Technology

• Defense Secretary Pete Hegseth has confirmed $5.1 billion in IT savings
• They will be realized by terminating contracts with Accenture and others
• DOGE has now enabled the DOD to save nearly $6 billion, Hegseth says`

$5.1 billion in IT contracts have been canceled by the Department of Defense (DOD) under the guidance of Elon Musk’s Department of Government Efficiency (DOGE), new evidence has shown.

Defense Secretary Pete Hegseth signed a memorandum detailing the cancellations, which mostly consisted of “consulting and other non-essential services.”

As Musk’s unit suggests, the DOD also claims to have enhanced efficiency by removing unnecessary steps and duplication, allowing it to allocate cash to other core areas of the agency’s scope, such as “better health care for our warfighters and their families.”

DOD terminates $5.1 billion in IT contracts

The terminated contracts include “$1.8 billion in consulting contracts the Defense Health Agency awarded to various private sector firms, a $1.4 billion enterprise cloud IT services contract awarded to a software reseller, and a $500 million Navy contract for business process consulting.”

Affected companies include Accenture, Deloitte, and Booz Allen.

Hegseth also noted a $500 million contract awarded to the Defense Advanced Research Projects Agency for IT help desk services was "completely duplicative" of services the Defense Information Systems Agency already provides.

The DOD also terminated 11 contracts related to diversity, equity and inclusion, climate change, and COVID-19.

Although the savings only account for a touch over half a percent of the DOD’s $883.7 billion budget, Musk’s goal to iron out unnecessary spending is at least commendable.

However critics have argued there are more factors to consider than simply how much a contract is costing, urging the government to agree on contracts that meet its requirements with more efficiency and to implement performance metrics.

More broadly for the DOD, this latest announcement follows from the recently confirmed $580 million cut to certain programs, contracts and grants, bringing the running total to nearly $6 billion in savings enabled by the DOGE.

You might also like

• Cut the costs with the best free office software
• Google Workspace is offering huge discounts for the US government
• We’ve rounded up a list of the best productivity tools and best online collaboration apps

The evolution of artificial intelligence (AI) continues to push boundaries, captivating the collective imagination of industries and societies alike. Today, the conversation is expanding beyond generative AI (Gen AI) into agentic AI and artificial general intelligence (AGI).

These advancements promise significant yet complex shifts in business dynamics, organizational structures, innovation processes, and workforce management. As we stand at the threshold of these transformative changes, our focus must remain clear: responsible and human-centered progress.

Reimagining organizational structures

The implications of agentic AI and AGI for organizations are profound. The rise of AI agents will disrupt hierarchical workflows, paving the way for more dynamic and decentralized operational models. In this new paradigm, Chief Knowledge Officers (CKOs) and other roles central to knowledge engineering will become pivotal. These roles ensure that knowledge, not just data, is used effectively, aligning the capabilities of intelligent systems with business strategy.

EY teams have embraced this shift, leveraging AI as a catalyst for redefining how work is structured and executed. The aim has been to develop a flexible, multi-model approach that exemplifies AI’s potential to create positive outcomes when adapted to a variety of business functions. This is not a one-size-fits-all journey; it is an ongoing process of refining options through real-world applications, where EY is often ‘Client Zero’, and learnings from our internal use cases help set the standard for client transformations.

The transformational promise of AGI

Unlike task-specific AI, AGI aspires to surpass human cognitive abilities across a variety of functions. It represents a leap from automated assistance to deep, strategic collaboration. While today’s AI can parse complex data sets and generate creative solutions, AGI could operate as a strategic partner, driving decisions and sparking innovation in ways that exceed the human thought process.

This advancement will challenge traditional enterprise models, shifting from workflow-driven operations to goal-oriented strategies. The essence of business will no longer be confined to what processes can achieve but will be reframed around outcomes that align with strategic objectives. Such transformation requires robust ethical frameworks and adaptive governance that keep human oversight at its core.

Innovation and workforce adaptation

The workforce of the future will need to evolve alongside these technological advances. We have already seen a shift from traditional data science roles to knowledge engineers, AI governance experts, and AI ethics specialists. These roles are becoming essential as AI evolves from a support tool to an integral component of strategic business planning.

Upskilling and reskilling initiatives are more important than ever. AI’s integration into the workplace should not be seen in isolation but as a catalyst for broader workforce transformation. The shift to objective-driven technology requires that businesses equip their employees with new capabilities and build their confidence. At EY, research and experience have shown that embedding human-centric support systems and properly prioritizing human factors can significantly raise the success rate of transformation programs.

Gen AI has already started democratizing skillsets, enabling more individuals to access roles that were previously restricted to highly specialized workers. For instance, AI can simplify complex programming tasks, opening pathways for mid-skilled professionals to contribute at a higher level. This democratization fosters inclusivity and helps bridge the productivity gap across various sectors.

Aligning AGI with societal and ethical values

One fundamental principle I champion at EY is that AI must enhance human capabilities. This belief extends to AGI. While the promise of AGI brings excitement, it also raises valid concerns around confidence, transparency, and control. The lessons learned from current AI deployments underline the importance of robust ethical frameworks, regulatory compliance, and proactive dialogue among industry, governments, and technology developers.

The EU AI Act exemplifies the kind of forward-thinking regulation needed to ensure these technologies develop in a way that aligns with societal values. At EY, a commitment to aligning AI deployment with stringent ethical and operational standards has set a precedent for responsible AI use. As AGI approaches, maintaining these guardrails becomes even more critical. AI must operate with transparency and accountability, building and reinforcing the confidence human beings feel toward the technology we use.

Preparing for AGI’s disruptive potential

If we manage AGI’s development responsibly, it has the potential to revolutionize the enterprise, not as a mere tool but as a collaborative partner. The strategic potential of AGI lies in its capability to autonomously achieve business outcomes, shifting the enterprise focus from process optimization to holistic, outcome-driven strategies. However, this transformation will only be as successful as our commitment to embedding human-centric, ethical practices into the core of AGI’s deployment.

I believe AGI’s introduction will also drive the creation of new roles, including those focused on AI governance, policy, and ethics. The current wave of regulatory discussions is a positive indicator that we are moving toward a global standard for safe and effective AI governance. This evolution, coupled with the emergence of specialized roles and innovative approaches, will help organizations navigate the complexities of integrating AGI while maintaining ethical and operational integrity.

The future is collaborative

In the grand scheme of technological evolution, we are on the brink of an era that promises to redefine industries and reshape society. Agentic AI and AGI will play transformative roles, and their impact on businesses and the workforce will be monumental. Yet, the focus must always return to ensuring that these technologies serve to elevate human potential and align with our shared values.

Our guiding principle is surely clear: AI must be human-centered, and this philosophy will be the cornerstone as we step into the AGI future. This approach is not just ethically sound—it’s essential for building confidence, driving innovation, and ensuring that AI’s profound capabilities are harnessed to benefit businesses and societies alike.

The views reflected in this article are the views of the author and do not necessarily reflect the views of the global EY organization or its member firms.

We've featured the best online collaboration tool.

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

Cyberattacks on public institutions are becoming more frequent and sophisticated, and with our reliance on technology, the stakes have never been higher. According to the UK Government’s Cyber Security Breaches Survey 2024, half of businesses (50%) and around a third of charities (32%) experienced cyber breaches or attacks last year. Given the scale of these attacks, it is no surprise that government bodies, which hold vast amounts of highly sensitive data, are also attractive targets.

Cybercriminals can exploit vulnerabilities in public sector IT systems and processes to target them with phishing scams, distributed denial of service (DDoS) attacks, ransomware and more. Sensitive data including personal records, financial details, and classified intelligence are valuable to cybercriminals, making public institutions prime targets for data theft and ransomware.

In addition to this, geopolitical tensions mean that attacks aimed at disrupting the availability of critical systems are also commonplace. For instance, the attack on the British Library in October 2023, which impacted digital services for months and had an estimated direct cost of £600,000 is just one example of how damaging these incidents can be.

The true cost of cyberattacks: more than just money

The financial impact of cyberattacks on government bodies can be staggering. In 2023, UK organizations spent an average of £1,100 per cyberattack, but for public institutions, the costs can be far greater. The 2025 NAO report on Government Cyber Resilience highlights how these attacks not only disrupt essential services but also lead to huge multifaceted costs including initial response, regulatory fines, recovery, litigation, and in some cases, ransom payments.

The financial impact isn’t the only problem. Cyberattacks also erode public trust which can prove to also be costly, especially in today's volatile political climate. When government agencies fail to protect personal data or ensure services continue running smoothly, confidence in public institutions undoubtedly takes a hit.

Rebuilding that trust can take years, and in the meantime, the British public may become wary of engaging with digital government services. This would ultimately make public services less efficient and more costly to run.

Outdated systems, outdated defenses and the cyber skills gap

One of the biggest challenges in tackling cyber threats in the UK is the cybersecurity skills shortage. The UK Government’s Cyber Security Skills in the UK Labour Market 2024 report found that nearly half (44%) of businesses have skills gaps in basic technical areas. Public institutions, which often work within tight budgets, can struggle to attract and retain cybersecurity professionals, leaving them vulnerable to attacks.

At the same time, many government bodies are still relying on outdated IT systems. The same 2025 NAO report found that 228 legacy systems that lack security patches are still in use across government agencies, with 53% (120 systems) having no fully funded plan for replacement or upgrades.

Many of these systems operate on outdated software that no longer receives security updates, making them easy targets for hackers. While replacing these systems is costly and complex, failing to do so only increases the risk of future cyber incidents.

Bridging the gap

Tackling these cybersecurity challenges requires investment, but being proactive is far more cost-effective long-term than constantly firefighting cyber breaches. Another key priority is improving incident response capabilities.

Cyberattacks are not just a possibility but an inevitability, and government agencies must be equipped to detect and rapidly respond to threats. Investing in advanced monitoring systems, real-time threat intelligence sharing, and dedicated incident response teams can significantly reduce the impact of cyberattacks, ensuring that disruptions are minimized, and services can quickly recover.

Increasing government funding and strategic planning for cybersecurity is also crucial. Cybersecurity cannot be treated as an afterthought, it must be a core component of government budgets and long-term strategy. Allocating sufficient resources for risk assessments, cybersecurity training, and continuous upgrades to IT infrastructure will ensure that public services remain secure and resilient against evolving threats.

Addressing the cyber skills gap is another fundamental aspect of strengthening the cybersecurity of UK Public Services. The shortage of skilled professionals means many public sector organizations lack the expertise needed to defend against and recover from cyberattacks.

The government must take action to attract and retain cybersecurity professionals by offering competitive salaries, investing in robust training programs, and promoting cybersecurity careers within the public sector. Additionally, upskilling existing employees with cybersecurity training can help build a more resilient internal security workforce, reducing the reliance on external expertise.

Securing the UK’s public services

The increasing threat of cyberattacks on UK public services cannot be ignored. With financial losses increasing, public trust on the line, and outdated IT systems creating massive vulnerabilities, there is an urgent need for action. By modernizing IT infrastructure, strengthening incident response, increasing funding, and addressing the cyber skills gap, the UK government can safeguard public services against cyber threats.

Failing to act now will only lead to greater costs, more disruptions, and an even more vulnerable public sector. Therefore, cybersecurity must be a top priority to protect public services, data, and finances from an ever-growing digital threat.

We've featured the best encryption software.

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

Another week, another OpenAI announcement. Just last week the company announced ChatGPT would get a major memory upgrade, and now CEO, Sam Altman, is hinting at more upgrades coming this week.

On X (formerly Twitter), Altman wrote last night, "We've got a lot of good stuff for you this coming week! Kicking it off tomorrow."

Well, tomorrow has arrived, and we're very excited to see what the world's leading AI company has up its sleeve.

We're not sure when to expect the first announcement, but we'll be live blogging throughout the next week as OpenAI showcases what it's been working on. Could we finally see the next major ChatGPT AI model?

Good afternoon everyone, TechRadar's Senior AI Writer, John-Anthony Disotto, here to take you through the next few hours in the lead up to OpenAI's first announcement of the week.

Will we see something exciting today? Time will tell.

Let's get started by looking at what Sam Altman said on X yesterday. The OpenAI CEO hinted at a big week for the company, and it's all "kicking off" today!

we've got a lot of good stuff for you this coming week!kicking it off tomorrow.April 13, 2025

• Windows 11 24H2’s April patch is causing some odd bugs
• There are multiple problems with Windows Hello logins
• The update is also completely failing to install for some people

Windows 11 24H2’s latest patch is causing some trouble with failures to install (once again) and also Windows Hello is ending up broken in some cases.

Let’s start with the bug in Windows Hello, which is the feature that allows for secure logins to your PC (via facial recognition using a webcam, or other methods besides).

Windows Latest reports that some users have found that the cumulative update for April 2025 (known as KB5055523) is affecting those who use Windows Hello for facial recognition with a privacy shutter over the webcam.

How does a facial login even work with a privacy shutter obscuring the camera? Well, it works just fine because the tech leverages an infrared sensor on the camera which can function through a physical shutter (a plastic slider that more privacy conscious folks use to rule out any danger of them being spied on through the webcam).

The problem is that the April update appears to mess with the infrared sensor so login no longer works unless you open that privacy shutter – which is hardly ideal.

Windows Latest found the issue present on an HP Spectre laptop, also highlighting a report from a Windows 11 user who has been affected by this bug and tells us: “I’ve reproduced the issue several times, with several clean installs. The webcam we’re using is the Logitech Brio 4k, with the latest firmware and drivers, which is compatible with Windows Hello.”

There are other reports on Reddit, too, such as this one: “Just wanted to see if anyone else is experiencing issues with Windows Hello face recognition since the latest update (KB5055523). Before, I had the camera (Brio 4K) covered and it would work fine. Now, I have to remove the privacy cover for it to recognize me. It doesn’t seem to be able to unlock with the infrared camera alone now.”

There are a number of replies from folks echoing that they have the same bug affecting their logins. The only solution is seemingly to uninstall the April update (which strongly suggests this is a problem caused by this latest patch from Microsoft).

That said, Windows Latest does offer a possible solution, which is achieved by heading into Device Manager (just type that in the Windows search box, and click on it). In there, you need to click on the little arrow next to where it says ‘Cameras’ and then you might see two cameras listed – the IR (infrared) camera and RGB/color (normal) camera. Right click on the latter and disable it, but leave the infrared camera on. Now, if you go and set up Windows Hello again, it may work correctly.

Windows Latest doesn’t guarantee this, though, and says this will only do the trick for some Windows 11 users. If you’re desperate for a fix, you can give this a whirl, as you could be waiting a while for the official solution from Microsoft. Just don’t forget that you will, of course, have to reenable the main (RGB) camera after any fix is applied.

Interestingly, in the official patch notes for the April update, Microsoft does flag a problem with Windows Hello – but not this one. Rather, this is a separate issue, albeit one that shouldn’t affect those running Windows 11 Home.

Microsoft explains: “We’re aware of an edge case of Windows Hello issue affecting devices with specific security features enabled … Users might observe a Windows Hello Message saying ‘Something happened and your PIN isn't available. Click to set up your PIN again’ or ‘Sorry something went wrong with face setup.’”

Microsoft provides instructions on what to do if you encounter this particular problem via its April patch notes (under known issues).

However, this bug only affects those using System Guard Secure Launch (or Dynamic Root of Trust for Measurement), which is only supported on Windows 11 Pro (or enterprise editions). So as noted, if you’re on Windows 11 Home, you shouldn’t run into this hiccup.

(Image credit: Shutterstock / Khosro) Installation blues

Windows Latest also picked up a second major problem, namely installation failures (which are nothing new). These seem to be happening again with the April 2025 update, with the usual meaningless error messages accompanying an unsuccessful attempt (hexadecimal stop errors such as ‘0x80070306’).

The tech site notes that it has verified reports of the update failing to install in this way, or its progress getting stuck at either 20% or 70% in some cases, never actually completing. Windows Latest informs us that Microsoft is seemingly investigating these installation failures.

There are other reports of this kind of problem on Microsoft’s Answers.com help forum (and some possible suggestions of solutions from a customer service rep, some of which have reportedly worked for some, but not for others). In the case of the original poster of this thread, the update was repeatedly failing to install and showing as ‘pending restart’ which is an odd twist in the tale.

Overall, then, there’s some degree of weirdness going on here, as the Windows Hello failures are rather odd, as are some of these installation glitches. Am I surprised at that, though? No, because Windows 11 24H2 has produced some very off-the-wall bugs (and general bizarreness besides) since it first came into being late last year.

If you’re in the mood for some highlights of the more extreme oddities that have been inflicted on us by version 24H2, read on…

(Image credit: Microsoft) 3 of the weirdest Windows 11 bugs

1. Language swap goes completely off the rails

Some of the most peculiar bugs I’ve ever seen have surfaced in Windows 11, and particularly in the 24H2 update.

One of my favorites – if that’s the right word (it probably isn’t if you were affected by this problem) – is the spanner in the works somewhere deep in Windows 11 that caused the operating system to be displayed as a mix of two different languages. This happened when some users changed the language in Windows 11 from one choice to another, after which a good deal of the operating system’s menus and text remained in the original language. Confusing? No doubt. How did it even happen? I haven’t got a clue.

2. Baffling deletion of Copilot

Last month, Microsoft managed a real doozy by allowing a bug through that actually ditched Copilot. Yes, in a time where the software giant is trying desperately to promote its AI assistant and rally support, last month’s patch uninstalled the Copilot app for some Windows 11 users. That was highly embarrassing for Microsoft, especially as some folks felt it was the first Windows bug they were pleased to be hit with.

3. See more – or less – of File Explorer

A really memorable one for me was late last year when Windows 11 24H2 was beset by a problem whereby a menu in File Explorer (the folders on your desktop) flew off the top of the screen. Yes, the ‘See more’ menu offering more options to interact with files went past the border of the screen, so most of it wasn’t visible – you saw less of it, ironically. And that meant you couldn’t use those non-visible options.

Again, how did Microsoft break a major part of the Windows 11 interface in such a fundamentally crude way? Your guess is as good as mine, but I suspect the transition to a new underlying platform for Windows 11 24H2 had something to do with it. (This bug has only just been fixed, by the way, and that happened with this most recent April update).

You may also like...

• Microsoft is digging its own grave with Windows 11, and it has to stop
• Windows 11 is getting a very handy change to the taskbar, as Microsoft takes a leaf from Apple’s Mac playbook
• Microsoft looks to be making a big change to how you install and log in to Windows 11 – and I’m not happy about it at all

• Google revealed prototype AR glasses at TED
• Those glasses are apparently being made by Samsung under Android XR
• The AR glasses are reportedly set to release in 2026

The Ray-Ban Meta smart glasses have been a wild success for Meta, and it appears that Samsung and Apple have taken notice as the duo plans to launch their own Android XR and Apple XR glasses.

We got our first peek at the Android XR glasses last week during a TED talk hosted by Google's head of augmented reality and extended reality, Shahram Izadi.

The specs are a significantly smaller package than the Project Moohan Android XR headset Google and Samsung are collaborating on (via Axios) – and you can see the headset behind him on a shelf to get some sense of the size difference.

In the surprise demo, Izadi used the glasses to perform a few tasks including live translation from Farsi to English, scanning a book using its in-built cameras, and helping them find their keys using a feature called ‘Memory.’ They also pack in a display so the wearer can receive information visually, not just via audio cues.

(Image credit: Jason Redmond / TED)

Now The Korea Economic Daily is claiming this prototype is Samsung-made under the duo’s existing Android XR partnership – and it’s slated to launch next year.

While leaks should be taken with a pinch of salt we’ve heard plenty of reports that Meta plans to launch its Ray-Ban Meta smart glasses with a display later this year, so it’s not out of the question that an Android XR device could follow soon after.

What’s more, given how impressive the prototype seems to be already – being able to perform a slew of tasks in a sleek package – it again seems likely that Google and Samsung aren’t far from having a consumer-ready product. I just hope they aren’t as pricey as Meta’s glasses are rumored to be.

What's Apple up to?

(Image credit: Apple)

Apple is also keen to create lightweight AR specs, according to Bloomberg's Mark Gurman (behind a paywall). According to people familiar with Apple's AR plans "It’s the only thing he’s really spending his time on from a product development standpoint.”

Though according to Gurman, Apple’s glasses might still be a few years away.

Because true AR glasses aren't yet achievable in Apple's mind (according to people in the know) it apparently first wants to focus on equipping its Apple Watch and AirPods with AI cameras to achieve some of the functionality of the Ray-Ban Meta AI glasses we have today.

Given Apple’s struggles with AI on the iPhone 16 it makes sense that it wouldn’t currently want to focus on AI smart glasses, though it does feel it could be late to the glasses party considering how soon Samsung, Google and Meta are expected to be launching specs.

As with all leaks and speculation we’ll have to wait and see what Apple has up its sleeve, but it could benefit from the same advantage Google will – one that Meta lacks.

The smartphone synergy strategy

The most striking part of Google’s prototype is the XR device looks an awful lot slimmer than other prototype AR glasses we’ve seen – such as the Meta Orion AR specs and Snap’s 5th-gen Spectacles developer kit. That’s because it is apparently leveraging Google’s biggest advantage: an Android phone according to Izadi.

He explained, "These glasses work with your phone, streaming back and forth, allowing the glasses to be very lightweight and access all of your phone apps."

Because they lack their own phone brands to rely on, Meta and Snap would ideally want users to rely solely on their own standalone platform – one which they have complete control over to introduce the apps and experiences they most want to build. Control they don’t have when merely piggybacking off Android via phone apps.

Google, seeing as it runs the Android ecosystem, likely isn’t keen for people to abandon its phones quite yet.

Snap's spectacles are cool, but bulky (Image credit: Future)

So its glasses are instead designed to leverage the processors in your phone rather than a sophisticated chipset built into the glasses themselves (though we suspect they still have a little onboard processing power for simple display and camera operations).

This allows Google to remove some of the bulk a beefy chipset requires – such as sophisticated cooling and a bigger battery – to create a slimmer end product without sacrificing overall performance (at least in theory).

Apple could leverage a similar smartphone-to-glasses relationship with its own AR specs, though Meta and Snap’s third-party approach has one advantage in that they are system agnostic. You can use which smartphone you want, or you might not even need a smartphone at all.

Hopefully Google’s symbiotic relationship between its glasses and phones results in smart specs that aren’t as pricey as its competitors, as they don’t need to pack in as many components. For example, Meta’s rumored upcoming XR glasses with a screen are expected to cost as much as $1,400 (£1,100 / AU$2,200).

For now we'll need to wait and see, but Google and Samsung's smart specs look promising, and if they are set to launch next year 2026 can't come soon enough.

You might also like...

• Pico 4 Ultra VS Meta Quest 3: the battle of the best mid-range VR headsets
• This beginner-friendly train-driving simulator is the chillest Meta Quest 3 game I've played – but it's not for everyone
• You might get a free Meta Quest 3 to use on your next flight, but I'm not keen on the advertising it'll serve you

• Security researcher finds more than 30 unlisted Google Chrome extensions
• Cumulatively they have more than four million users
• They are potentially hazardous, with a variety of security risks

A cybersecurity researcher from Secure Annex recently discovered more than 30 unlisted browser extensions that put more than four million of its users at different security risks.

In a detailed analysis, researcher John Tuckner explained software developers will sometimes unlist their extensions if they’re not operating properly.

However, he also suggested that malicious actors might unlist them to make it harder for security teams to detect and flag them. After all, these hidden tools cannot easily be found via search engines, or public directories.

Monitor your credit score with TransUnion starting at $29.95/month

TransUnion is a credit monitoring service that helps you stay on top of your financial health. With real-time alerts, credit score tracking, and identity theft protection, it ensures you never miss important changes. You'll benefit from a customizable online interface with clear insights into your credit profile. Businesses also benefit from TransUnion’s advanced risk assessment tools.

Preferred partner (What does this mean?)View Deal

Flagging for malicious behavior

“Many companies provide their software through unlisted extensions because it makes it harder for any normal user to find the extension and then hit a wall when it isn’t functional,” he said. “It has also been known as a way to target users to install a malicious extension while being really hard to detect by security teams.”

Some of the extensions Tuckner found, like "Fire Shield Extension Protection," request excessively broad permissions. These permissions include access to users' web traffic, stored cookies, and even browser tabs, which opens the doors to the misuse of potentially sensitive data.

“While the management API is requested, so is access to many more permissions that provide the ability to interact with web traffic on all URLs, access cookie storage, manage browser tabs, and execute scripts!,” Tuckner explained.

Secure Annex's analysis flagged these extensions for potentially malicious behavior, such as accessing stored cookies, or matching signatures associated with known malware. The researcher suggested users remove these unlisted extensions, since their hidden and overly intrusive nature creates unnecessary vulnerabilities.

Fortunately, Tuckner did not find any extensions stealing login credentials or payment information.

However, he stressed that this level of obfuscation for software that can be remotely controlled could mean it can be used as an infostealer. “That is ultimately the problem and threat these extensions pose when they can be controlled remotely.”

We have reached out to Google for comment.

Via Ars Technica

You might also like

• Malicious "polymorphic" Chrome extensions can mimic other tools to trick victims
• Take a look at our guide to the best authenticator app
• We've rounded up the best password managers

• The Samsung Galaxy Z Fold 7 has been spotted on Geekbench
• The listing reveals Android 16, which could make this one of the first phones with that software
• There's also mention of 12GB of RAM and a Snapdragon 8 Elite chipset

The Samsung Galaxy Z Fold 7 will probably launch in July, and one sign that it’s approaching is that it has been spotted on a benchmarking site – revealing some details about the phone in the process.

Leaker Tarun Vats (via Phone Arena) has found a Geekbench listing for the Samsung Galaxy Z Fold 7, and it mentions Android 16 – which, if the phone does launch in July, would mean it’s likely to be one of the first handsets running Android 16.

That’s pretty speedy for Samsung, considering the company still hasn’t got Android 15 running on most of its phones, but it’s something we’ve heard about more than once, so there’s a good chance Android 16 (with Samsung’s One UI 8 overlay) really will be available on the Galaxy Z Fold 7 at launch.

Galaxy Z Fold7 spotted running Android 16 with One UI 8 on GeekBench!!OpenCL-Score: 18143Build Version: F966USQUOAYD7 pic.twitter.com/2QGU1udttxApril 12, 2025

A faster chipset but no increase in RAM

This benchmark also mentions 12GB of RAM, which is the same amount as you’ll find in the Samsung Galaxy Z Fold 6. So it would be nice to see Samsung increase this, but given that the Samsung Galaxy S25 Ultra also has 12GB we didn’t really expect to see more than that here.

Finally, it has a motherboard that’s listed as “sun”, which is the codename for the Snapdragon 8 Elite, so that’s probably the chipset this phone will use.

That’s good but unsurprising news – good because that’s the most powerful chipset currently available to Android phones. Unsurprising because the Samsung Galaxy Z Fold 6 used the previous generation of this chipset (the Snapdragon 8 Gen 3).

Both that RAM amount and chipset have also previously been leaked, so there aren’t really any surprises here, but this is just one more piece of evidence suggesting that these are the specs we’ll get.

We’ll probably find out for sure in July, but stay tuned to TechRadar in the meantime for any additional leaks and rumors.

You might also like

• Samsung’s One UI 7 update reportedly delayed due to a serious bug, extending the wait for Android 15
• Samsung Galaxy S25 Edge: latest news, rumors, and everything we know so far
• Samsung Galaxy Z Flip 7: latest news, rumors, and everything we’ve heard so far

• Laboratory Services Cooperative confirmed suffering a data breach in October 2024
• An investigation uncovered that the crooks stole sensitive data
• The information has not yet surfaced on the dark web

Laboratory Services Cooperative (LSC), a significant US-based laboratory for healthcare providers, has revealed it suffered a data breach and lost some member information.

In a press release published on its, LSC said it identified “suspicious activity” within its network on October 27, 2024. It notified the police and brought in third-party cybersecurity experts to analyze the incident and assess its effects.

The investigation concluded in February 2025, and showed that “certain LSC patient and worker-related data might be affected.”

Monitor your credit score with TransUnion starting at $29.95/month

TransUnion is a credit monitoring service that helps you stay on top of your financial health. With real-time alerts, credit score tracking, and identity theft protection, it ensures you never miss important changes. You'll benefit from a customizable online interface with clear insights into your credit profile. Businesses also benefit from TransUnion’s advanced risk assessment tools.

Preferred partner (What does this mean?)View Deal

Payment info stolen

According to BleepingComputer, roughly 1.6 million people are affected. The stolen data has not yet leaked online, and no one claimed responsibility just yet.

The data stolen varies from case to case, LSC says, adding that it may include contact details (names, addresses, phone numbers, emails), as well as medical and clinical information (date(s) of service, diagnoses, treatment, medical record number, lab results, patient/accession number, provider name, treatment location, and related-care details), health insurance information (plan name, plan type, insurance companies, and member/group ID numbers), and most worrisome - billing claims and payment data.

The data could also include claim numbers, billing details, bank account details (bank name, account number, routing number), billing codes, payment card details, balance details, and similar banking and financial information.

Furthermore, crooks may have taken people’s Social Security numbers, driver’s license or state ID numbers, passport numbers, dates of birth, demographic data, student ID numbers, and other forms of government identifiers.

“For LSC workers, the information involved may also include details about their dependents or beneficiaries if that information was provided to LSC,” the press release reads.

LSC acts as a cooperative, meaning it's owned and governed by its member physicians and clinics. The breach mostly affects people who did tests through some Planned Parenthood centers that use LSC for their testing, it was said.

You might also like

• Thousands of healthcare records exposed online, including private patient information
• Take a look at our guide to the best authenticator app
• We've rounded up the best password managers

I spent hours testing the Roborock Saros Z70 and its robotic arm. Here's what you need to know about the $1,899 robot vacuum.

• The Wheel of Time's cast hope that they'll return for a fourth season
• Amazon hasn't renewed the high fantasy series for another installment yet
• The Prime Video show's best-rated season indicates that it should be, though

The Wheel of Time's cast aren't sure if Amazon will continue to weave its incredibly dense source material into a fourth season.

Speaking to TechRadar before the show's third season began airing in mid-March, Josha Stradowski and Daniel Henney revealed they hadn't heard if the high fantasy series would be renewed again. It's possible, then, that The Wheel of Time's third season could mark the end of the road for Prime Video's TV adaptation.

However, Stradowski and Henney, who play Rand al'Thor and Lan Mandragoran, are holding out hope for some good news once the Amazon TV Original's latest chapter ends. And, with season 3's final episode due to be released this Thursday (April 17), there's a chance that a season 4 announcement could be made in the hours or days after its arrival on one of the world's best streaming services.

"We don't know much," Henney told me when I asked if anyone had informed them about the show's renewal. We're just hopeful and have all of our fingers crossed.

Season 3's trippy fourth episode was the Prime Video Original's best entry yet (Image credit: Amazon MGM Studios)

"Season three is something we're incredibly proud of," Henney continued. "Watching it back, I couldn't be happier with how it turned out. I'm sure Josha will agree that this is The Wheel of Time we set out to make."

"Yeah, we're confident [that a fourth season will be greenlit]," Stradowski added.

"When I was reading the book series, it wasn't until book four [The Shadow Rising, which season three is based on] that I thought 'Wow, okay, now I understand why so many copies have been sold'. I feel like season three is where we found our form, so we've all got our fingers firmly crossed."

May the Light protect you

Poring over the viewership data for season 3 like... (Image credit: Amazon MGM Studios)

It's hard to determine if one of the best Prime Video shows (in my view, anyway) will get a fourth outing.

For starters, Amazon rarely reveals official viewing figures for any of its new movies or shows. On the off-chance that it does, it's usually to boast about the millions of people who tuned in to watch the premiere of a new TV Original or the debut of any of the new Prime Video movies it develops in-house.

The best information we have from a viewership perspective, then, is from third-party sources, such as audience engagement experts Nielsen. Companies like this are our best bet at understanding how The Wheel of Time season 3 performed, albeit with a couple of caveats.

Indeed, Nielsen's weekly viewing figures only take a project's US audience into consideration. Additionally, their data is based around the number of minutes viewed, which can somewhat skew the figures when trying to work out how many people actually watched the show.

The tapestry of Episode 7. pic.twitter.com/g8UBcloijWApril 13, 2025

All of that said, Nielsen hasn't released any numbers for season 3 yet. The audience aggregation firm releases figures one month behind where we're at in the calendar. For example, the latest weekly Top 10 information comprises viewership data for the week running March 10 to 16. At the time of publication (April 14), that's around 30 days ago.

With The Wheel of Time's third season not debuting until March 13, it'll be a few more days before we find out how well its three-episode premiere performed. That'll give a clearer picture as to whether a fourth season will be greenlit.

There is hope that Amazon's TV retelling of Robert Jordan's best-selling novels will get another installment, though. Based on its Rotten Tomatoes score, season 3 is the highest-rated entry yet, with general audiences (82%) and critics (97%) loving what they've seen. Compared to season 2 (80% and 86%) and season 1 (61% and 81%), this season is by far the best since the show made its debut in August 2021.

I was similarly effusive in my praise for this chapter, too. In my review of The Wheel of Time season 3, I called it "a spellbinding return to form for Prime Video's fantasy TV underdog". Based on this season's seven episodes to date, which I really enjoyed, I'd be gutted if it's canceled just as it's starting to get really interesting, too.

You might also like

• 'There's a reason why we do it': The Wheel of Time showrunner responds to fans who are still upset over the Prime Video show's plot alterations
• Learn more about The Rings of Power season 3
• Or get the lowdown on Invincible season 4

• Garmin begins its beta update cycle for the Garmin Instinct 3
• Among 30 updates are eight new features
• Those features include rucking, Passcodes, and Garmin Connect+

Garmin is beginning to roll out updates for the new Garmin Instinct 3. The updates, currently in beta, include 22 minor changes, fixes, updates, and eight new features, including support for the new Garmin Connect+ premium service.

The Garmin Instinct 3's new features, currently only available to members of Garmin's public beta program as Beta Version 9.07, will roll out to all watches once testing is complete. While there are 30 updates in total, the eight significant new features are listed below:

• Garmin Connect+ support
• Rucking workout profile
• Pack weight option added to the Rucking profile
• Passcode support
• Breathing Variations
• Daily Suggested Workouts added to Workout app for Solar/Instinct E
• Morning report animations
• Extra locations in Weather widget

The rest of the changes are fixes to various bugs and preventing potential crashes. Of the eight significant changes above, Passcode is a big one, allowing you to lock your watch for added security, and Rucking is a great workout profile for those looking to make their hikes more challenging by adding a weighted backpack. The Pack Weight functionality is a nice touch, allowing you to granularly track your improvements by weight in addition to distance.

We previously reported on Garmin adding Rucking, and were mildly surprised it didn't come as part of the Instinct 3 at launch, as this chunky watch is tailor-made for rough trails. However, it may simply not have been ready, or the Pack Weight functionality may have created additional complications when designing the workout profile.

(Image credit: Garmin)

However, support for Garmin Connect+, Garmin's new premium subscription service, is likely to have a somewhat mixed reception, to say the least, while the other features are expected to be welcome changes for Instinct 3 owners. If you were following the premium tier's launch, you'll know Garmin Connect+ was met with enormous backlash, as we reported after spending two days fielding reader comments and emails.

Users' main concerns are with the relatively basic and unhelpful AI advisor service, Active Intelligence, and the risk of future new features being placed behind the GC+ paywall. Users feel strongly about having to fork over additional monthly charges to get the most from Garmin, even after buying an expensive watch.

However, fortunately, there has been no mention of any of the new features listed above being locked behind Garmin Connect+, which is great news for ruckers and those interested in protecting their watches with a passcode.

You might also like...

• Garmin has just crossed a powerful bike light with a 4K dash cam – and as a cyclist, I’m all in
• 5 fitness apps that can help you build muscle in the gym
• Hot on the heels of Garmin's controversial subscription plan, Polar's just unveiled its own

The Moto G Power may not be the cheapest of Motorola's budget phones, but it's worth the upcharge.

The Galaxy XCover7 Pro and Galaxy Tab Active 5 Pro place durability at the forefront. And with more years of OS and security updates, they can last even longer.

A relic of historic redlining that restricted marginalized peoples from buying homes still persists in internet access today.

• Newly uncovered Geekbench benchmarks suggest the rumored Motorola Razr Plus could best the Galaxy Z Flip 7 specs-wise
• The new Razr Plus could sport the Snapdragon 8 Elite chipset and 16GB of RAM
• The Galaxy Z Flip 7 in turn may only get 12GB of RAM, and either the Snapdragon 8 Elite or Exynos 2500 chipset.

The Samsung Galaxy Z Flip 7 hasn’t even been confirmed by Samsung yet, but it already seems like a powerful rival is on the way in the form of the rumored 2025 Motorola Razr Plus.

According to newly uncovered Geekbench benchmark reports, this year’s rumored Motorola Razr Plus model comes equipped with a Qualcomm Snapdragon 8 Elite chipset, the Adreno 830 GPU, and an impressive 16GB of RAM.

The Geekbench report – spotted by Xpertpick – also suggests the phone will run an Android 15-based operating system, most likely a version of Motorola’s Hello UI Android wrapper.

Those specs – while still unconfirmed – could give the also unannounced Samsung Galaxy Z Flip 7 a run for its money. As we previously reported, the Galaxy Z Flip 7 could come with either a Snapdragon 8 Elite or Samsung’s own Exynos 2500 chipset.

The Samsung Galaxy Z Flip 6 is our pick for the best flip phone currently available (Image credit: Samsung)

The Snapdragon 8 Elite would be a safe bet to match or exceed the same chipset found in the Razr Plus – it’s typical for Samsung phones to come with special “for Galaxy” versions of Qualcomm chipsets.

However, a Galaxy Z Flip 7 equipped with the Exynos 2500 chipset wouldn’t have such a certain edge over the Razr Plus – issues with Samsung’s Exynos chipset platform are well documented, and have prevented the tech giant from fitting its flagship phones with its own chipsets for the past few years.

That’s not to mention that the Galaxy Z Flip 7 is expected to launch with 12GB of RAM to the Razr Plus’ rumored 16GB – that difference could give the Razr Plus an edge for multitasking and on-device AI processing.

Additionally, the current gen Motorola Razr Plus starts at $1,000 / £1,000 / AU$1,700, while the Samsung Galaxy Z Flip 6 starts at $1,099 / £1,049 / AU$1,799. Assuming neither phone sees a price hike with their next iteration, that’ll be another win for the Razr Plus.

As PhoneArena notes, Motorola is expected to release the new Razr lineup, which will launch as the Razr 60 lineup outside of the US, on April 24 – so we don’t have too long to wait to find out the actual specs of this new foldable.

Until then, the above is effectively entirely based on rumors – still, it’s always nice to hear of phone makers challenging Samsung in the foldable space. Let us know if you’re looking forward to either of these phones in the comments below.

You might also like

• Samsung’s One UI 7 update reportedly delayed due to a serious bug, extending the wait for Android 15
• ‘It will take time for foldables to become mainstream’: OnePlus defends decision to delay the OnePlus Open 2 and bow out of the foldable phones race in 2025
• Google's new Android update will give your devices a battery boost – plus a helpful new feature for Google Wallet

Every week, there's a new AI trend, but this viral ChatGPT image-generation prompt might be a step too far - turning our beloved pets into humans.

As a dog dad who treats my French Bulldog, Kermit, like an absolute prince, the idea of seeing him as a human made me feel uneasy. My focus at work, however, is creating good content, and if that means I need to see my dog as a human, then I'll suck it up and take one for the team.

If you want to cut to the chase, here's the prompt: "Can you create an image of this dog/cat as a human?" You'll need to upload an image of your pet, and ChatGPT may ask follow-up questions to get a better idea for the image generation.

Once you've uploaded your image and asked your prompt, it's time to wait for the result...

I don't want to see this

(Image credit: Future / ChatGPT)

I uploaded a photo of Kermit to ChatGPT, and after a considerable wait (please make the image generation quicker, OpenAI), I got to see my furry friend as a human. I wasn't quite prepared to see a grown man sitting on the yellow armchair Kermit likes to frequent, and I immediately felt a sense of uncanny valley.

Sat on Kermit's throne was what appeared to be a happy-go-lucky man in his early 20s, wearing a yellow t-shirt with a frog on it. Kermit is three years old, which would line up with the age of ChatGPT's image, but apart from that, I'm not really convinced AI has captured him in human form.

Just to see what another dog would look like, I asked TechRadar's Editor-In-Chief, Marc Mclaren, to do the same for his black Labrador, Rigby. Weirdly, Rigby as a human looks fairly similar to Kermit, despite being a completely different breed. Marc's response to the ChatGPT image was, "Pretty spot on to be honest. The son I never had."

You know, the more I think about the concept of seeing your pet as a human, the more it makes me feel uneasy. Last year, I got to see Shazam's AI-powered collar in action, and while I'm now convinced that product might be vaporware, I left the impressive demo questioning what it means to be a pet owner.

You see, you get a connection with your dog that's completely different from any human relationship you have in your life. A dog is your best friend; it revolves its whole life around your existence, and it understands despite the inability to verbally communicate.

Shazam's idea of adding a voice to man's best friend completely overlooks the emotional connection you have with your animal, instead introducing technology to attempt to vocalize feelings.

Now, why am I talking about an AI voice collar? Well, I feel like even the concept of imagining a dog as a human overlooks what a dog is in the first place. I get this is just a fun trend, but I have a feeling I'll struggle to look at Kermit in the same way for the next few days, as he sleeps on my lap, sleeps in my bed, and cries for food at dinner time.

What about a human as a dog?

(Image credit: Future / ChatGPT)

After feeling a bit revolted by the human version of my little pup, I decided to creep myself out some more by asking ChatGPT to turn a photo of me into a dog.

You can do the same by asking, "Can you create an image of this human as a dog?" Again, ChatGPT may ask for extra context to determine the breed, etc.

At first, ChatGPT took a while to determine what a bald man with a beard would look like in dog form. It took so long, in fact, that the image generation tool completely failed altogether.

I tried again, and after a considerable amount of time, ChatGPT showed me an image of what appeared to be a Labrador or a Golden Retriever with a human beard. I looked at the image, shrieked, and decided that was enough ChatGPT image generation for the day.

I think imagining an animal as a human or vice versa might be my least favorite ChatGPT trend yet, although I fully expect to see another even creepier trend appear on social media by the end of the week.

You might also like

• Will Adobe's AI agents cause the death of creativity?
• I refuse to jump on ChatGPT’s Studio Ghibli image generator bandwagon because it goes against everything I love about those movies
• Spotify is about to be flooded with AI-made ads, and I wonder if it will make much of a difference to businesses

• Mythic Quest has ended after four seasons at Apple TV+
• Season 4's finale episode will premiere next week with a new ending
• The executive producers have thanked Apple TV+ in a statement

I bring disappointing news, because Mythic Quest season 4 is the end of the road for the hit comedy series. It's one of the best Apple TV+ shows for a reason, and we're all sad to see it go.

TechRadar contributor Samuel Roberts called it a "masterpiece", and it's one of the many reasons we crowned Apple TV+ as Streaming Service of the Year in 2024.

According to exclusive reports from Variety, Mythic Quest has not been renewed beyond its fourth season and a statement has been released thanking Apple TV+ as well as the cast, crew and fans.

“Endings are hard. But after four incredible seasons, ‘Mythic Quest’ is coming to a close,” series executive producers Megan Ganz, David Hornsby and Rob McElhenney said. “We’re so proud of the show and the world we got to build – and deeply grateful to every cast and crew member who poured their heart into it.

They added: "To all our fans, thank you for playing with us. To our partners at Apple, thank you for believing in the vision from the very beginning. Because endings are hard, with Apple’s blessing we made one final update to our last episode – so we could say goodbye, instead of just game over.”

What do we know about the final episode of Mythic Quest season 4?

(Image credit: Apple)

Unlike regular TV shows which only get one ending, Mythic Quest has been allowed to respawn and try again, very reminiscent of video games and a fitting send off.

According to Variety, an updated finale episode with a brand new ending will premiere next week on Apple TV+. This might provide some sort of relief for disappointed fans, considering the season finale did end on a cliffhanger.

So, we might be getting a satisfying end. Or, an even more frustrating one, depending on how benevolent the creators were feeling. You'll have to wait and see for yourself when the new episode drops.

An air date for the new episode hasn't been confirmed beyond "next week", frustratingly, but if it's following in the footsteps of Mythic Quest's previous release schedule, it may arrive on Wednesday, April 23.

You might also like

• Apple TV+ unveils the trailer for a new comedic sci-fi series that sees Alexander Skarsgård as a sentient AI cyborg
• Slow Horses season 5: everything we know so far about the popular Apple TV+ show's return
• Everything new on Apple TV+ in April 2025: The Studio, Your Friends and Neighbors, Government Cheese, and more

• Microsoft is testing new features for Defender for Endpoint
• The features will block traffic from and to undiscovered endpoints
• The goal is to minimize malicious lateral movement

Microsoft wants to minimize the risk of undiscovered endpoints by adding a new feature to its Defender for Endpoint product, which will automatically block all traffic to and from such devices.

Such devices are a major security risk because they can bypass monitoring, lack security controls, and potentially serve as entry points for cyberattacks or data exfiltration.

The company is currently testing a new capability that will contain the IP addresses of devices that have not been discovered or onboarded to Defender for Endpoint.

Monitor your credit score with TransUnion starting at $29.95/month

TransUnion is a credit monitoring service that helps you stay on top of your financial health. With real-time alerts, credit score tracking, and identity theft protection, it ensures you never miss important changes. You'll benefit from a customizable online interface with clear insights into your credit profile. Businesses also benefit from TransUnion’s advanced risk assessment tools.

Preferred partner (What does this mean?)View Deal

Automatic protection

"Containing an IP address associated with undiscovered devices or devices not onboarded to Defender for Endpoint is done automatically through automatic attack disruption. The Contain IP policy automatically blocks a malicious IP address when Defender for Endpoint detects the IP address to be associated with an undiscovered device or a device not onboarded," Microsoft said.

"Through automatic attack disruption, Defender for Endpoint incriminates a malicious device, identifies the role of the device to apply a matching policy to automatically contain a critical asset. The granular containment is done by blocking only specific ports and communication directions."

We do not know yet when the feature will be released for the users, but we do know that it will be available on Defender for Endpoint-onboarded devices running Windows 10, Windows 2012 R2, Windows 2016, and Windows Server 2019+.

Microsoft also explained that there is a way to stop the product from containing different IP addresses by restoring the connection. That can be done via the “Contain IP” menu in the “Action Center”, which will have an “Undo” button.

Via BleepingComputer

You might also like

• Microsoft Defender will now warn users about insecure Wi-Fi networks
• Take a look at our guide to the best authenticator app
• We've rounded up the best password managers

• A new report claims Apple is developing two new Vision Pro headset models
• One version will apparently be lighter and more affordable
• The other will connect to a Mac for a lower-latency experience

Despite its technical brilliance, Apple’s Vision Pro headset has struggled to sweep consumers off their feet, with serious questions being raised about the future of the virtual reality (VR) headset. In spite of this potential crisis, though, Apple apparently has a plan to right the ship and get the Vision Pro back on track.

At least, that’s according to Bloomberg journalist Mark Gurman’s latest Power On newsletter. There, Gurman says that Apple is working on two new Vision Pro models that could shake up the device and potentially give it a new foothold in the industry.

Neither of these models will be simple refreshes. While Gurman says that Apple had previously considered a few minor tweaks – like changing the headset’s chip from the M2 to the M5 chip – Apple is apparently now mulling more ambitious plans.

Instead, Apple is “now looking to go further,” Gurman says, with one device that is both cheaper and more lightweight than the current Vision Pro, which should make it both more accessible to consumers and reduce the neck strain that can sometimes occur after long periods of use.

The second Vision Pro in development is “more intriguing,” Gurman believes. Supposedly, this model will connect directly to a Mac, which will allow for greatly reduced latency. Some Vision Pro headsets are being used in surgery and in flight simulators, Gurman says, which are two areas where reduced latency will be incredibly important.

Addressing two key problems

(Image credit: Apple)

Interestingly, Gurman likens the second Vision Pro to a product Apple reportedly canceled in January: a set of lightweight glasses outfitted with augmented reality (AR) capabilities that would be tethered to a Mac. The ultimate goal for Apple is to create an entirely wireless pair of AR glasses that a user can wear all day, but it will take many years before that device is ready for prime time.

Still, even if the next Vision Pro isn’t as lightweight as a pair of glasses, anything that reduces its weight will be a welcome change. I don’t own a Vision Pro myself, and one of the main reasons is that I’ve heard the stories of the pain caused by wearing it for too long. If I’m going to make a significant investment in a product like this, I want to be able to use it without risking soreness after just a few hours’ usage.

The other reason I’ve steered clear is the eye-watering price, with $3,500 feeling incredibly steep when the Vision Pro is still in its early stages. I know I’m not alone here, with multiple reports suggesting that sales of the device have been very slow.

If Apple can address both of those problems, I might finally be tempted to fork out for a Vision Pro, particularly for its Ultrawide Mac Virtual Display mode. We've previously described that as a "lightbulb moment" for the device, given it gives you a massive 32:9 aspect ratio – the equivalent of two displays side-by-side. That would definitely be more enticing on a headset that's more comfortable, and also potentially not too much more expensive than buying an extra monitor.

I’m not entirely convinced yet, but at least Apple seems to be heading in the right direction.

You might also like

• Vision Pro at one – I love Apple's revolutionary headset, so why do I hardly ever use it?
• Report: Apple is stopping work on a pair of smart glasses that would have connected to the Mac
• The best VR headsets in 2025: top virtual reality headsets from Meta, Valve and more