TechRadar News

• Pixel phones could soon support themes, according to new code
• It lets customizations be applied more quickly
• The feature may arrive with Android 16

After showcasing the big update at Google I/O 2025 earlier this month, Google is all set to launch Android 16 in the coming days, at least as far as Pixel phones are concerned – and it looks as though Pixel devices are also in line for a major customization upgrade.

The team over at Android Authority has spotted evidence of a Pixel themes feature hidden in the code for the latest Android 16 beta. While the feature hasn't gone live yet, it looks as though it's close to completion.

Many other Android handsets, including the best Samsung Galaxy phones, already offer themes: they let you group together wallpapers, color schemes, and icon packs together in a single bundle, and you can then apply them with a single tap.

It looks as though themes would be available via the current wallpaper and lock screen settings on Pixel phones, available through the Wallpaper and style section in Settings. It would also be quite a prominent option, based on the existing code.

Coming soon?

A new visual language is heading to Android (Image credit: Google)

It's not clear whether you'll be able to view a gallery of themes provided by Google and perhaps other users, as well as creating your own customization combinations. On Samsung phones, there's a large gallery of theme options to pick from.

Another question we have is whether this feature is going to be ready in time for the official, stable Android 16 release: the rumor is that we might see that coming on Tuesday, June 3, though that's yet to be confirmed.

It's possible that Google will wait for a later Android 16 release before pushing out themes and making them available to Pixel owners. Other Android phone makers will then be able to use (or not use) Google's theme code as they see fit.

Upcoming changes we do know about in Android 16 include better battery health monitoring (for newer Pixels), improved security features, and a revamped, more modern look for the interface that's being called Material 3 Expressive.

You might also like

• Samsung's next foldables will launch with Android 16
• These are the best Pixel phones you can buy
• Android 16 may add its own version of iOS StandBy Mode

• Anthropic CEO Dario Amodei claims major white-collar layoffs are on the horizon
• Unemployment could spike to 20%
• Anthropic develops AI models designed to do tasks traditionally done by humans

The CEO of AI firm Anthropic, creators of ChatGPT rival ‘Claude’, has warned the US Government about the possible ‘mass elimination’ of entry-level jobs across law, technology, finance, and other white-collar fields.

Dario Amodei predicted up to half of all entry-level white-collar jobs could be taken by AI, leading to unemployment of as much as 20% in the next one to five years.

Amodei warned most workers are “unaware” that this job apocalypse is imminent; "It sounds crazy, and people just don't believe it", adding AI has the potential to affect society in positive and negative ways; "Cancer is cured, the economy grows at 10% a year, the budget is balanced — and 20% of people don't have jobs."

Shifting sands

It’s worth noting Anthropic is one of the organisations leading the charge in developing technology designed to replace humans.

The warnings that Amodei and others share about the potential impact on unemployment and short-term economic turmoil come with no mitigation plan; "We, as the producers of this technology, have a duty and an obligation to be honest about what is coming."

Axios outlines that Steve Bannon, long time advisor to President Trump and influential media personality, also envisions a scenario in which entry level positions are “eviscerated,” which correlates with the almost 2% rise in tech sector layoffs seen thanks to AI already in 2025.

"I don't think anyone is taking into consideration how administrative, managerial and tech jobs for people under 30 — entry-level jobs that are so important in your 20s — are going to be eviscerated."

This isn’t necessarily the only outcome long term though, with over 55% of companies in the UK who replaced workers with AI ultimately coming to regret their decision, indicating that perhaps the technology isn’t as potent its marketing suggests.

Via Axios

You might also like

• Take a look at our picks for the best AI tools around
• Check out our choice for best antivirus software
• Generative AI isn't biting into wages, replacing workers, and isn't saving time, economists say

• Microsoft is reportedly sidelining its Xbox native handheld in favor of Windows 11 handheld gaming PC optimization
• The 'Project Kennan' handheld collaboration with Asus is still underway
• It appears to be a direct response to Valve's SteamOS success after it became available for non-Steam Deck handhelds

Valve has recently taken the spotlight on handheld gaming PCs, providing SteamOS support for non-Steam Deck handhelds – and as a result, Microsoft has seemingly set aside one of its most-anticipated devices to compete with the Linux-based operating system.

According to an exclusive from our friends at Windows Central, Microsoft has reportedly shelved its Xbox native handheld momentarily, as it will now shift its focus to Windows 11's optimization for third-party handheld gaming PCs – and 'Project Kennan', a new Xbox-themed handheld in collaboration with Asus.

This comes shortly after the launch of a new Lenovo Legion Go S, an officially-licensed SteamOS handheld gaming PC – and it appears as though the revelations of the SteamOS version outperforming the Windows 11 model in gaming may have pushed Microsoft to change its priorities.

Windows 11 is yet to feature a 'handheld' mode or anything that focuses on portability to help streamline the handheld gaming experience. While software like Armoury Crate (alongside tools like Xbox Game Bar) has helped handheld gamers with quick game access and other necessary TDP adjustment settings, it's had a hard time matching up to the SteamOS experience for gamers.

It's not exactly clear whether the codenamed handheld, Project Kennan, will come in the form of a long-awaited Asus ROG Ally successor. However, it's clear that Valve's success has lit a fire under Microsoft to take handheld gaming seriously – and rightly so, as many would argue.

Analysis: Gamers, this is exactly why voicing your frustrations matters

(Image credit: Microsoft)

If there's anything Microsoft's reported actions highlight, it's that voicing your complaints and frustrations does indeed work. I say this because I can almost guarantee that without the praise of SteamOS and the recent criticisms of Windows 11, Microsoft wouldn't lift a finger.

The same case applies to the sudden rise of game prices. I'm still very concerned and adamant about shooting down $80 game prices, since Nintendo kick-started it. Instead of just 'accepting it' and letting it become a precedent, let your displeasure about it be known as much as possible.

Microsoft, Sony, and Nintendo are all more than capable of doing better and being reasonable with consumers, and the only thing I've ever seen work is protesting against measures that are out of line.

I could go on about Sony's PS Plus and Microsoft's subscription services, just for console gamers having access to multiplayer, and just how absurd that truly is – but ultimately, it's an example of what happens when we let things slide or become normalized.

Regardless, I'm pleased to hear that Valve has essentially been a catalyst for Microsoft to take handheld gaming and Windows 11 optimization more seriously, because it's been needed for a very long time now.

You may also like...

• New update for Windows 11 24H2 is the most feature-packed for some time, delivering some seriously useful additions
• The new SteamOS beta already has fixes for non-Steam Deck handhelds – here's why it's better than Windows 11 for gaming
• Forget the Nintendo Switch 2 – I’m more excited that Microsoft could be making Windows 11 gaming handhelds even better with a new ‘handheld mode’

As demand for AI and cloud computing soars, pundits are suggesting that the world is teetering on the edge of a potential data center crunch—where capacity can’t keep up with the digital load. Concerns and the hype have led to plummeting vacancy rates: in Northern Virginia, the world's largest data center market, for example, vacancy rates have fallen below 1%.

Echoing past fears of "peak oil" and "peak food," the spotlight now turns to "peak data." But rather than stall, the industry is evolving—adopting modular builds, renewable energy, and AI-optimized systems to redefine how tomorrow’s data centers will power an increasingly digital world.

1. Shift Toward Modular and Edge Data Centers

Future data centers will increasingly move away from massive centralized facilities alone, embracing smaller, modular, and edge-based data centers. The sector is already splitting out in hyperscale data centers one end and smaller, edge-oriented facilities on the other.

Smaller, modular and edge data centers can be built in a few months and tend to be located closer to end users to reduce latency. Unlike the huge campuses of hyperscale with facilities often covering millions of square feet these smaller data centers are sometimes built into repurposed buildings such as abandoned shopping malls, empty office towers, and factories in disuse, helping requalify ex-industrial brownfield areas.

Leaner centers can be rapidly deployed, located closer to end users for reduced latency, and tailored to specific workloads such as autonomous vehicles and AR.

2. Integration with Renewable and On-Site Power Sources

To address energy demands and grid constraints, future data centers will increasingly be co-located with power generation facilities, such as nuclear or renewable plants. This reduces reliance on strained grid infrastructure and improves energy stability. Some companies are investing in nuclear power. Nuclear power provides massive, always-on power that is also free of carbon emissions. Modular reactors are being considered to overcome grid bottlenecks, long wait times for power delivery, and local utility limits.

Similarly, they will be increasingly built in areas where the climate reduces operational strain. Lower cooling costs and access to water enables the use of energy-efficient liquid-cooling systems instead of air-cooling. We will be seeing more data centers pop up in places like Scandinavia and the Pacific Northwest.

3. Operational Optimization

Artificial intelligence will play a major role in managing and optimizing data center operations, particularly for cooling and energy use. For instance, reinforcement learning algorithms are being used to cut energy use by optimizing cooling systems, achieving up to 21% energy savings.

Similarly, fixes like replacing legacy servers with more energy-efficient machines, with newer chips or thermal design, can significantly expand compute capacity, without requiring new premises.

4. Hardware Density and Efficiency Improvements

Instead of only building new facilities, future capacity will be expanded by refreshing hardware with newer, denser, and more energy-efficient servers. This allows for more compute power in the same footprint, enabling quick scaling to meet surges in demand, particularly for AI workloads. These power-hungry centers are also putting a strain on electricity grids.

Future data centers will leverage new solutions such as load shifting to optimize energy efficiency. Google is already partnering with PJM Interconnection, the largest electrical grid operator in North America, to leverage AI to automate tasks such as viability assessments of connection applications, thus enhancing grid efficiency.

Issues are typically not due to lack of energy but insufficient transmission capacity.

In addition to this, fortunately, data centers are usually running well below full capacity specifically to accommodate future growth. This added capacity will prove useful as facilities accommodate unexpected traffic spikes, and rapid scaling needs without requiring new constructions.

5. Geographically and Politically Responsive Siting

Future data center locations will be chosen based on climate efficiency, grid access, and political zoning policies but also availability of AI-skilled workforce. Data centers aren’t server rooms—they’re among the most complex IT infrastructure projects in existence, requiring seamless power, cooling, high-speed networking, and top-tier security.

Building them involves a wide range of experts, from engineers to logistics teams, coordinating everything from semiconductors to industrial HVAC systems. Data centers will thus drive up the demand for high-performance networking, thermal, power redundancy, and advanced cooling engineers.

Looking to the future

It’s clear that the recent surge in infrastructure demand to power GPUs and high-performance computing, for example, is being driven primarily by AI. In fact, training massive models like OpenAI’s GPT-4 or Google’s Gemini requires immense computational resources, consuming GPU cycles at an astonishing rate. These training runs often last weeks and involve thousands of specialized chips, drawing on power and cooling infrastructure.

But the story doesn’t end there: even when a model is trained, running these models in real-time to generate responses, make predictions, or process user inputs (so-called AI inference) adds a new layer of energy demand. While not as intense as training, inference must happen at scale and with low latency, which means it’s placing a steady, ongoing load on cloud infrastructure.

However, here’s a nuance that’s frequently glossed over in much of the hype: AI workloads don’t scale in a straight-forward, linear fashion: doubling the number of GPUs or increasing the size of a model will not always lead to proportionally better results. Experience has shown that as models grow in size, the performance gains actually may taper off or introduce new challenges, such as brittleness, hallucination, or the need for more careful fine-tuning.

In short

In short, the current AI boom is real, but it may not be boundless. Understanding the limitations of scale and the nonlinear nature of progress is crucial for policymakers, investors, and businesses alike as they plan for data center demand that is shaped by AI exponential growth.

The data center industry therefore stands at a pivotal crossroads. Far from buckling under the weight of AI tools and cloud-driven demand, however, it’s adapting at speed through smarter design, greener power, and more efficient hardware.

From modular builds in repurposed buildings to AI-optimized cooling systems and co-location with power plants, the future of data infrastructure will be leaner, more distributed, and strategically sited. As data becomes the world’s most valuable resource, the facilities that store, process, and protect it are becoming smarter, greener, and more essential than ever.

We list the best colocation providers.

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

Artificial Intelligence (AI)’s role in web hosting is set to redefine the tech industry over the next twelve months, but not in the way you might expect. While current AI web hosting tools are largely focused on front-end solutions like marketing features, such advancements only scratch the surface of what AI can really offer the industry. These AI tools automate basic tasks and provide quick fixes, making them popular, despite failing to address the real challenges.

The real potential of AI in web hosting lies in solving deeper, backend challenges that directly impact a website’s stability, performance and security. In this article, I’ll explore AI’s shift in web hosting from front-end features to backend solutions, that drive real improvements in these areas.

Why AI in web hosting is missing the mark

The race is on to adopt AI – the problem is, in a bid to win, some companies have taken shortcuts to the finish line, adopting only surface level integrations. Let’s take an AI customer chatbot for example. Yes, AI chatbots provide customers with an instance service and can improve customer satisfaction, however they’re rendered useless if your website takes fifteen seconds to load.

According to Cloudways’ research, 60% of UK online shoppers will abandon a purchase if a website takes too long to load or has minor errors, with 27% of shoppers facing such problems in the last 12-months. Customers are facing real problems with slow loading times, and complicated checkout processes.

By only focusing on flashy, front-end solutions, small and medium-sized businesses (SMBs) – which make up 99.8% of the UK business population – are missing a huge opportunity to improve the overarching website experience, fundamental to the customer.

How AI can be used to solve real challenges for SMBs

The majority of SMBs don’t have dedicated website experts; websites are instead managed by a member of the team as a bolt-on to their day job. This setup works fine—until something goes wrong. When issues arise, troubleshooting can take days or even weeks for complex problems, potentially impacting sales. On the other hand, AI can proactively detect, diagnose, resolve, and even prevent website issues from occurring with minimal human intervention.

Modern hosting environments must navigate complex challenges like database optimization, traffic spikes, security threats and bot activity. By leveraging AI-driven analytics and machine learning models, hosting providers can predict potential failures, automatically allocate resources, and mitigate risks before they impact performance. Integrating AI into the process can enable predictive maintenance, reducing server downtime by 30% or greater enhancing the reliability of web hosting services.

AI can also be integrated into intelligent monitoring systems, proactively detecting and resolving technical issues before they impact users. By analysing real-time data, AI can predict server strain, automatically scale resources, and optimize database performance, all without manual intervention. This ensures websites remain fast and reliable, even during high-demand events like product launches or ticket sales. AI tackles the behind-the-scenes challenges customers don’t see, but businesses spend hours trying to fix.

These examples of integrations aren’t flashy front-end solutions but solve real business challenges. They integrate AI to deliver customers with a seamless, reliable experience.

The human-AI partnership: the key to unlocking the future of web hosting

The future of successful web hosting will be shaped by the integration of AI and human expertise. In my opinion, AI should be used to enhance, not replace, human skills, with technical professionals still playing a key role in strategic decision-making and solving complex problems.

The human touch will always be essential for building relationships and understanding unique business needs.

With 80% of businesses in Europe expected to integrate AI-powered software by next year, SMBs will increasingly demand hosting solutions that simplify complexity through intelligent automation. The most successful businesses in the coming months will blend AI efficiency with human insight, allowing time and money to focus on growth and innovation while AI handles day-to-day operations. The industry is evolving from reactive problem-solving to predictive, proactive hosting management.

AI’s true potential in web hosting over the next twelve months lies in tackling fundamental technical challenges rather than just offering surface-level features. Businesses should focus on AI hosting capabilities that deliver measurable improvements in performance, security and reliability. The future of web hosting will be defined by providers who successfully blend AI automation with human expertise, ensuring both the technical side and customer relationships are prioritized.

Organizations should seek hosting partners who use AI to solve real business challenges, rather than simply following technology trends, ensuring they stay ahead in a competitive digital landscape.

We Compiled a list of the best website monitoring 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

• Over 1.6 million files have been discovered online by researchers
• These seem to belong to Etsy, Poshmark, and TikTok Shop customers
• Personally Identifiable Information is included

Two apparently unsecured Azure Blob Storage containers holding a combined 1.6 million files have been discovered by CyberNews researchers, allegedly belonging to online shopping platforms Etsy, Poshmark, and TikTok Shop.

The researchers say these files contained personally identifiable information, such as full names, home addresses, email addresses, and shipping order details.

Anyone who uses these services should keep a close eye on their accounts and take a look at the best identity theft monitoring tools if they are concerned.

Save up to 68% on identity theft protection for TechRadar readers!

TechRadar editors praise Aura's upfront pricing and simplicity. Aura also includes a password manager, VPN, and antivirus to make its security solution an even more compelling deal.

Preferred partner (What does this mean?)View Deal

Customers at risk

Both of the exposed instances “contained shipping email confirmations in HTML format,” researchers confirmed, and the vast majority of users exposed are in the United States, with some from Canada and Australia.

The exact origin or ownership of the datasets is not yet known, but the nature of the information suggests that these belonged to one particular storefront (across multiple shopping platforms), in particular a Vietnamese-based embroidery service.

It’s also not known whether cybercriminals have accessed these datasets, but only an internal forensic audit would reveal this information.

Researchers outlined the risk this brings to those exposed, such as convincing social engineering attacks from cybercriminals posing as Etsy or TikTok shop - urging customers to give their details, resulting in potential financial loss.

“With access to personal information like full names and addresses, attackers could impersonate trusted shipping providers or Etsy itself, making fraudulent communications seem more credible and urging victims to take actions such as confirming personal details, making payment, or clicking malicious links,” the researchers said.

Data leaks are unfortunately all too common for internet users today.

We recommend regularly checking whether your details have been exposed, using services like Have I Been Pwned - and monitoring your accounts, statements, and transactions - and immediately reporting any suspicious or unexpected activity with your bank or credit card provider.

You might also like

• Take a look at our picks for the best malware removal software around
• Check out our choice for best antivirus software
• Adidas confirms customer data stolen in worrying cyberattack

• EA has canceled Cliffhanger Games' Black Panther game and shuttered the studio
• EA Entertainment president Laura Miele told staff that the decision was made to "sharpen our focus and put our creative energy behind the most significant growth opportunities"
• Going forward, EA will focus on franchises like Battlefield, The Sims, Skate, and Apex Legends, while also supporting games like Motive's Iron Man, the third Star Wars: Jedi game, and Mass Effect 5

Electronic Arts has canceled Cliffhanger Games' Black Panther game and closed down the studio.

In an email sent to staff (as seen and reported by IGN), EA Entertainment president Laura Miele shared the announcement and said the decision to cancel the game and close the studio was to "sharpen our focus and put our creative energy behind the most significant growth opportunities."

The number of employees affected by the decision hasn't been confirmed at this time, but IGN understands that the total number of layoffs is less than the 300 roles that were cut last month across Respawn and several other studios.

It's also understood that EA has laid off some individuals from its mobile and central teams.

"These decisions are hard," Miele wrote in the email. "They affect people we’ve worked with, learned from, and shared real moments with. We’re doing everything we can to support them - including finding opportunities within EA, where we’ve had success helping people land in new roles."

Miele's email also mentioned that EA will now focus on a small handful of franchises going forward, like Battlefield, The Sims, Skate, and Apex Legends.

The president also said that EA will continue to invest in Motive's Iron Man game and Respawn's third Star Wars: Jedi game, the Mass Effect 5, while also maintaining its mobile business

Cliffhanger's Black Panther project was announced in 2023 and was being worked on by developers who previously led the Shadow of Mordor series.

Platforms weren't confirmed at the time, and there was no trailer to accompany the game's reveal, but we did know that it was an "original, third-person, single-player Black Panther game".

At the time, project lead Kevin Stephens said the game was going to be a "definitive and authentic Black Panther experience".

"Wakanda is a rich superhero sandbox, and our mission is to develop an epic world for players who love Black Panther and want to explore the world of Wakanda as much as we do," Stephens said.

You might also like...

• Best PS5 games 2025 - top PlayStation 5 titles to play right now
• Sony's Days of Play sale kicks off tomorrow and will bring price cuts to the PS5 Pro, DualSense Edge, the PS5 Call of Duty: Black Ops 6 bundle, and a host of price cuts on accessories and games
• Ghost of Yotei pre-orders live: Collector's Edition up in the US, Standard Edition retailers, and where to buy in the UK

• LexisNexis has suffered a data breach after a cyberattack
• Personal information was taken affecting around 360,000 users
• Not everyone is happy with the response timeline from the organization

Data analytics and risk management firm LexisNexis has disclosed a cyberattack that resulted in data theft affecting 364,333 individuals.

In a notification letter sent to those affected, the company claims an “unauthorized party” gained access to a third-party software development platform and stole the data.

According to the firm, no sensitive personal information was accessed, nor was financial or credit card information, and the organisation’s infrastructure, systems, and products also remain uncompromised.

Save up to 68% on identity theft protection for TechRadar readers!

TechRadar editors praise Aura's upfront pricing and simplicity. Aura also includes a password manager, VPN, and antivirus to make its security solution an even more compelling deal.

Preferred partner (What does this mean?)View Deal

Information affected

“Our Information Security team, in consultation with a forensic firm, immediately began investigating and confirmed that some data which was held in GitHub... was acquired by an unknown third party. Specifically, we have determined that some software artifacts as well as some personal information was accessed” LexisNexis told The Register.

The leaked information includes names, phone numbers, email addresses, home addresses, SSNs, and driver’s license details - enough to spark concern for anyone affected. Take a look at the best identity theft monitoring services if you are concerned.

Not everyone is impressed with LexisNexus’ response timeline, though. Dr Ilia Kolochenko, CEO at ImmuniWeb explains;

“The timeline of the incident detection and disclosure is a bit surprising for a company offering legal and other comparatively sensitive services: the incident reportedly happened in December 2024, was detected in April 2025 after receiving information from the attackers, while disclosed only in May. Given that a lot of personal data was reportedly compromised, the incident detection and response timeline is pretty far from being perfect, to put it mildly.”

“The legal consequences of this incident may cost a lot of dollars to the breach company – being composed of regulatory penalties, legal fees and a likely settlement agreement with the victims. Sadly, as practice demonstrates, the victims will likely get paltry two- or three-digit compensation for the incident in the best-case scenario.”

LexisNexis is far from the first company to be affected by a breach like this, with firms like Co-op and Marks and Spencers offering apologies for the effects of cyberattacks that hit the retailers in May 2025.

You might also like

• Take a look at our picks for the best malware removal software around
• Check out our choice for best antivirus software
• PowerSchool hit by cyberattack which saw student and teacher data stolen

• Garmin is apparently working on a screenless, Whoop-style recovery tracker
• The band is set to launch 'within two months' according to leak site The5KRunner
• Details include vibration alarms, LED battery life indicators and tap gestures – just like Whoop

Yes, it's been rumored before, but it sounds as though we could be seeing a very different style of fitness tracker from Garmin in a couple of months.

Garmin has consistently been known for its smartwatches – it's why we have a dedicated best Garmin watches list, after all – but the company is reportedly set to debut a Whoop-style screenless tracker "within two months", according to leak site The5kRunner.

Having apparently seen a confidential source document, The5KRunner refers to it as a "sleep band" stated to provide comfort during use. This is a nice option to have, as Garmin's chunky watches often aren't ideal for use during sleep. A buckle mechanism and overlapping adjustable strap makes it sound similar to Whoop.

The comparisons don't stop there. Features listed include vibration alarm that wakes the wearer, tap gestures to turn off or snooze the alarm, and an LED battery indicator – all emblematic of Whoop.

Sleep tracking is a given, considering the device is said to pack a smart-wake alarm. Smart-wake alarms, a feature recently introduced in the Garmin Vivoactive 6, only work if the wearable is actively tracking sleep stages via heart rate and other metrics.

Whoop's got some competition

A long time ago, based on my colleague Lance Ulanoff's findings during CES 2023, I predicted that screenless smartwatches would be the future of fitness.

While that hasn't strictly proven true until now, 2025 has seen new screenless optical heart rate monitoring fitness tech from Whoop, Coros, and now (apparently) Garmin, not to mention the advent of the best smart rings.

Fitness tracking is going discreet, focusing on the holistic overall benefits of an active lifestyle rather than specifically tracking workouts.

While the best running watches will always have a place, I think an optical armband-based Whoop competitor from a company like Garmin, with an enormous built-in infrastructure for sleep, health and workout tracking, could be a real winner.

You might also like...

• I used the Garmin Instinct 3, Whoop MG, Apple Watch Ultra 2, and AllTrails all at once on a hike – here's how they compare
• Samsung Galaxy Watch 8 and 8 Classic reportedly certified – but no mention of the Galaxy Watch Ultra 2
• Users of the new Whoop MG fitness band are reporting widespread failures

• Retro compact has no rear LCD and film-style digital filters
• It offers instant one-button Wi-Fi transfer to cloud or mobile app
• Kickstarter campaign launching "mid-June" according to website

The upcoming Echolens is a digital camera that stands out from the pack. Designed to recreate the experience of shooting on analog film, it has no rear screen and built-in film-style filters to create warm, vintage photographs. And the X100VI look-a-like is also going to be really, really affordable.

We’ve seen cameras like the Echolens before. We don’t mean compact, fixed-lens and pricey cameras like the imperious Fujifilm X100VI, the stunning Leica Q3 43 or even the recent fun-oriented Fujifilm X, although there’s definitely a bit of all three in the Echolens’ eye-catching retro styling. We mean in pared-back, cheap point-and-shoots like the Camp Snap, which also has no screen and a refreshingly simple approach to shooting.

While details on its lens, sensor and other specs are yet to be revealed, the Echolens appears to be a little more advanced than the Camp Snap. For one thing, it has an LCD indicator on its top plate to keep count of your shots.

After 54 snaps have been captured, you’ll have to offload them to your smartphone or the cloud via the camera’s built-in Wi-Fi – a feature that’s supposed to 'echo' the act of changing the roll of film in your analog camera. This offloading apparently takes just one button press and no time at all.

The bottom plate features a shot count indicator and image transfer button. (Image credit: Echolens)Stripped-back shooting

The Echolens website promises a minimalist approach to photography, stripping away much of the tinkering that can get in between the photographer and their subject.

That means simplified controls (there’s a shutter button, an image transfer button and a dial to switch the flash on and off, and that seems to be it), a viewfinder for composition and nothing in the way of AI image enhancement or over-sharpening. Instead, the user can pick from a range of digital ‘film’ styles designed to replicate the color profiles of classic 35mm stocks.

A USB-C port is also present for recharging, and Echolens’ makers claim the battery will last for several days on a full charge.

The Echolens' top plate has a shutter button and a flash control – nothing else. (Image credit: Echolens)

While we really need more detail before we can get too excited about this camera – and all the usual crowdfunding caveats still apply – we hopefully won’t have to wait long. Echolens is slated for a launch on Kickstarter in mid-June 2025, with a full release to follow at an unspecified later date.

What we do know, roughly at least, is the price: the Echolens site’s FAQ says the camera will be around $150 (or around £110 / AU$230), but early bird backers on Kickstarter will be able to secure a pre-order for under $100 (about £75 / AU$155). Which seems like a brilliantly affordable price tag if this thing lives up to its billing.

Of course, backers sometimes have a lengthy wait for the finished product after a Kickstarter campaign closes, so we’ve reached out to Echolens to try and find out more about their camera and campaign. We’ll update this story when we have further info.

The flash dial simply toggles between on and off, apparently. (Image credit: Echolens)You might also like

• Fujifilm’s X half is a tiny retro compact that’s big on wacky film photography features, and I love it
• Pentax 17 review: a modern analog charmer
• Fujifilm’s affordable Instax Mini 41 impressed me with its elegant design and simple operation

• Windows 11 22H2 and 23H2 have been hit by a bug that can leave a PC unable to boot
• Microsoft is investigating this recovery error glitch in the May update
• Windows 11 Home users are unlikely to be affected, we’re told, but that doesn’t mean they can’t run into this showstopping affair

Those who haven’t yet installed the latest cumulative update for Windows 11 in May – by which I mean the upgrade released the week before last, not the most recent optional update – should be aware of a nasty bug that some users are encountering.

Microsoft has said that it’s investigating whether the May update can cause a showstopping error, which means your PC won’t boot. However, this warning is delivered with a welcome caveat that it isn’t likely to affect Windows 11 Home users. Still, that possibility isn’t ruled out, and given the gravity of the bug, it’s worth bearing in mind if you’re about to pull the trigger on this cumulative update.

Bleeping Computer reports that Microsoft has updated its release health dashboard with information on the bug, and how the upgrade might fall over with a ‘recovery error 0xc0000098 in ACPI.sys’ message.

This is no simple error or installation failure, though, because it leaves Windows 11 unable to boot and requires a repair to be run on the system. And whether that repair will be successful is anyone’s guess, as sometimes, these are the kind of situations that are tricky to recover from – and the OS won’t necessarily pull itself out of the hole that’s been dug.

The good news is that affected devices running Windows 11 22H2 and 23H2 – note that 24H2 is not affected here – are few in number. Primarily, this problem pertains to virtual machines, not physical Windows 11 PCs.

Microsoft explains: “Home users of Windows using Home or Pro editions are unlikely to face this issue, as virtual machines are mostly used in IT environments.”

However, being ‘unlikely’ does not equate to being impossible, and so there must have been some reports of Windows 11 PCs (as opposed to virtual machines) being affected by this unfortunate issue.

I did find one such recent report on Reddit, although it isn’t explicitly linked to the latest update being installed, so take that with a grain of salt. The error message was slightly different as well, but in its warning, Microsoft acknowledges that a file name other than ACPI.sys might appear in the error.

Analysis: A cautious approach may be wise here

(Image credit: MAYA LAB / Shutterstock)

Microsoft wouldn’t announce such an investigation lightly, so if you haven’t yet installed the May update, I think that it’s worth considering delaying the installation of that upgrade for a little longer. Mainly to allow for Microsoft to complete its investigation here, after all, you’ve waited a couple of weeks (for whatever reason) now, so what’s another week, or perhaps a bit longer?

It is possible to pause OS updates for over a month, even on Windows 11 Home, though you’ll be forced to apply any given patch after 35 days of deferring it.

In this additional time, you can wait for Microsoft to complete its deeper dive into what’s going on here, and hopefully that’ll lead to a fix being implemented soon, too.

Granted, you’d probably be very unlucky to encounter this recovery error, but even a small chance is off-putting, given that it could be a potentially thorny problem to extricate yourself from. Ultimately, the choice is yours, and if you’ve already installed the May update without encountering any such error, I should make it clear that you've got nothing to worry about.

Microsoft isn’t having a great time of things lately when it comes to bugs of a seriously unpleasant nature, as there was a real sting in the tail for some Windows 10 users with the May update, too.

You might also like...

• Hate Copilot? You might change your mind if this hint that it’ll extend your laptop’s battery life is true
• Can’t upgrade to Windows 11? This Linux project wants to save your old PC from the scrapheap when Windows 10 support ends
• Been hiding from Windows 11 24H2 due to the fuss about all the bugs? There’s nowhere to run now as Microsoft’s made the update compulsory

The 2025 edition of Netflix Tudum, aka the now-annual fan event that celebrates everything about the world's best streaming service, is almost upon us.

This year's installment will take place on Saturday, May 31 at 5pm PT / 8pm ET – non-US readers can work out when it'll begin using the aforementioned times – and there'll be plenty of announcements made on the night.

We already know that TV hits including Stranger Things, Wednesday, and Squid Game will be featured at Netflix Tudum 2025. However, Netflix has stayed tight-lipped on what it'll actually reveal during its live event, which you can watch on Netflix if you didn't manage to secure a ticket to the in-person event.

That won't stop me from speculating on what'll be shown, mind you. With that in mind, here are four things I predict we'll see at Tudum 2025.

1. Stranger Things season 5 release date and format, plus a teaser trailer

Me and who as we sit around our computers hoping for some Stranger Things 5 news on Saturday? (Image credit: Netflix)

As far as many people are concerned, these are the biggest reveals that Netflix could make at Tudum 2025.

We already know that the hugely popular show's final season is set to be released sometime in 2025. Indeed, filming wrapped on Stranger Things 5 in December 2024 but, while a big Stranger Things season 5 leak seemingly suggested it'll air in late 2025, we still don't know when it'll make its actual debut.

Nevertheless, Tudum 2025 is the perfect time for Netflix to finally confirm when Stranger Things season 5 will be released. I predict that the streaming giant will also tell us whether it'll launch in full, or be released in two parts like Stranger Things season 4 was.

Lastly, I'd be surprised if a teaser trailer isn't unveiled to further generate hype for one of the best Netflix shows' final hurrah. We've had to live on behind-the-scenes morsels for years at this point, so some actual footage of season 5 would go down a storm.

2. An official trailer for Squid Game season 3

Unchain Squid Game season 3's official trailer, you cowards! (Image credit: Netflix)

It's hard to believe that another incredibly successful Netflix TV Original – Squid Game – will end this year, too.

The Korean-language survival drama is set to end when Squid Game season 3 arrives on June 27. But, despite the fact that we were treated to an exciting Squid Game 3 teaser in early May, an official trailer still hasn't been released online.

Tudum 2025 is the ideal time to drop a much longer trailer that'll make me and millions of other Netflix users even more stoked for the series' last chapter.

3. More footage of Wednesday season 2

This is the exact same pose and facial expression I'll have if we don't see more of Wednesday season 2 (Image credit: Netflix)

Unlike the aforementioned pair, Netflix hasn't been shy about disclosing information about Wednesday season 2.

In the last few weeks alone, we've received word about Wednesday 2's dual release dates and first-look at its new cast members, and been treated to a finger-clickingly good behind-the-scenes featurette of the show's sophomore outing.

I'm hungry for even more, though, so I'd like to see another first-look clip or two for Wednesday's second chapter. I don't really care what's shown, either. Indeed, I'll take an extended sequence of Jenna Ortega's eponymous character interacting with Joanna Lumley's Grandmama, a tension-filled reunion with Hunter Dooham's Tyler/Hyde, or something more kooky. Just give me something, please Netflix.

4. Release date reveals for Netflix's 2025 movie line-up

"No Knives Out 3 release date yet? It makes no goddamn sense! Compels me, though..." (Image credit: https://x.com/rianjohnson)

It's not all about the biggest shows that Netflix has to offer. There are numerous new Netflix movies that'll launch on the platform before the end of 2025, but lots of them still only have placeholder launch windows.

There's no better moment to announce some proper release dates for some of Netflix's most anticipated new movies of the year, then, than at Tudum 2025. I'm hoping for official launch dates for Wake Up, Dead Man: A Knives Out Mystery, Guillermo del Toro's Frankenstein, The RIP, and The Twits, but I won't turn my nose up at any other major reveals that are made.

What are you hoping to hear more about at Netflix Tudum 2025? Will you be tuning in to watch it live, or do you plan on catching up on the biggest news once it's wrapped up? Let me know in the comments!

You might also like

• These are the best Netflix movies to stream today
• 'Is this for real?': Lilo and Stitch fans are furious about the live-action remake's ending, so they're watching the original movie on Disney+ instead
• Avengers: Doomsday could spell the end for another original Avenger as Marvel actor's thank you video sparks MCU exit rumors

• Doda, a district in India's Jammu and Kashmir region, issued a 2-month VPN ban over national security concerns
• The ban was enforced on May 2, 2025, with some VPN users reportedly detained after that date
• Experts are warning against the attack on people's digital rights, while questioning the legality of the ban

Doda, a district in India's Jammu and Kashmir region, enforced a 2-month VPN ban on May 2, 2025, over national security concerns.

Authorities issued the ban under Section 163 of the new BNSS law (the equivalent of the Indian Penal Code), citing individuals and groups misusing virtual private network (VPN) tools "to circumvent lawful cyber restrictions and access prohibited applications," the order reads.

Digital rights and legal experts have so far strongly criticized this decision. They warn against the attack on people's digital rights and security, while questioning its legality.

Is Doda's VPN ban legal?

The Doda's VPN ban aims to "curtail the use of VPNs in the interest of public safety," and it's extended to all individuals, institutions, internet service providers, and cyber cafes operating in the district.

This means that all residents are barred from using one of the best VPN services or similar tools for two months, to avoid being liable for legal action.

Repercussions for allegedly using VPNs seem to have already begun. In a report published on May 18, Indian publication The Wire reported allegations that an unspecified number of Doda residents under "technical surveillance" have already been arrested as found to be using VPNs.

Two days before, Doda's authorities had already confirmed to have "detained several individuals across the district for violating the directive."

This is why many experts have been raising concerns about how the ban can impact people's rights, like free speech, free access to information, and privacy.

TechRadar needs you! We want to know what you think about the world of VPNs. Whether you're a novice or a VPN pro, we want to hear your thoughts. Don't worry, though, your responses are completely anonymous, and it takes less than a minute to complete!

To take part, click the link below: https://futurenet.questionpro.eu/tr-vpn

Talking to TechRadar, Senior Policy Counsel and Encryption Policy Lead at Access Now, Namrata Maheshwari, explains that the right to the internet and the right to privacy are protected fundamental rights under the Indian Constitution.

Maheshwari also pointed out how the Indian government has already implemented "an onerous mandate" for VPN providers when it enforced a new data retention law in 2022.

A host of online services are now required to store users' data for up to five years and share it with authorities on request, in fact, something that led to an exodus of major VPN companies from the country.

VPNs are not illegal."

Namrata Maheshwari, Access Now

Yet, "VPNs are not illegal," said Maheshwari, adding that a blanket ban on all VPNs for two months is "unnecessary and disproportionate."

"People have a fundamental right to information and freedom of expression, and restrictions on these rights ought to be legal, necessary, and proportionate – a standard this ban does not meet," she added.

The ban isn't affecting only individual users, though, but also businesses and institutions that regularly use these services for security purposes.

According to Maheshwari, the ban could then end up increasing vulnerabilities in the country's cyberspace, ultimately weakening security as opposed to strengthening it.

She said, "Even in an emergency, preventative measures cannot be this vague and overreaching, particularly where less intrusive means are available."

You might also like

• India ordered X to block 8,000 accounts – here's what we know so far
• Proton Mail hit with blocking order in India
• VPNs are disappearing from India's app stores – and a 2022 law may be the culprit

• World War Z VR, a new first-person, single-player game, is coming to SteamVR and Meta Quest on August 12
• Saber Interactive's new game will allow players to take on hordes of over 200 zombies at once in VR
• Pre-orders are now available on the Meta Store and Steam

Saber Interactive has announced the next game in its World War Z franchise, World War Z VR, and it's launching this summer.

World War Z VR is a brand new first-person, single-player game based on the Paramount Pictures 2013 film and Saber’s game franchise, and is set to release on August 12 for SteamVR, Meta Quest 3, Meta Quest 2, and Meta Quest Pro.

"World War Z VR faithfully reimagines the intense action of the original World War Z game to create the most immersive VR zombie experience ever," Saber said. "Take on hordes of over 200 zombies on screen at a time, an astounding technical feat for VR which will put your skills to the ultimate test."

In this new installment, players will find themselves taking on hordes of zombies in three locations: New York, Tokyo, and Marseille. There will be seven playable characters to choose from, and after selecting their operative, players can start missions with a pre-selected AI squad, equipped with a wide variety of weapons and perks.

The VR missions will present unique challenges for players and even offer three difficulty modes to choose from. Zombies will also adapt to their surroundings and recognize the players' playstyle to hunt them down, while special infected enemies raise the stakes even higher.

Pre-orders are now available on the Meta Store and Steam for $19.99 / £14.99. Those who secure a pre-order will also gain access to the Golden Skin Pack, which features four exclusive gold weapon skins.

Steam users can also pre-order a special game bundle that includes both the new World War Z VR and the original World War Z, featuring a 20% discount.

You might also like...

• Sony's Days of Play sale kicks off tomorrow and will bring price cuts to the PS5 Pro, DualSense Edge, the PS5 Call of Duty: Black Ops 6 bundle, and a host of price cuts on accessories and games
• Leaker claims Sony is now working on a new PlayStation handheld console
• Ghost of Yotei pre-orders live: Collector's Edition up in the US, Standard Edition retailers, and where to buy in the UK

• A new update for the Philips Hue mobile app is rolling out now
• The app now suggests adding new sensors and switches to a zone during the setup process
• This avoids a problem where lighting scenes are unavailable later on

If you've ever struggled to set up new sensors and switches for your Philips Hue lights, help is at hand. A new app update is rolling out now that automatically suggests adding accessories to a zone when you connect them for the first time, so you don't miss out on any features later on.

A zone is a set of lights and other accessories that you want to control together as a group. They don't necessarily have to be in the same place or include all the lights within a room. For example, you might want to create a zone specifically for your home entertainment lighting, or one that includes all the lights on a particular floor.

Previously, assigning new devices to a room or zone was an optional step within the Philips Hue app, which you might miss if you were in a hurry. If you did, you wouldn't be able to use all of the switch or sensor's functions afterwards. For example, you may be unable to view all the available scenes (ready-made lighting profiles).

The new app update was spotted by the editor of Hueblog.com, who says he's often contacted by readers stuck in this frustrating 'accessory trap'. From now on, the app will suggest creating at least one zone during the setup process so you don't suffer that problem, and will note that you can always hide zones in the future if you prefer.

Simpler setup

There is also a change to the setup process when you're installing a new Philips Hue security camera. As Notebookcheck explains, the app now offers you the chance to set up motion-controlled lighting during the initial setup process.

Both of these are relatively minor changes, and there's still no news about the upcoming Philips Hue AI assistant. However, they address some common problems that you might face when extending your Philips Hue setup and make your best smart lights even better.

You might also like

• Nanoleaf's new indoor and outdoor smart lights are its funkiest designs to date
• Philips Hue bulb types explained: choose the right lights for your home
• If the Philips Hue Play HDMI Sync Box is too expensive, this new budget-friendly alternative from WiZ could be exactly what you need

• NordVPN has launched post-quantum encryption (PQE) across all its applications
• The provider launched its first iteration of PQE in 2024, but only for its Linux VPN app
• This comes as traditional encryption methods are set to become obsolete with the advent of quantum computers

NordVPN has just taken an important step forward into its post-quantum transition, extending support for quantum-safe encryption across all its applications.

While the provider launched its first iteration of post-quantum encryption (PQE) in September 2024, only Linux users could enjoy this protection. Now, all NordVPN users can enjoy PQE no matter the platform they use.

Already rated as the best VPN service on the market by TechRadar's reviewers, NordVPN joins a handful of VPN providers that also include quantum-safe cryptography in their software.

Why and how use NordVPN post-quantum encryption

As quantum computing advances, NordVPN's CTO, Marijus Briedis, explains, RSA-based key exchange encryption methods that most VPN protocols use today will eventually become vulnerable.

Experts predict, in fact, that traditional encryption will soon be broken by the ability of quantum computers to perform computations that machines can't handle, within minutes.

While this is expected to happen between five to 10 years, cybercriminals have already started to conduct what's known as "store now, decrypt later (SNDL) attacks." The idea is simple – to collect encrypted data now so they can crack it in the future when quantum computing is finally up to the task.

This is exactly why the cryptographic world is already catching up. Last August, the National Institute of Standards and Technology (NIST) released the first three quantum-resistant encryption standards to better support software providers in their transition. NordVPN joined the post-quantum revolution a few months after that, adding one of these algorithms (ML-KEM) to its WireGuard-based Nordlynx protocol, which is known for its speed and security.

"By integrating PQE into our VPN infrastructure, we're taking a proactive step to ensure long-term confidentiality and resilience for our customers' data, both now and in a post-quantum future," said Briedis.

TechRadar needs you! We want to know what you think about the world of VPNs. Whether you're a novice or a VPN pro, we want to hear your thoughts. Don't worry, though, your responses are completely anonymous, and it takes less than five minutes to complete!

To take part, click the link below: https://futurenet.questionpro.eu/tr-vpn

As mentioned earlier, before the latest release, only the Linux VPN app included PQE support.

This choice allowed the NordVPN team to analyze how the new quantum-safe algorithm, which was implemented in a hybrid way alongside classic encryption methods, could affect performance like connection speeds and latency.

"The gathered data served as a stepping stone in the transition to quantum-resistant encryption for the rest of our platforms," said Briedis. "The Linux case demonstrated that we successfully maintained the highest level of user experience in terms of connection time and speed during the transition."

As a result, NordVPN rolled PQE across all its platforms – from Windows and macOS to Android and iOS, even including its Android TV and tvOS apps.

Users need to actively enable this extra protection by heading to the Connections tab in the app's Settings and switching on the toggle. The PQE feature will automatically activate whenever the user connects via the NordLynx protocol.

Which VPN providers offer post-quantum encryption?

NordVPN is not the only virtual private network (VPN) provider currently including PQE support on its service.

ExpressVPN added quantum-safe encryption on its proprietary Lightway protocol in October 2023, by default across its Android, iOS, Linux, Mac, and Windows apps. Last January, the team took a further step forward by upgrading to the NIST standards.

After pioneering the post-quantum transition in 2017, Mullvad extended PQE support for all platforms when it released post-quantum protection for iPhone in September 2024. The provider now plans to migrate to the new NIST standards, too.

AdGuard, Windscribe, and PureVPN are other providers currently offering PQE protection, with Surfshark and NymVPN also expected to join the list soon.

You might also like

• VPNs need to be ready for the post-quantum world – here's what's being done
• Is the internet ready for the "biggest security threat of all time"?
• NordVPN has now a full GUI Linux VPN app

• Google found Chinese hackers abusing Google Calendar
• The service was used to host malicious instructions and to exfiltrate results
• ToughProgress campaign was carried out by Chinese state-sponsored hackers APT41

Chinese state-sponsored hackers known as APT41 have been seen abusing Google Calendar in their newest attacks, using it as part of the C2 infrastructure.

Google’s Threat Intelligence Group (TIG) recently discovered the technique, dismantled the setup, and introduced changes to prevent similar attacks in the future.

The attack starts from a previously compromised government website - TIG did not explain how the site was compromised, but said it was used to host a .ZIP archive. This archive is then shared, through phishing emails, with potential targets.

Reading the calendar

Inside the ZIP are three files: a DLL and executable files posing as JPGs, and a Windows shortcut file (LNK) posing as a PDF document.

When the victim tries to open the fake PDF, it runs the shortcut which, in turn, activates the DLL.

This file, in turn, decrypts and launches the third file, which is the malicious payload dubbed “ToughProgress”.

The malware then reads additional instructions shared in two specific events in Google Calendar. The commands are found either in the description field, or hidden events.

To share the results, the malware would create a new zero minute calendar event on May 30, and share the data, encrypted, in the calendar event description.

Since the malware is never actually installed on the disk, and since the C2 communication happens via a legitimate Google service, most security products will have trouble spotting the attack, Google suggests.

To tackle the threat, TIG developed custom detection signatures to identify and block APT41’s malware. It also took down associated Workspace accounts and calendar entries. Furthermore, the team updated file detections and added malicious domains and URLs to the Google Safe Browsing blocklist.

Google also confirmed that at least a few companies were targeted: “In partnership with Mandiant Consulting, GTIG notified the compromised organizations,” it said.

“We provided the notified organizations with a sample of TOUGHPROGRESS network traffic logs, and information about the threat actor, to aid with detection and incident response.”

It did not say how many companies were affected.

Via BleepingComputer

You might also like

• Google warns of legit VPN apps being used to infect devices with malware
• Take a look at our guide to the best authenticator app
• We've rounded up the best password managers

• The Aviot WA-J1 will ship worldwide in July
• Three drivers combined: piezo, planar and dynamic
• Price converts to about $270 / £200 / AU$424

The Japanese audio brand Aviot has introduced what it says is the world's first "Tribrid" 3-driver headphones, the WA-J1 – and it's made a special edition in collaboration with the J-pop star Hyde.

What's a Tribrid? According to Aviot, it's one piezo driver, one planar magnetic driver, and one dynamic driver, and together they deliver "overwhelming information density and unprecedented acoustic performance".

And they do it for an incredibly long time: battery life is a claimed 140 hours, though it's not specified if that's with active noise cancellation or not (I have to assume not).

(Image credit: Aviot)When three drivers are better than one

As Aviot explains, the piezo driver is there for the high frequencies, and the planar and dynamic drivers are placed coaxially with it to take care of the mids and lows, all with very low distortion and very little phase disturbance. The result, the firm says, is very high precision.

Supported audio formats include LDAC and Bluetooth LE Audio (that latter one via a future update) as well as the obligatory AAC and SBC, and there's adaptive noise cancelling with a transparency mode.

The headphones also offer 3D spatial audio. That option is adjustable via the control dial, enabling you to spatialize audio to suit your personal preference – from very flat to very reverberant sonic spaces. And for gamers, there's a low-latency mode to reduce lag.

The headphones will go on sale in Japan in July 2025 with a price tag of 39,600 yen – roughly $270 / £200 / AU$424 – and will be available to ship worldwide.

You might also like

• The best headphones 2025 tested by our experts, for all budgets
• The best over-ear headphones in 2025 for all budgets, tested by us
• The best wireless headphones to suit every listener and budget

Thanks to intensified activity on the part of law enforcement agencies worldwide – resulting in the shutdown of highly profitable and destructive ransomware-as-a-service (RaaS) groups such as LockBit – we’ve seen some encouraging signs within the ongoing fight against ransomware threats. Annual ransomware payment totals fell from $1.25 billion in 2023 to $813.55 million last year.

Meanwhile, the global year-to-year rise in the rate of these attacks is on the decline, with 5,289 incidents in 2024, up 15 percent from 4,591 in 2023. But, that percentage increase is significantly lower than the 77 percent increase seen in 2023 (2,593 attacks in 2022). So, the overall incident growth rate appears to be leveling off somewhat.

However, it’s no time to get complacent. In our own research, we’ve found that ransomware variants grew to 101 in 2024, up from 70 in 2023. This rise in variants remains a foreboding indicator of accelerated attacks to come, signifying rebranding efforts on the part of cybercriminals. They’re responding to law enforcement activity with craftier variants and more precisely targeted campaigns aimed at victims with deep pockets. As a result, they’re boosting their efficiency and efficacy.

Ransomware-as-a-service

With the rise of RaaS over the years, cybercriminals now work much like a business, leveraging a subscription-based market in which customers (most commonly referred to as affiliates) pay for software created by ransomware operators, for the purpose of launching attacks.

By its very nature, RaaS has lowered the barrier to entry because customers no longer need to be experts in coding. They can simply pay for the product to launch their ransomware campaigns, and their subscriptions typically entail full-service offerings of software support, malware and infrastructure. Such a model created a strong incentive for more cyber actors, who are perhaps less skilled or experienced, to join the ransomware landscape, as long as they agreed to share a cut of the ransom with the RaaS operator, of course.

Key trends

If this sounds like a high-level, enterprise approach to crime, well, that’s because it is. Subsequently, we’re seeing the following key trends reshaping the very essence of the ransomware experience in 2025 – trends that can help victims make better decisions as they assess whether to proceed with payment negotiations or not:

(Dis)honor among thieves

Among cyber attack techniques, ransomware stands out because there are inevitably person-to-person interchanges, those which often touch upon human emotions and elements of trust. Or mistrust.

Victims, already rattled by the reality of the situation, find themselves in dialogue with the ransomware operator and must evaluate whether the operator will make good on terms if the ransom is paid. “What do we know about this group?” victim organizational leaders will ask. “Do they seem too aggressive? Are they pressuring us without any intention of deleting the data they stole, or never bothering us again? Or do they have a track record of respecting negotiated agreements?”

The weaponization of data

RaaS groups aren’t just stealing victims’ data. They’re weaponizing it as a means to increase the pressure to pay. They will now go through financial records, cyber liability limits and additional information to determine whether a victim is positioned to pay – and how much.

Artificial intelligence (AI) tools enable them to explore further at a larger scale, such as the exploitation of human resources (HR) records to reach employees and/or senior executives and tell them that their records have been compromised. With this, the RaaS operator tries to apply pressure on employees and executives to convince their organization to pay the ransom.

Outsourcing and automation

RaaS operators are fully capable of using the same tech and personnel management tools as legitimate companies. Outsourcing allows them to rebrand quickly if they feel the heat around the corner. They can put people and resources in place with a spinoff group and then migrate to the new brand when they have to make the current one “disappear” to avoid getting arrested.

Automation will enhance efficiencies and expand the scale of victim targeting and payments. Instead of communicating with targeted organizations person-to-person, a chatbot instantly will direct organization members to a log-in page, and they’ll use a transaction number to process payment and secure the stolen information.

Deal or no deal?

Ultimately for victims, ransomware comes down to one simple proposition: Deal or no deal? This is a highly individualized decision, based upon the size of the organization, industry, and the potential consequences of taking a “no payment” position.

But the more companies know about the RaaS operator, the better informed their choices. If they know the background of the operator and how adept they are at deploying AI and additional resources to “go deep” within their environment – then they’ll have a stronger sense of likely outcomes of paying or not paying. If they’re aware that the “new” group which compromised them is actually a spinoff of an old one, then they can research the reputational history of the former brand.

Given that AI and automation should only dramatically expand the capabilities of the RaaS underworld as a collective whole, organizations will need to stay on top of these trends to make the best decision for their teams.

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

• The Order: 1886 creative director Ru Weerasuriya has revealed that the game was supposed to be the first game in a trilogy
• The Order 1891 and The Order 1899 were planned, and the second game was in the early stages of development before it was canceled
• The sequel would have featured bigger fight sequences and a multiplayer mode that was cut from the first game

The PlayStation 4 exclusive The Order: 1886 was originally supposed to be the first game in a trilogy.

That's according to The Order: 1886 creative director Ru Weerasuriya, who revealed in a new French-language interview with YouTuber Julien Chieze that two more games were planned: The Order 1891 and The Order 1899 (via VGC).

Weerasuriya said the second game was in the early stages of development before it was canceled and that he wrote a 10-page pitch for the sequel, which would have featured bigger fight sequences, as well as a multiplayer mode.

It was also revealed that a multiplayer mode was in development for the first game, but it was shelved for the sequel.

Although the third game never made it to development, Weerasuriya said that he planned for where the story would go if developer Ready at Dawn had been able to complete the trilogy, and that other installments in the series would have taken the franchise to the 20th century.

While it's never been revealed why The Order: 1886's sequel was canceled, Weerasuriya suggested that the first game's mixed critical response was a factor.

The Order: 1886 was released in 2015 and is set in an alternate version of Victorian London, "where science and myth collide". It's been more than 10 years since the game made its PS4 debut, and although it's playable on PS5 through backward compatibility, The Order: 1886 is still exclusive to PlayStation.

You might also like...

• Sony's Days of Play sale kicks off tomorrow and will bring price cuts to the PS5 Pro, DualSense Edge, the PS5 Call of Duty: Black Ops 6 bundle, and a host of price cuts on accessories and games
• Leaker claims Sony is now working on a new PlayStation handheld console
• Ghost of Yotei pre-orders live: Collector's Edition up in the US, Standard Edition retailers, and where to buy in the UK