TechRadar News

• The latest Oppo Find N5 image shows a near-invisible crease
• We should see the phone called the OnePlus Open 2 internationally
• A launch is happening within the next two weeks

We know that we're just days away from the launch of the Oppo Find N5 – which will almost definitely be available in some markets as the OnePlus Open 2 – and a new teaser image of the foldable hints at one way it could best the Samsung Galaxy Z Fold 6.

The image was posted by Pete Lau, Chief Product Officer at Oppo and previously the founder of OnePlus. It shows the Oppo Find N5 on the left, up against an unnamed foldable – which looks very much like the Galaxy Z Fold 6 launched last July.

When it comes to crease visibility, the Oppo Find N5 is the clear winner. This is something Oppo and OnePlus have form for too: in our OnePlus Open review, we praised the foldable for having a screen crease that was "nearly invisible and barely tactile".

It would seem the Oppo Find N5 and OnePlus Open 2 will be even better in this regard. It looks as though the phone will outdo the Galaxy Z Fold 6 in terms of screen size too – the Samsung handset has a main display measuring 7.6 inches, corner to corner (and the OnePlus Open had a 7.82-inch panel).

Coming soon

IPX6 ✅ IPX8 ✅ IPX9 ✅Immensely durable, rain or shine. #OPPOFindN5 pic.twitter.com/Yg0G3Qm8Z0February 7, 2025

Lau has also been promoting the improved waterproofing on the upcoming Oppo Find N5. The phone is apparently "immensely durable, rain or shine", and should offer more protection against the elements than the Oppo Find N3 (and OnePlus Open).

We've already heard from other Oppo executives that waterproofing is going to be one of the key features on the upcoming foldable phone. Much has been made of its thinness too: even thinner than an M4 iPad Pro, apparently.

Camera upgrades over the previous model have also been rumored, and if you add in a processor bump as well, this sounds like it could be a contender for the top spot in our best foldables list – but we'll have to wait and see.

Oppo has given us a rough idea of when its Find N5 handset will make an appearance – around February 19 or February 20 – but we don't have an exact date yet. The phone is getting a global launch, but certain parts of the world should also see the device appear as the OnePlus Open 2 later in 2025.

You might also like

• Our full review of the Samsung Galaxy Z Fold 6
• 5 things we want from the OnePlus Open 2
• Apple’s foldable iPhone specs hinted at in fresh leak

• A new feature could almost miraculously cut energy consumption of data centers by 30%
• Interrupt request suspension dynamically alters CPU power usage and can be done via OS
• Hyperscalers are likely to be the big winners and it will be interesting to see how it impacts AI

Data centers reportedly account for between 2-4% of total electricity consumption around the world, something hyperscalers are understandably looking to reduce wherever possible.

Potential solutions include implementing next-generation architectures like Hyperconverged Infrastructure (HCI) and employing advanced cooling techniques.

Professor Martin Karsten at the Cheriton School of Computer Science, within the University of Waterloo in Ontario, Canada, has a cheaper, easier solution. He claims that data center energy consumption could be cut by up to 30%, simply by changing a few lines of Linux code.

Small change, big impact

Working with Joe Damato at Fastly, Professor Karsten has developed a small, non-intrusive kernel change of just 30 lines of code that uses IRQ (interrupt request) suspension to reduce unnecessary CPU interruptions and improve Linux’s network traffic processing. This tweak has now been published as part of Linux’s newest kernel, release version 6.13.

This code change, which reportedly improves Linux networking efficiency and boosts throughput by up to 45% without increasing latency, is based on a research paper called “Kernel vs. User-Level Networking: Don’t Throw Out the Stack with the Interrupts”, which Professor Karsten authored with former master’s student Peter Cai in 2023.

"We didn’t add anything,” Professor Karsten said of the code change. “We just rearranged what is done when, which leads to a much better usage of the data center’s CPU caches. It’s kind of like rearranging the pipeline at a manufacturing plant so that you don’t have people running around all the time.”

The Professor believes this small adjustment could have a huge impact. “All these big companies - Amazon, Google, Meta - use Linux in some capacity, but they’re very picky about how they decide to use it. If they choose to ‘switch on’ our method in their data centers, it could save gigawatt-hours of energy worldwide. Almost every single service request that happens on the Internet could be positively affected by this.”

Aoife Foley, IEEE senior member and professor in the School of Mechanical and Aerospace Engineering at Queen’s University Belfast, welcomes the potential savings but observes that it’s going to take a lot more than just changing a few lines of code to address the broader energy challenges.

“There is a long way to go yet," she says. "These facilities represent huge electricity demands, adding pressure to electricity grids and increasing the challenge of energy transitions, especially in smaller countries. Although it is impossible to calculate precisely, the entire ICT sector is estimated to account for about 1.4 per cent of CO₂ emissions globally. Infrastructure and operations leaders have a responsibility here and need to consider the unnecessary waste associated with data storage and commit to generating power from more renewable sources.”

Yandex recently released an open source tool called Perforator, which takes a similar approach to Professor Karsten’s research, helping businesses optimize their code, reduce server load, and ultimately lower energy and equipment costs.

Sergey Skvortsov, who leads the team behind Perforator, told us: “This latest research confirms what we have long believed: optimizing code is one of the most effective ways to reduce data center energy consumption. Perforator helps businesses identify and fix inefficient code, cutting CPU usage by up to 20% and reducing infrastructure costs - without sacrificing performance. With data centers consuming up to 4% of global electricity, tools like Perforator can play a crucial role in making tech infrastructure more sustainable.”

You might also like

• Why the data center of 2030 will look very different from today
• Data centers are transforming waste heat into community energy assets
• The AI boom is spurring an energy crisis: is there a sustainable way out?

Google really wants you to know that Gemini 2.0 Flash is a big deal. It's the first generally released member of the Gemini 2.0 family, and the company is hyping it up as faster, smarter, and more powerful than ever, hinting without outright claiming it should replace OpenAI’s ChatGPT in people's hearts and smartphones.

Gemini 2.0 Flash is essentially a high-efficiency version of the model that is faster but supposedly maintains as good an accuracy rate. I decided to play around with Gemini's new model, testing it out in some basic ways the average person would use it. To spice things up, I sent the same prompts to ChatGPT.

Working it out

(Image credit: Screenshots of Google Gemini and ChatGPT)

I started with a prompt about fitness as it's the kind of thing a lot of people turn to the internet to help with. I think a good AI chatbot should not only suggest a reasonable workout routine but also explain the reasons for its suggestions, variations on the exercise, and more information about how to be healthy and avoid injury.

I wanted to see which chatbot would provide the most well-rounded response, but both honestly did quite well. Gemini included links and tips for its workout, while ChatGPT had a whole weekly schedule for workouts and its own set of helpful advice on getting in shape safely. I had to call it a tie, though the multi-day schedule from ChatGPT did stand out somewhat.

Translation Total

(Image credit: Screenshots of Google Gemini and ChatGPT)

(Image credit: Screenshots of Google Gemini and ChatGPT)

Translations are always a popular use of AI chatbots, and this prompt covers three vastly different languages: one classical, one historical, and one fictional. Latin requires proper grammatical structure, while Middle English isn’t just modern English with “ye” thrown in; it has distinct grammar and vocabulary. Of course, Klingon follows an entirely different syntax, created artificially with its own fake etymology. A smart AI should recognize these differences and ask for and provide natural-sounding translations of a common saying instead of just swapping words.

ChatGPT broke down each translated phrase and explained the word comparisons, including references to changes in phrasing. Gemini was more concise but included explanations for each translation in terms of alternate phrasing and spelling, as well as calling out when it might be a bit too literal. The Latin translations were the same, but the Middle English and Klingon versions differed. To be honest, I like the Gemini Middle English version more than ChatGPT's, but the other aspects also put Gemini over the top as the winner here.

Crossword Key

(Image credit: Screenshots of Google Gemini and ChatGPT)

For the final test, I used logic, word association, vocabulary, and pattern recognition. In other words, a crossword puzzle. Any AI chatbot should be able to quickly identify words that match letter patterns and match the definition.

Crossword clues can be tricky because words often have multiple meanings. Both handled it well, though Gemini decided to go for the British spelling for unknown reasons.

Gemini Flash matches ChatGPT

Both Gemini 2.0 Flash and ChatGPT offer incredibly advanced AI models, and for most users, the difference comes down to preferences, ecosystem, and specific use cases rather than raw ability. Any clear technological edge that might have existed a year ago has essentially closed for the average person. Even 'personality' differences are likely due to the prompts used as much as any underlying programming.

If you primarily operate in Google's ecosystem, Gemini 2.0 Flash has obvious appeal. Its integration with Google Search, Google Docs, Gmail, and other Google services makes great for anyone who relies on these tools daily. You can use it to draft emails, summarize documents, and even assist with research, all within the Google suite. If speed and efficiency are priorities, Gemini is there for you.

On the other hand, ChatGPT is well-established and has deep partnerships. In particular, ChatGPT is embedded throughout Microsoft Office tools like Word and Excel. ChatGPT is often more conversational and expressive in my experience, though again, that is likely contingent on the prompts used.

At the core, both models have reached a level where their intelligence feels equally matched. There’s no longer a clear “better” option. You have to decide which AI fits more seamlessly into your daily life. Google and OpenAI are in an ongoing race, constantly refining their models and adding new capabilities. Gemini might take the lead in one update, only for ChatGPT to counter with an even stronger feature. There’s no wrong answer or best option. Whether you choose Gemini 2.0 Flash or ChatGPT, you’re getting one of the best AI models ever built. You just have to decide who you want to translate your Klingon for you.

You might also like...

• DeepSeek has already dethroned ChatGPT – is it coming for Midjourney and DALL-E next?
• OpenAI is making ChatGPT's Canvas feature smarter and sleeker
• ChatGPT Tasks can start taking over your calendar and remind you to finish your to-do list

Sonos has had a tumultuous few months, all of which date back to what was now clearly a rushed app design that led to a terrible experience for many customers and a long, messy process of repairing those relationships. It overshadowed the long-anticipated launch of the Ace over-ear headphones, though the more recently launched Arc Ultra soundbar’s impressive specs did shine through.

More recently, CEO Patrick Spence left, and a new leadership was ushered in. Earlier this week, it was announced that Sonos would further reduce its overall headcount, laying off 200 employees. Still, the new interim CEO Tom Conrad is shedding a bit more light on the internal changes – and much of it can be read in a way that’s returning to form.

While Conrad is quick to repeat that a lot more work needs to be done, he’s almost looking back in a good way. “I'm returning Sonos to a scrappier and more focused enterprise,” said Conrad, who’s focused on changing how Sonos operates. “I've reorganized our product and engineering staff into functional teams for hardware, software, design, quality and operations and away from dedicated business units devoted to individual product categories. This allows us to bring together rightsized cross functional projects that maximize our efficiency as we continuously evaluate, prioritize and focus on the highest value market opportunities,” explained Conrad.

(Image credit: Sonos México)

It’s a re-do on a previous structure that clearly did not work in a fluid way, most notably with folks mostly just working on individual parts of the business. Quite possibly just the app or maybe even more segments, that didn’t come together to improve the entire ecosystem of Sonos. That was likely a miss, and we saw some of the potential outcomes come to fruition, but Conrad and the teams left at Sonos have now been reorganized to better look at the brand as a whole.

For instance, the app is likely no longer siloed, with a team focused solely on it; rather, it might be more integrated across the entire lineup. How can the app better work with the Arc Ultra while improving interoperability with future products or creating a new flow to help you regroup existing, connected products when new ones are introduced?

Conrad described the setup at Sonos as being layered with redundancies, which likely made it hard to work across departments and, therefore, across product lines. Similarly, with a focus on “improving the Sonos experience to a place that exceeds the expectations of all of our customers,” he’s ensuring the company is operating in a way to hit this, or at least setting it up with that hope.

In the same vein, Conrad also stated that even with the near-constant app updates and other changes, "our core experience still needs significant improvement.” And that certainly should bring some hope to Sonos customers, long-standing ones and future ones. The restructure, while tough, and it is never good to see employees let go at any business, might help Sonos right the ship a bit more and ensure that a disaster on the level of the app doesn't occur again.

As for what might be next, our Managing Editor of Home Entertainment, Matthew Bolton, gave his expert analysis on the rumored streaming box.

And you can see the full transcript of the Sonos Quarter One of 2025 earnings call here.

You might also like

• 'We’ve let them down': Sonos gets a new CEO after its disastrous year, and it looks like better times are ahead
• The leaked Sonos streaming box could be a huge hit for custom home theaters, if it adds 2 key missing features from the current line-up

• North Korean hackers are using LinkedIn to scam jobseekers
• The fake job offers often promise well-paid remote work
• But the victims are eventually infected with malware

A long-running campaign by notorious North Korean hacking group Lazarus has seen job hopefuls scammed in many different ways, including downloading malware disguised as interview software, fake coding tests, infostealers, and some companies have even accidentally hired North Korean hackers as remote IT workers.

Now, a new facet of the ‘Contagious Interview’ campaign has arisen, and this time, hackers are using LinkedIn to scam victims, research from Bitdefender warns.

LinkedIn can be a fantastic tool for professionals to network, and many businesses use the app to recruit new employees, and now, it turns out, so are the Lazarus group.

Malicious offers

The fake recruitment scams ultimately result in the victim being infected with malware, and the hackers tend to target jobseekers in high profile industries, like defense, aerospace, or engineering - looking to exfiltrate classified or sensitive information, or even corporate credentials.

The fake jobs researchers observed in these scams were often remote work, flexible and well paid, sometimes involving cryptocurrencies as payment. These are designed to be enticing offers, so be wary of anything that looks a little too good to be true.

Scammers will message a victim via LinkedIn, then requesting a CV or personal GitHub repository link (which could be used to harvest personal information). From there, the ‘recruiter’ shares a ‘feedback’ document, which infects the victim with malware.

There are some warning signs to look out for, like vague job descriptions, poor communications, and users without popper documentations. Make sure to vet any job offers, applications, and interview offers thoroughly - and don’t click any links from unknown sources.

In February 2025, Apple delivered a new patch on Xprotect, its on-device malware removal tool to block variants of the macOS ‘FerretFamily’ - which had been found disguised as Chrome or Zoom installers targeting applicants.

You might also like

• Take a look at our pick of the best antivirus software
• Check out our recommendations for the best small business router picks around
• “Everyone will experience a hack” - how incident response can protect your organization

• Gravy Analytics is being sued for failing to protect personal data
• The suit comes after 17TB of records were allegedly stolen from the firm
• The hack on a data broker has sparked four lawsuits so far

A complaint has been filed in the federal court of Northern California outlining allegations that data broker Gravy Analytics has failed to properly safeguard vast amounts of personal data, which may now have been stolen.

This is the fourth such lawsuit since January 2025, when screenshots were posted to Russian cybercrime forum XSS fuelling fears that a staggering 17TB of records were swiped from the analytics firm’s AWS S3 storage buckets.

This breached information puts the privacy of millions at risk, and outlines the enormous risk when personal data is harvested and stored by private companies.

This week's suit alleges a huge archive of geo-locations from smartphone devices - here’s what we know so far.

Insufficient data protection

Arguing the firm had a duty to protect the data it collected and stored, the lawsuit points to the risk of identity theft for anyone whose information was compromised.

The latest complaint, reported by The Register, alleges "the hacked Gravy Analytics data included tens of millions of mobile phone coordinates of devices inside the US, Russia, and Europe, obtained through individuals’ use of major mobile applications such as Tinder, Grindr, Candy Crush [and more”.

The first breach was reported in early January 2025 after a hacker threatened to publish stolen location data, customer lists, and personal information harvested by Gravy Analytics and stolen in a huge hack.

Gravy Analytics has since been banned by the FTC from selling sensitive location data, alongside its subsidiary Venntel, after the FTC alleged the two violated the FTC Act by ‘unfairly selling sensitive consumer location data, and by collecting and using consumers’ location data without obtaining verifiable user consent for commercial and government uses.’

There are plenty of popular apps which collect your data, and often this is sold on to brokers for profit. Because a lot of this collection occurs through the ‘advertising ecosystem’ rather than a code the app creators themselves develop, this data collection is ‘likely happening without users’ or even app developers’ knowledge’.

The collection of personal information by the data broker industry comes with some serious risks and the industry is largely unregulated in the US, so the protections provided by laws like GDPR don’t apply.

The specific details of the hack aren't yet known, but keeping your organization safe is about anticipating and preparing for a potential attack, says Pierre Noel, Field CISO EMEA at Expel.

"The solutions to prevent a major security incident are well known- adequate protection, detection, and swift incident response. However, the real challenge lies in human nature: we instinctively believe cyberattacks only happen to others, rather than ourselves".

Take control of your data

If you regularly use the internet, unfortunately, it’s pretty likely that your information has fallen into the hands of a third party, whether it’s a company you use and gave permission to, appeared in a data breach, or whether it's been sold on legally to a broker.

“Data Privacy Day serves as a crucial reminder to safeguard sensitive information in an era where data dominates” comments Dr Ellison Anne Williams, CEO and founder of Enveil.

“As we navigate an increasingly interconnected world and transformative technologies such as AI grow their foothold in the digital economy, finding ways to protect data privacy and mitigate risk will be essential.”

Because of this, a market has opened up for the best personal data removal services, which can be a really powerful tool in helping you scrub you or your employees safe by removing your information from data brokers.

If you’re in the EU or UK and are protected by GDPR but still want to completely disappear your online persona - we still have some tricks for you.

The first is to delete your social media accounts. As much as you might enjoy scrolling, the Cambridge Analytica scandal told us that social media platforms have been used to harvest your data and map out your personal relationships and personality - so if you really value your privacy, the socials have to go.

Once those are gone, you’ll need to scour through your other accounts. Innocuous accounts like shopping sites or dating profiles will more than likely be monitoring your purchases or selling your demographic information, so removing these is also key. A great tip is to search your inbox for ‘sign-up’ or related phrases to get a reasonably accurate list to work from.

Going forward, staying anonymous online will be much easier with a Virtual Private Network (VPN). These essentially encrypt your internet traffic so that your browsing history isn’t recorded, and hides your IP address, so your location can’t be shared. We’ve listed the best VPN services to keep yourself safe.

You might also like

• Check out our list of the best firewall software around today
• Foh&Boh data leak leaves millions of CVs exposed - KFS, Taco Bell, Nordstrom applicants at risk
• We've also rounded up the best malware removal software on offer right now

• The company behind Arma and DayZ suffered a disruptive cyberattack
• Bohemia Interactive game's servers are inaccessible for days
• The community is speculating about the causes of the outage

If you’re a DayZ or Arma Reforger player, you may have noticed issues connecting to your favorite games these past couple of days, as the company’s servers are reportedly suffering a disruptive Distributed Denial of Service (DDoS) attack, rendering playing the games impossible.

The company behind these games is called Bohemia Interactive, a Czech-based developer and publisher who, according to media reports, first started experiencing problems with its servers several days ago.

A few days later, it confirmed the issues and apologized to the community, however five days later, the problems still seems to persist, as there are no updates on the company’s X page.

Political motives

In the latest tweet, published on February 5, it said: “We understand the frustration of the community with the server difficulties over the past week. And we know that many of you want answers. However, we cannot divulge details about the issue or the specifics of our efforts. We know this naturally leads to much community speculation, which we cannot prevent.”

BleepingComputer had also reported the Bohemia Interactive website was not functioning - however our tests show it seems to be back up and running.

Since the company did not want to disclose details about the issues, speculation is running rampant. On Reddit, a group named 'styled squad reborn' assumed responsibility for the attack, the publication found. The motives are unknown, but the group did say it wanted to see “drastic changes” in the company’s games, as well as in the company itself.

It denied claims it was asking for money. Some people also think this is related to the Russia - Ukraine war, and how it was portrayed in the games.

Until Bohemia comes forward with more details, there’s no way of knowing what actually happened. We have reached out to the company with additional questions and will update the article if we hear back.

You might also like

• DDoS attacks see a huge rise as criminals get braver and more ambitious
• We've rounded up the best password managers
• Take a look at our guide to the best authenticator app

OpenAI has changed the way that its new ChatGPT o3-mini model displays its chain of thought (Cot) to “make it easier for people to understand how the model thinks," however it has faced an almost immediate backlash from users and accusations that it is copying the way that DeepSeek's R1 model displays its reasoning.

Speaking to TechRadar, OpenAI said: “Users have told us that understanding how the model reasons through a response not only supports more informed decision-making but also helps build trust in its answers."

"While the model’s raw CoT remains hidden as it’s hard to understand, we’ve found a balance: the model can think freely, and then it organises those thoughts so that they are easy to read. To improve clarity and safety, we’ve added an additional post-processing step where the model reviews the raw chain of thought, removing any unsafe content and then simplifies any complex ideas. Additionally, this post-processing step enables non-English users to receive the CoT in their native language, creating a more accessible and friendly experience.”

The new approach effectively provides summaries of the model’s reasoning instead of showing you the raw data. However, despite the new approach being available in both the o3-mini for free and paid users and o3-mini-high for paid users, the response on X to its changes was not entirely positive.

Mayo Oshin responded on X, posting: “We'd appreciate if you showed the full chain of thought, not just summarised version....thank you”, and Conor responded with, “its still a summary and not real CoT, which is disappointing."

Some other users responded by suggesting that OpenAI was simply responding to the threat offered by the new DeepSeek by copying the way it presented the reasoning chain in its R1 model. “Finally DeepSeek changing the O-World for us,” replied Hamza. “So OpenAI copied Deepseek's Chain of Thought feature?” said Ignis Rex, and “That moment when China is the one innovating, and US is the copycat,” said Josip Tomo Licardo.

A big step forward

Personally, I much prefer the new way of presenting reasoning in ChatGPT o3-mini. If you try the model now (just hit the Reasoning button) before you enter your prompt, you’ll find that much more information is provided on how ChatGPT is coming to its conclusion. The previous scarcity of reasoning information, when compared to DeepSeek, was something I criticized 03-mini for in the last few days. So, from my perspective, I think the new approach is a big step forward for ChatGPT o3-mini.

I’m not alone in welcoming the new feature, either. X user Roman Pshichenko said, “Hilariously, the chain of thought is more endearing and funny than the stiff response,” and “That's a nice improvement! I'd love to see some examples of how this update impacts it's reasoning process," said Jesus Vazquezon X.

You may also like

• ChatGPT Search is now free for everyone, no OpenAI account required – is it time to ditch Google?
• ChatGPT on WhatsApp can now see, hear, and remember your conversations from elsewhere
• OpenAI's Deep Research smashes records for the world's hardest AI exam, with ChatGPT o3-mini and DeepSeek left in its wake

• Amazon’s last quarter capex was $26.3 billion last quarter, with growth set to continue
• Company plans to spend big on AI projects in 2025 to keep pace with rivals
• Microsoft and Google are also spending more on AI

Amazon's CEO appears to have indirectly revealed the company plans to allocate around $100 billion to capital expenditure in 2025, with much of the money destined for AI.

The news comes as the ecommerce and cloud computing giant announced its fourth-quarter and year-end financial results.

Amazon revenue for the final three-month period of 2024 climbed 10% year-over-year to $187.8 billion, boosting its full-year revenue to $638.0 billion, up 11% compared with 2023.

Amazon will invest even more money in 2025

Its Amazon Web Services cloud divison, accounted for around 15-17% of its entire revenue, depending on whether we take a quarterly or annual overview. Both periods saw 19% increases year-over-year.

Speaking with investors on an earnings call, CEO Andy Jassy suggested Amazon may spend more than $100 billion in the year ahead: “We spent $26.3 billion in capex in Q4, and I think that is reasonably representative of what you expect an annualized capex rate in 2025.”

Although a lack of figures doesn’t allow us to quantify the bold claims, Jassy added: “The vast majority of that capex spend is on AI for AWS.”

It’s not just Amazon looking to increase its capital expenditure to keep up with demand for artificial intelligence. Google recently revealed it would be spending $75 billion this year, with Microsoft set to invest $80 billion.

It’s no surprise Amazon’s investments will far exceed its key hyperscaler rivals – Canalys has its market share at a staggering 33%; Microsoft and Google occupy 20% and 10% of the market respectively. Just last month, Amazon unveiled an $11 billion investment into data centers in the state of Georgia.

The consensus is that investing more will help make AI more accessible and increase its use cases, therefore bringing more money in for the companies.

You might also like

• Check out our roundup of the best AI tools
• We’ve listed the best cloud hosting providers
• Microsoft reveals surprise plan to spend $80bn on AI data centers

• Yandex developed and open-sourced Perforator which identifies inefficiencies in applications
• This in turn allows DevOps and SysOps to optimize applications and reduce infrastructure costs
• Savings can be billions of dollars for hyperscalers but even SMBs with on-prem servers can benefit

Yandex, Russia's largest tech company, has released an open source tool designed to identify and evaluate code inefficiencies across an entire code base.

“Perforator” can help a range of businesses - from tech giants like Google and Microsoft to SMBs - reduce their infrastructure costs by up to 20%, the company says, helping DevOps and SysOps teams pinpoint the most resource-intensive sections of code and offers detailed suggestions for how to optimize them. Using eBPF technology it can run programs within the Linux kernel safely and without slowing down systems.

The tool supports a range of popular languages, including C, C++, Go, Rust, Python, and Java. Analytics and data visualization are offered through flame graphs, making problem diagnostics much more manageable, and use of profile-guided optimization (PGO) can accelerate C++ programs by up to 10%. Yandex says it intends to improve future versions with better integration for Python and Java, along with more precise event analysis.

Saving hyperscalers a fortune

“Perforator helps businesses get the most out of their servers without sacrificing performance," said Sergey Skvortsov, a senior developer at Yandex who leads the team behind the tool. "Using Perforator, businesses can optimize their code, reduce server load, and ultimately lower energy and equipment costs."

Yandex says Perforator is useful for large data centers, tech corporations, and even smaller businesses with limited resources. Instead of investing in additional equipment, they can use the tool to get more out of their existing infrastructure. Perforator also includes handy A/B testing capabilities.

The tool has been used within Yandex services for over a year and is now available globally for companies, developers, and researchers. Because it runs directly on a company’s servers, Perforator reduces reliance on external cloud providers, making it ideal for organizations with strict data security requirements.

For hyperscalers, the potential savings could be huge, but SMBs with on-prem servers can also see substantial cost reductions. "Perforator can benefit companies of all sizes, from small businesses with 10-100 servers, which can save millions of dollars per year, to larger enterprises with thousands of servers and more, where savings can reach hundreds of millions or even billions of dollars annually,” Skvortsov said.

Perforator’s source code is available for free on GitHub, alongside Yandex's other open-source tools.

You might also like

• This is the best small business software available right now
• And these are the best IT infrastructure management services
• Russia's largest tech company just released a free AI tool

Looking for a different day?

A new Quordle puzzle appears at midnight each day for your time zone – which means that some people are always playing 'today's game' while others are playing 'yesterday's'. If you're looking for Friday's puzzle instead then click here: Quordle hints and answers for Friday, February 7 (game #1110).

Quordle was one of the original Wordle alternatives and is still going strong now more than 1,100 games later. It offers a genuine challenge, though, so read on if you need some Quordle hints today – or scroll down further for the answers.

Enjoy playing word games? You can also check out my NYT Connections today and NYT Strands today pages for hints and answers for those puzzles, while Marc's Wordle today column covers the original viral word game.

SPOILER WARNING: Information about Quordle today is below, so don't read on if you don't want to know the answers.

Quordle today (game #1111) - hint #1 - Vowels How many different vowels are in Quordle today?

• The number of different vowels in Quordle today is 5*.

* Note that by vowel we mean the five standard vowels (A, E, I, O, U), not Y (which is sometimes counted as a vowel too).

Quordle today (game #1111) - hint #2 - repeated letters Do any of today's Quordle answers contain repeated letters?

• The number of Quordle answers containing a repeated letter today is 2.

Quordle today (game #1111) - hint #3 - uncommon letters Do the letters Q, Z, X or J appear in Quordle today?

• No. None of Q, Z, X or J appear among today's Quordle answers.

Quordle today (game #1111) - hint #4 - starting letters (1) Do any of today's Quordle puzzles start with the same letter?

• The number of today's Quordle answers starting with the same letter is 2.

If you just want to know the answers at this stage, simply scroll down. If you're not ready yet then here's one more clue to make things a lot easier:

Quordle today (game #1111) - hint #5 - starting letters (2) What letters do today's Quordle answers start with?

• A

• A

• H

• T

Right, the answers are below, so DO NOT SCROLL ANY FURTHER IF YOU DON'T WANT TO SEE THEM.

Quordle today (game #1111) - the answers

(Image credit: New York Times)

The answers to today's Quordle, game #1111, are…

• AFOOT
• ALIKE
• HUMUS
• TOWEL

Another close shave. After a couple of unlucky guesses and one wild one (Pushy was an extremely unlikely punt) I just about managed to complete today’s puzzle.

The word AFOOT is automatically connected to “the game is afoot”, an idiom uttered by Sherlock Holmes to Watson when his detecting senses are pricked by a case. The phrase was not coined by author Sir Arthur Conan Doyle, but by Shakespeare's Henry V and his impassioned plea for the English to take up arms.

Whatever its provenance, for me today’s Classic game was not afoot, but the Sequence most definitely was and I somehow pulled rabbits out of hats to conjure up the four words. A very WOOZY experience.

How did you do today? Let me know in the comments below.

Daily Sequence today (game #1111) - the answers

(Image credit: New York Times)

The answers to today's Quordle Daily Sequence, game #1111, are…

• WOOZY
• WHOOP
• NOMAD
• BLOKE

Quordle answers: The past 20

• Quordle #1110, Friday 7 February: PETAL, ABASE, AMONG, CORER
• Quordle #1109, Thursday 6 February: MADAM, SCAMP, FAUNA, TRACK
• Quordle #1108, Wednesday 5 February: SIXTH, VERVE, BLURB, CACTI
• Quordle #1107, Tuesday 4 February: BIRTH, DRAWN, MILKY, JOIST
• Quordle #1106, Monday 3 February: SLEET, TENSE, SEIZE, SCAMP
• Quordle #1105, Sunday 2 February: LOUSY, PENNE, CHUMP, SALVE
• Quordle #1104, Saturday 1 February: FACET, BLISS, FLUFF, FRANK
• Quordle #1103, Friday 31 January: PARER, SYNOD, BRAWN, BANJO
• Quordle #1102, Thursday 30 January: WEEDY, CHIRP, STALE, SUGAR
• Quordle #1101, Wednesday 29 January: CROWN, SCENE, AGILE, CRANK
• Quordle #1100, Tuesday 28 January: TOKEN, DAISY, INANE, PLIED
• Quordle #1099, Monday 27 January: ANGEL, GLOBE, MAYOR, BRAVE
• Quordle #1098, Sunday 26 January: WATER, ORBIT, BRIDE, TOUCH
• Quordle #1097, Saturday 25 January: GIVER, GROOM, GUSTO, MAKER
• Quordle #1096, Friday 24 January: GRIND, STOCK, SERVE, TOUCH
• Quordle #1095, Thursday 23 January: RUGBY, SASSY, WORRY, BANAL
• Quordle #1094, Wednesday 22 January: SLANT, TRUNK, WOOZY, EATEN
• Quordle #1093, Tuesday 21 January: CHART, VIGOR, PRINT, SPAWN
• Quordle #1092, Monday 20 January: SIXTY, THONG, TATTY, ROBIN
• Quordle #1091, Sunday 19 January: WREST, RINSE, SCOUR, CANNY

Looking for a different day?

A new NYT Strands puzzle appears at midnight each day for your time zone – which means that some people are always playing 'today's game' while others are playing 'yesterday's'. If you're looking for Friday's puzzle instead then click here: NYT Strands hints and answers for Friday, February 7 (game #341).

Strands is the NYT's latest word game after the likes of Wordle, Spelling Bee and Connections – and it's great fun. It can be difficult, though, so read on for my Strands hints.

Want more word-based fun? Then check out my NYT Connections today and Quordle today pages for hints and answers for those games, and Marc's Wordle today page for the original viral word game.

SPOILER WARNING: Information about NYT Strands today is below, so don't read on if you don't want to know the answers.

NYT Strands today (game #342) - hint #1 - today's theme What is the theme of today's NYT Strands?

• Today's NYT Strands theme is… Westminster assembly

NYT Strands today (game #342) - hint #2 - clue words

Play any of these words to unlock the in-game hints system.

• NUDE
• GOING
• SPOON
• RIDER
• KING
• DROWN

NYT Strands today (game #342) - hint #3 - spangram What is a hint for today's spangram?

• Well bred

NYT Strands today (game #342) - hint #4 - spangram position What are two sides of the board that today's spangram touches?

First side: left, 4th row

Last side: right, 5th row

Right, the answers are below, so DO NOT SCROLL ANY FURTHER IF YOU DON'T WANT TO SEE THEM.

NYT Strands today (game #342) - the answers

(Image credit: New York Times)

The answers to today's Strands, game #342, are…

• WORKING
• HOUNDS
• HERDING
• TOYS
• TERRIERS
• SPORTING
• SPANGRAM: DOG GROUP

• My rating: Moderate
• My score: 2 hints

Despite being a big fan of the movie Best In Show – which parodies the prestigious and pretentious dog show world – it took me ages before I realized what today’s theme was about.

Being British, Westminster automatically triggers thoughts of the Houses of Parliament, Big Ben and all that tourist malarky, rather than The Westminster Kennel Club and its many canine categories.

Despite my ignorance, this was still a really enjoyable puzzle, with dozens of hint words lurking in the grid to throw us off scent. Top marks, Strands.

How did you do today? Let me know in the comments below.

Yesterday's NYT Strands answers (Friday, 7 February, game #341)

• EMBROIDERY
• LACE
• FRINGE
• TASSEL
• SEQUINS
• SPANGRAM: CLOTHING ACCENTS

What is NYT Strands?

Strands is the NYT's new word game, following Wordle and Connections. It's now out of beta so is a fully fledged member of the NYT's games stable and can be played on the NYT Games site on desktop or mobile.

I've got a full guide to how to play NYT Strands, complete with tips for solving it, so check that out if you're struggling to beat it each day.

Looking for a different day?

A new NYT Connections puzzle appears at midnight each day for your time zone – which means that some people are always playing 'today's game' while others are playing 'yesterday's'. If you're looking for Friday's puzzle instead then click here: NYT Connections hints and answers for Friday, February 7 (game #607).

Good morning! Let's play Connections, the NYT's clever word game that challenges you to group answers in various categories. It can be tough, so read on if you need Connections hints.

What should you do once you've finished? Why, play some more word games of course. I've also got daily Strands hints and answers and Quordle hints and answers articles if you need help for those too, while Marc's Wordle today page covers the original viral word game.

SPOILER WARNING: Information about NYT Connections today is below, so don't read on if you don't want to know the answers.

NYT Connections today (game #608) - today's words

(Image credit: New York Times)

Today's NYT Connections words are…

• FAST
• BUNCH
• TEAM
• PLAYER
• SOUL
• STAT
• SOON
• RUFFLE
• LATER
• GATHER
• BIO
• FINGER
• NOW
• JUNK
• THEN
• PUCKER

NYT Connections today (game #608) - hint #1 - group hints

What are some clues for today's NYT Connections groups?

• YELLOW: Ballers deets
• GREEN: Ruche it up  
• BLUE: Cometh the hour 
• PURPLE: Eat it

Need more clues?

We're firmly in spoiler territory now, but read on if you want to know what the four theme answers are for today's NYT Connections puzzles…

NYT Connections today (game #608) - hint #2 - group answers

What are the answers for today's NYT Connections groups?

• YELLOW: INFO ON A BASEBALL CARD
• GREEN: SCRUNCH, AS FABRIC
• BLUE: TIME ADVERBS 
• PURPLE: — FOOD 

Right, the answers are below, so DO NOT SCROLL ANY FURTHER IF YOU DON'T WANT TO SEE THEM.

NYT Connections today (game #608) - the answers

(Image credit: New York Times)

The answers to today's Connections, game #608, are…

• YELLOW: INFO ON A BASEBALL CARD BIO, PLAYER, STAT, TEAM
• GREEN: SCRUNCH, AS FABRIC BUNCH, GATHER, PUCKER, RUFFLE
• BLUE: TIME ADVERBS LATER, NOW, SOON, THEN
• PURPLE: — FOOD FAST, FINGER, JUNK, SOUL

• My rating: Easy
• My score: Perfect

I completed today’s Connections without ever knowing what the four groups were until getting them correct – I just thought they belonged together.

It’s part of the two-tiered wonder/satisfaction/disappointment loop of this particular puzzle. You can whiz through it and still feel as if you know nothing – and I know absolutely nothing about INFO ON A BASEBALL CARD – which in turn can make you feel either blessed or as if you have gotten away with something.

I’m definitely in the latter camp. “He got away with it (until now)” is all that will need to be said in my obituary.

How did you do today? Let me know in the comments below.

Yesterday's NYT Connections answers (Friday, 7 February, game #607)

• YELLOW: BE IN CHARGE OF DIRECT, HEAD, LEAD, RUN
• GREEN: THINGS MADE OF RUBBER BALLOON, ERASER, GALOSH, TIRE
• BLUE: ASSOCIATED WITH THE GENIE IN "ALADDIN" BLUE, EARRING, LAMP, WISHES
• PURPLE: RED -_ BULL, CROSS, HERRING, VELVET

What is NYT Connections?

NYT Connections is one of several increasingly popular word games made by the New York Times. It challenges you to find groups of four items that share something in common, and each group has a different difficulty level: green is easy, yellow a little harder, blue often quite tough and purple usually very difficult.

On the plus side, you don't technically need to solve the final one, as you'll be able to answer that one by a process of elimination. What's more, you can make up to four mistakes, which gives you a little bit of breathing room.

It's a little more involved than something like Wordle, however, and there are plenty of opportunities for the game to trip you up with tricks. For instance, watch out for homophones and other word games that could disguise the answers.

It's playable for free via the NYT Games site on desktop or mobile.

• New documentation suggests we'll see Fire TV devices run on Android 14
• That's despite claims Amazon was moving Fire TV to a new platform
• A big product launch is planned for February 26

Amazon appears to be working on some new Fire TV Sticks that run on Android 14, and we could get our first good look at them before the month is out.

Eagle-eyed reporters for AFTVNews spotted the mention of new hardware on one of Amazon’s developer pages, titled ‘Developing for Amazon Fire TV Devices Running Android 14’.

The page opens by saying that “Android 14-based Fire TV is based on API level 34. The following sections explain some of [the] important changes that you should consider when you build apps for Android 14-based Fire TV.”

There are no specific models listed, but the term ‘Android 14’ reappears on another Fire OS page, and with a big Amazon product launch only a few weeks away, we’re curious about whether a new Fire TV lineup is about to appear.

We don’t yet have any Fire TV devices running on Android 14 – the version of Google’s ubiquitous operating system that launched in 2023 – and it looked like Amazon was keen to start shifting its streaming hardware onto the new Linux-based VegaOS, which runs on the latest versions of the Echo Spot and Echo Show.

However, the documentation suggests that Amazon’s Fire TV streaming sticks and smart TV collaborations could still run on a version of the Android platform, and with no current devices running it, it seems likely that new ones are coming.

The last Fire TV Stick was the HD model launched in 2024, and we wouldn’t be surprised to see some updates to older Fire TV hardware for the coming year.

(Image credit: Getty Images) You, me, and Alexa

The big launch event is on February 26, and is expected to showcase a next-generation, AI-powered iteration of the Alexa smart assistant – which could well be integrated into any new-for-2025 Fire TV hardware.

This new Alexa should be better at understanding natural speech, and handling multiple contextual prompts in sequence – allowing for a more streamlined, conversational interaction – and even act as an AI agent to perform tasks on your behalf.

These kinds of AI enhancements are still in public beta for the most part, and you may find yourself paying $5-10 dollars per month for the privilege of testing these not-yet-perfect features out. But it shows that Amazon is stepping up to the AI plate and making sure it’s part of this new frontier in smart assistants. Given how ubiquitous Alexa is across the best smart home devices, it’s something we should all take notice of.

Here’s hoping the new Fire TV brings back our ability to hide apps on the homepage, though…

You might also like…

• Meta could launch 6 new AI wearables in 2025, according to leaked memo
• Google Gemini's new model is a brainstorming AI partner
• TikTok owner ByteDance has a new AI video creator

• Severance director Ben Stiller has delivered a promising update on season 3
• Stiller says a writer's room has been assembled and script work is already underway
• Apple hasn't officially confirmed that a third season of its hit series is in development

Ben Stiller has confirmed that work on Severance season 3 is already underway, despite the fact that Apple hasn't officially renewed its biggest TV Original.

As part of a profile piece in a recent issue of The Hollywood Reporter (THR), Stiller revealed that, even though an announcement hasn't been made, season 3 was in the early stages of development. Stiller, who serves as Severance's lead director and one of its many executive producers, says a writer's room has been assembled, and that progress is being made on next season's scripts.

Confirmation that Severance season 3 is in development will come as no great surprise to fans of the Apple TV Plus series. The sci-fi mystery-thriller was already one of the tech giant's biggest original shows following its critically-acclaimed first season. However, ever since Severance season 2 was released on January 17, its popularity has skyrocketed as more and more people have tuned in to take in its 'new weird' story, be enthralled by its compelling and complex cast of characters, and offer up their own theories on what's going on. It seems inevitable, then, that a third season will be announced sooner rather than later.

It's unclear if Severance's third chapter will be its last. Speaking to TechRadar prior to one of the best Apple TV Plus shows' return, creator Dan Erickson said he already knows how many seasons it'll take to tell Severance's whole story. He has an idea of what its final scene will be, too, for what it's worth. Considering how successful Severance has become, I'd be very surprised if its to-be-announced third installment is the show's final entry, especially when it's been such a money spinner for Apple TV Plus. That said, if it only takes three seasons to wrap up Mark Scout and company's various journeys, maybe it will be after all.

Will there be a three-year gap between Severance seasons 2 and 3?

Severance fans are hoping that we'll be reunited with Dylan and Irving before 2028 (Image credit: Apple TV Plus)

I hope not! And it sounds like Stiller is in the same boat. Per Rebecca Keegan, the writer of THR's article, Stiller said that he would like Severance season 3 to launch on one of the world's best streaming services much sooner than January 2028. That would officially mark the three-year period between the debuts of season 2 and season 3.

Read more of our Severance season 2 coverage

- Severance star Patricia Arquette says Harmony Cobel's loyalty to Lumon will be tested in season 2
- Severance actor Tramell Tillman teases what lies in store for Seth Milchick in season 2 of the hit Apple TV show
- Yes, Severance season 2 episode 1 is hiding a Keanu Reeves cameo in plain sight
- 5 of the best Severance season 2 goat theories I've seen, from the bizarre to the quite plausible

There were multiple reasons why it took so long for Severance season 2 to see the light of day – the pandemic and 2023 Hollywood strikes being the two biggest factors for its delayed release. As long as there are no similarly big issues during season 3's development cycle, season 3 should be with us before we bid farewell to 2027.

But let's not get ahead of ourselves. There are still six more episodes of season 2 left to enjoy after its latest chapter, aka 'Woe's Hollow'. I bet you've got plenty of questions after the shocking and confusing events of that entry, too, so read my Severance season 2 episode 4 ending explained piece to find out what the heck just happened.

You might also like

• Severance season 2 has a brand new title sequence – here are the 5 major clues it gives us about the hit show's story
• Severance creator Dan Erickson reveals the funniest fan theory he's heard about the popular Apple TV Plus show: 'I think that's ridiculous'
• Everything new on Apple TV Plus in February 2025: The Gorge, Surface season 2, and more

• Sophos says the use of SVG files in phishing is on the rise
• SVG files bypass email protection and can display malicious hyperlinks
• The researchers shared a few tips on how to stay safe

Hackers are using .SVG files in new phishing attacks aimed at stealing people’s Office 365 login credentials, experts have warned.

A report from researchers at Sophos revealed the number of phishing attacks with .SVG files in attachments is on the rise. SVG (Scalable Vector Graphics) files are XML-based images that can be scaled without losing quality, making them ideal for web design, icons, and illustrations. Unlike raster images (e.g., PNG, JPG), SVGs use mathematical equations to define shapes, allowing them to remain crisp at any size.

Since SVG files are usually loaded natively inside a browser, they can contain anchor tags, scripts, and other kinds of active web content.

Defending against SVG attacks

Sophos notes the body of the phishing emails is nothing extraordinary. It’s the usual invoice/new voicemail/signature required type of email, with an .SVG attachment, that usually only displays a sentence or two, and a hyperlink. Sophos says that it’s seen these messages, especially the contents inside the SVG file, grow more sophisticated as the campaign progressed.

In any case, opening the SVG file brings up a new browser tab, and in it a hyperlink. Clicking the hyperlink redirects the victim to a fake Office365 login page that steals the login credentials and relays them to the attackers.

There are two ways to defend against these phishing emails, Sophos said. The best way (aside from not clicking on shady email attachments) is to open a known, benign SVG file on the computer, and instruct Windows to always open it in Notepad, or a similar non-browser program.

“Even if you accidentally click a malicious SVG in the future, it’ll only open in Notepad, throwing another roadblock in front of (potentially) being phished,” Sophos explained. “If, at some point, you find you need to work with real SVG files, follow the same steps again, and choose the graphics application you plan to use.”

The second way is to use a reputable email security program. Sophos said a detection signature was developed for the various kinds of weaponized files it recently observed.

You might also like

• Hackers are loading SVG files with multi-stage malware in new phishing attack
• We've rounded up the best password managers
• Take a look at our guide to the best authenticator app

In today's rapidly evolving digital landscape, enterprises face mounting pressure to deliver innovative solutions faster, at a higher quality, while managing costs effectively. Traditional approaches to software development, which rely heavily on expanding development teams to meet growing demands, are becoming increasingly unsustainable.

As we stand at the cusp of a technological revolution, AI presents a compelling alternative that could fundamentally transform how we approach software development and modernization. This transformative technology is particularly relevant in an era where digital transformation initiatives are accelerating across industries, and the demand for software development talent far exceeds the available supply.

Through the strategic implementation of AI, enterprises are discovering new ways to break free from the traditional linear relationship between revenue growth and workforce expansion.

The role of AI agents in modern software development

The role of AI agents in software development represents one of the most significant technological shifts in recent years. These agents are fundamentally changing how software gets built and delivered across enterprises.

Accelerated development timelines: AI agents demonstrate the capability to generate code up to 70% faster than human developers. By automating routine coding tasks, developers can redirect their focus towards strategic aspects of development. Furthermore, these agents enable rapid prototyping, facilitating quick iterations and faster time-to-market for new features and products.

Cost optimization: With AI agents, organizations can achieve significant cost reductions through decreased development hours, lower overhead costs compared to maintaining large development teams, and minimized expenses related to debugging and quality assurance activities.

Enhanced quality and consistency: AI agents consistently follow standardized coding practices, incorporate automated error detection and prevention mechanisms, and leverage built-in optimization capabilities to ensure superior performance. Additionally, they maintain comprehensive project documentation through systematic generation processes.

Scalability benefits: AI agents can efficiently manage multiple projects simultaneously while maintaining consistent performance levels regardless of workload. They also demonstrate remarkable adaptability by quickly learning and implementing new programming languages and frameworks.

The challenges of fully autonomous AI agents

Fully autonomous AI agents in software development, while powerful, present several challenges that need careful consideration:

Limited human oversight: Without real-time intervention mechanisms during code generation, it becomes challenging to incorporate domain-specific knowledge and capitalize on strategic decision-making opportunities at crucial development stages.

Trust and reliability concerns: There's inherent uncertainty about the reliability of automatically generated code—potential security vulnerabilities could go undetected and there’s limited ability to validate business logic implementation. Furthermore, ensuring compliance with industry standards and regulations becomes more complex.

Integration complexities: Organizations struggle with seamlessly integrating AI agents into existing development workflows, maintaining consistency with established coding standards, adapting to organization-specific requirements, and managing potential conflicts with legacy systems and architectures.

Knowledge transfer limitations: Reduced opportunities for knowledge sharing and team learning, increased risk of organizational dependency on AI systems without building internal expertise, and challenges in maintaining and updating organizational best practices create long-term risks.

The power of human-steered AI agents

To address the challenges listed above, AI agents steered by human-in-the-loop can be used to leverage the strengths of both humans and AI:

Controlled automation: Human experts and developers can guide and validate AI decisions at critical points in the development process, facilitating controlled automation.

Quality assurance: Quality assurance is enhanced as human experts can review and refine AI-generated code in real time, ensuring high standards are maintained.

Context integration: A system that incorporates human-steered AI agents allows for effective context integration, incorporating business requirements and domain knowledge seamlessly into the development workflow.

Risk mitigation: Continuous human oversight reduces the likelihood of errors and ensures alignment with business objectives.

Knowledge enhancement: These systems exhibit continuous improvement through knowledge enhancement, as AI systems learn from human feedback and become more refined over time.

The true power lies in creating a collaborative environment that fosters a symbiotic relationship. Developers can focus on high-level architecture and strategic decisions while AI handles routine coding tasks and initial implementations. This collaboration creates a feedback loop where humans provide corrections that improve AI performance, and teams maintain control while benefiting from AI acceleration.

Driving enterprise revenue growth through AI-enhanced development

The implementation of human-steered AI agents enables enterprises to achieve substantial growth without the need for headcount expansion. Here’s how:

Cost-effective scaling: With AI-assisted development solutions, enterprises can now handle a significantly larger project portfolio without that linear relationship between headcount and output. AI takes care of routine, repetitive tasks, allowing existing team members to focus on higher-value activities, and helping reduce per-project costs.

Accelerated time-to-market: Through AI-assisted development, organizations can now launch new features and products in a fraction of the traditional time. Companies can now respond to market opportunities almost in real-time, rather than being constrained by development bottlenecks.

Resource optimization: The key to successful resource optimization lies in maximizing the potential of existing talent by allowing them to focus on high-value strategic initiatives rather than getting bogged down by routine tasks. Through intelligent automation, organizations can eliminate redundant and repetitive work that previously consumed valuable human resources. This shift enables teams to redirect their energy towards complex problem-solving, innovation, and strategic decision-making.

Competitive advantage: When organizations leverage AI-assisted development capabilities, they gain the ability to stay consistently ahead of market trends through rapid development and deployment of new features. This speed advantage, combined with significantly reduced development costs, enables businesses to offer more competitive pricing while maintaining their profit margins.

Conclusion

As we move forward, organizations must adapt to remain competitive in an increasingly digital world. AI agents, enhanced by human-in-the-loop, represent not just an evolution, but a revolution in how enterprises approach technical innovation. The key to success lies in finding the right balance between human expertise and AI capabilities. Organizations should:

Evaluate current processes: Assess your existing software development workflows and identify areas where human-steered AI agents could drive improvements.

Start small, scale fast: Begin with pilot projects to experience the benefits of human-steered AI agents firsthand.

Prepare your team: Invest in training developers to work effectively with AI agents, maximizing the potential of this powerful collaboration.

By following these principles, enterprises can position themselves for sustainable growth in the digital age, leveraging the power of AI agents steered by human-in-the-loop to drive innovation and efficiency in software development.

We've featured the best web development 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

• The Asus ROG Astral LC RTX 5090 tips the scales at $3,409
• Some other RTX 5090 and 5080 models also command a hefty premium over MSRP
• Stock of these GPUs remains vanishingly thin anyway

Nvidia’s RTX 5090 and 5080 GPUs are off to a shaky start, not just because of the fact that these new Blackwell graphics cards aren’t available to buy – they sold out immediately, more or less – but also due to pricing on third-party boards.

Wccftech has been scanning pricing across the Asus and MSI Blackwell ranges, and the findings will likely scare, if not terrify, gamers out there who are mulling a high-end GPU purchase.

When stock does come back in – there are smatterings of live pre-orders here and there right now, but not much else, although be sure to keep an eye on our guides on where to buy the RTX 5090 and also the RTX 5080 GPU – price tags are looking seriously steep.

The crown for the most eye-watering price of all is awarded to the Asus ROG Astral LC RTX 5090, a behemoth of a flagship that admittedly pulls out all the proverbial stops (the ‘LC’ refers to the incorporated liquid cooling system), but it’ll set you back $3,409 in the US currently at the graphics card maker’s own store.

Even the cheapest RTX 5090 GPU that Asus currently sells, the TUF Gaming model, is a weighty $2,749, which is over a third more expensive than the MSRP as set by Nvidia (and the asking price for its Founders Edition). Remember, this is the base model from Asus, you can’t go any lower.

At least with the RTX 5080, there is an Asus model at the MSRP of $999 on its web store, the Prime model – so that’s something. If you want the TUF Gaming model of the RTX 5080, though, this will set you back $1,484, quite a hefty hike for a variant that isn’t a top-end card. (The ROG Astral, air cooled, RTX 5080 touches $1,649, all prices correct at the time of writing).

MSI does somewhat better at keeping a lid on pricing, but there are some difficult to swallow price tags nonetheless. There are no RTX 5080 models at MSRP on MSI’s US store, and the cheapest is the Ventus 3X at $1,139, although that’s not a big jump. The top-end RTX 5080 Suprim Liquid SOC is pitched lower than the Asus ROG Astral, though, at $1,499.

MSI’s RTX 5090 graphics cards kick off with the Ventus 3X which is at $2,379 on its store, more reasonably priced than the baseline flagship from Asus, but still a 20% hike (almost) over the MSRP. If you’re looking at the RTX 5090 Suprim Liquid SOC, though, this graphics card is an eye-watering $2,789 on the MSI store.

(Image credit: Shutterstock) Analysis: Okay, so a liquid-cooled flagship was never going to be cheap, but…

Before the new Blackwell graphics cards emerged, the prediction was that they could be priced considerably higher than the MSRPs. And that has indeed broadly come true, although part of the problem here is that third-party board makers seem to have inflated prices even further after all the clamor and stock shortages around the RTX 5090 and 5080 GPUs.

As an example at the highest-end, the mentioned Asus ROG Astral LC RTX 5090 is the priciest of all Blackwell GPUs at $3,409. However, it was announced at $3,100 (on Reddit, and caused quite a stir at that price), so somehow, there’s been another $300 heaped on top of the already exorbitant asking price.

I’m not impressed with these kinds of shenanigans as it’s bad enough that stock is stupidly difficult to get hold of, and scalpers are causing huge price inflation by reselling RTX 5080 and 5090 graphics cards, without the actual board makers themselves deciding to try and push even harder for profits.

Is this just market dynamics – a reflection of thin supply and big demand? Well, yes, it is to an extent, and doubtless pricing will settle down eventually anyway, if you can be patient. (Indeed, there are still some big question marks over how long we might be waiting for pricing to even out to a more acceptable level overall).

Meanwhile, it’s clearly true that asking prices remain something of a moot point seeing as there are hardly any RTX 5090 or 5080 graphics cards for sale at all, whether at recommended pricing, or third-party pricing.

You might also like...

• Nvidia's RTX 5080 has dethroned AMD's RX 7900 XTX at the same price – but good luck finding one
• Can’t buy an Nvidia RTX 5080 or 5090 GPU due to stock shortages? If you were hoping GeForce Now might provide a good alternative, we’ve got some bad news
• Nvidia releases stats that prove DLSS and Frame Generation are here to stay - sorry, angry gamers

• Android's security team partners with Mandiant FLARE for upgrade
• The capa open source binary analysis tool is being enhanced
• Gemini AI is being thrown in the mix, too

Google is ramping up its Android security protection with new app safety tools.

In a new blog post, Google’s Lin Chen announced the company’s Android Security and Privacy Team is partnering with Mandiant FLARE, to enhance the capa open source binary analysis tool. That way, the tool will be better at analyzing ARM ELF files, often used in Android malware.

Chen said this collaboration will help detect and highlight suspicious code behaviors in native files, enabling faster malware analysis and decision-making, with the help of Gemini AI.

Detecting malware in ELF

Describing how the new tools work, Chen shared a case study of an illegal gambling app disguised as a music app. This app, found on the Google Play Store, was secretly loading gambling websites for users in specific regions. It used different anti-analysis techniques (hiding key functions in a native ELF file, timezone detection, dynamic downloading and decryption of additional malicious code) to stay hidden in plain sight.

However, by leveraging static analysis and capa, Google's team identified these deceptive behaviors and successfully removed the app.

Capa detects malware capabilities in ELF files, and new rules have been developed specifically for Android, Chen further explained.

These rules identify behaviors like ptrace API calls (anti-debugging), extracting device and timezone info via JNI, downloading and decrypting code, using Base64 & Cipher API for encoding/encryption, allowing analysts to quickly locate suspicious functions, without needing to sift through mountains of obfuscated code.

Google also added Gemini AI to summarize the most suspicious functions highlighted by capa. The AI tool can do risk level assessments, insights into obfuscation, anti-debugging, and cloaking tactics, enabling a faster, and more effective, malware detection and rule-writing.

“Equipped with the fast-evolving Gemini, our analysts are able to spend less time on those sophisticated samples, minimising the exposure for malicious apps and ensuring the safety of Android ecosystems,” Chen concluded.

You might also like

• Dangerous new Android malware infects 11 million devices — here's what we know
• We've rounded up the best password managers
• Take a look at our guide to the best authenticator app

The recent announcement by the UK Government that it would designate data centers as Critical National Infrastructure (CNI), placing them in the same category as energy and water systems, is a clear recognition of the vital role they play in today’s economy and society. There are no official records of how many data centers there are in the UK, perhaps down to the sensitive nature of data itself, however, Statistica puts the current number at 514, the third highest number in one country worldwide, behind the United States and Germany.

It’s difficult to envisage any aspect of modern life that data centers don’t impact – from the way we communicate to our interactions with businesses and vital public services, including the NHS. We simply couldn’t function without them. While this government initiative is reassuring, its impact in real terms is questionable.

A data center itself is a building with power, cooling and network access. It’s the data and systems that reside within them which is of critical national importance. Safeguarding this is the responsibility of the data center's client - not the physical provider. However, the building provider can enforce physical security and ensure it’s at an appropriate level which mirrors cybersecurity defenses. Additional, specific issues need to be taken into account and adequately addressed, including approaches to cyber resiliency, power consumption and sustainability challenges.

The cyber resiliency challenge

It could be argued that the government is playing catch-up with its CNI initiative, particularly in relation to cyber security. In recent years, we’ve seen regulations come into effect, such as the National Institute of Standards Technology (NIST) security framework in the US, and the European Union’s Digital Operation Resilience Act (DORA) and NIS2, which between them, have become defaults for how data is managed and secured. That’s not to say that the UK CNI initiative isn’t valuable, because it is. Crucially, it sends out a clear message to data center operators that they have a responsibility to take specific steps to prepare for cyber-attacks and have robust recovery plans in place.

In reality, despite having the best enterprise-grade cyber defenses and strategies in place, some attacks will be successful. This is made clear by the most recent figures published by the UK Information Commissioner’s Office, revealing record levels of ransomware and cyber-attacks in 2023. It’s almost impossible to have foolproof cyber defenses, especially considering that attacks are constantly evolving and often originate from very well-resourced criminal gangs or even hostile nation states. For this reason, it has become best practice in the cyber security world to assume that it’s a matter of when, not if, a breach will occur. The CNI initiative, with the backing of the forthcoming Cyber Security and Resilience Bill, will mandate that data centers implement robust security protocols and adhere to higher security standards to protect sensitive data. This increased focus on security can only be beneficial.

Understanding the impact of data center CNI designation on sustainability

According to a 2022 National Grid study, UK data centers consume between 2–3 terawatt-hours (TWh) of power per year - enough electricity to power up to 800,000 homes annually. However, consumption is increasing and is set to jump dramatically over the next 10 years. Energy consumption is a well known issue, and data center operators have been working to address the enormous challenges this presents, especially in relation to carbon emission and sustainability commitments.

In addition, data center operators and in many cases, their customers will now have to contend with additional burdens that arise from CNI designation. In many cases this new directive will result in some duplication of infrastructure through the addition of more systems and software to protect data. This in turn will result in increased power consumption and generation of e-waste, which runs counter to sustainability goals. Under this new dynamic, it will be more important than ever for operators to look to solutions that are the most efficient in terms of density. From a customer perspective, the best way to approach this challenge is to measure efficiency on a per watt basis.

As we grapple with the issue of ever-increasing data storage needs, which is driving even greater demand on power grids, there’s a great deal of focus from some of the technology sector’s biggest players to find solutions. Microsoft, Google and Amazon think that powering data centers with small-scale nuclear reactors might be a novel solution.

It's now pretty clear that power has become the main limiting factor when it comes to building and expanding data centers, and the increased focus on AI is only making the situation worse, as training and using AI models use significantly more compute and associated power than traditional software.

It is important to recognize there are ways to address power related challenges now, with all-flash storage technology. Within a data center, storage infrastructure generally accounts for 20-25% of energy usage. There is a huge opportunity to adopt high-density flash storage systems, which consume significantly less power compared to legacy HDD-based systems, and even ones based on “off the shelf” commodity SSDs.

The challenge of critical data differentiation

Deeper consideration of what constitutes critical infrastructure raises a parallel question about what constitutes critical data that resides in data centers. Not all data is of equal importance in the context of CNI, with NHS records, telecommunication network security, continuity for facilities like water, and other information which ensures our safety and security surely taking priority over non-critical data such as social media feeds.

Yet this is a very challenging issue to address and one that will require the implementation of agreed frameworks for data classification. Current data protection regulations like GDPR may be helpful in some respects here, as they offer clear guidance on the classification of personal data and the responsibilities of an organization in its handling. GDPR also helps organizations understand the types of data they hold, which is a first and critical step in understanding how that data might be prioritized once frameworks have been agreed. AIOps tools will most likely play a key, vital role in helping organizations looking to better understand the data they hold.

Reaping the benefits of CNI data center designation

In announcing CNI data center designation, the UK government is effectively recognizing a need to better protect the nation’s personal and institutional data. At the same time, they are effectively giving backing to the UK’s data center operators and encouraging more investment in the ecosystem, which is already substantial and predicted to expand over the coming years.

This is borne out in a report issued by techUK in November, after the CNI data center initiative was announced, which predicts over 40,000 new jobs in the sector by 2035 based on increased investment. This is of course very positive. However a critical next step in ensuring the effectiveness of CNI data center designation, and the realization of the UK data center's potential, is the development of an agreed framework on data prioritization classification, increased cyber resiliency and more robust sustainability practices.

We've featured the best green web hosting.

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