenience; it embeds itself in the very systems Europeans use every day - dependence on foreign technology does not just present a financial vulnerability; it raises questions about surveillance, geopolitical leverage, and the future of innovation.
AI training programs outside Europe’s control can sweep in sensitive business data, while reliance on external platforms exposes companies to warrantless legal demands.
This arrangement has also fostered a talent and capital drain, as engineers and investors direct their focus toward Silicon Valley rather than strengthening European ecosystems, whether through proprietary services or alternative Linux distros.
Some argue that American technology simply offers the best tools available, which may be true in terms of efficiency and global reach, yet the consequences of reliance are increasingly hard to ignore, since the US can turn off the switch at any time, and thousands of companies will be in crisis.
The fact that so many European firms cannot operate without American software demonstrates the fragile nature of Europe’s autonomy.
Rather than securing independence, Europe risks locking itself further into external dependencies at a moment when political winds in Washington are shifting.
You might also like- Take a look at our roundup of the best business VPNs available now
- These are the best cloud document storage solutions to choose from
- Asus released new laptops with a 5-year BIOS update warranty and 1-year McAfee+ premium subs