TechRadar News

• The Samsung Galaxy Z Fold 7 could be less than 10mm thick when folded
• That would make it the slimmest Z Fold yet, having even the Z Fold Special Edition beat
• However, that would apparently mean it lacks a digitizer, and could have an S Pen "downgrade"

This year, it looks like Samsung might have become a bit obsessed with making its phones as thin as possible – even if doing so brings some downgrades.

Not only is there the upcoming Samsung Galaxy S25 Edge to consider, which has seemingly sacrificed a camera and perhaps also battery capacity to slim down, but now we’re hearing that the Samsung Galaxy Z Fold 7 could be incredibly slim too.

Two separate sources have said as much, with leaker @kro_roe (via GSMArena) claiming that the Galaxy Z Fold 7 will be the slimmest Z Fold ever, thanks in part to an S Pen downgrade.

The Fold 7 is the slimmest Fold ever, thanks to the S Pen downgrade.February 11, 2025

They don't get more specific about how slim or what kind of downgrade, but a report in ET News (via @Jukanlosreve) suggests that Samsung is aiming to make the Galaxy Z Fold 7 less than 10mm thick when folded.

For reference, the Samsung Galaxy Z Fold 6 is 12.1mm thick, and even the Samsung Galaxy Z Fold Special Edition is 10.6mm thick, so this should be a substantial reduction in thickness.

However, to achieve this reduction, Samsung will apparently remove the digitizer from the foldable screen, which is the display layer that interacts with the S Pen stylus. Removing it isn’t necessarily a downgrade, though, as ET News claims that Samsung will simply build technology into the pen itself that allows it to work without a digitizer.

Report: Samsung to Remove Digitizer on the Galaxy Z Fold 7According to multiple industry insiders cited by the media, it has been decided that the Fold 7 will not include a digitizer. They mentioned that in order to reduce the thickness, the digitizer will be omitted, and the…February 13, 2025

Still, since @kro_roe mentions a downgrade, it’s possible that Samsung’s solution will result in a less feature-packed S Pen. Perhaps it will lack Bluetooth connectivity, which is a sacrifice we've already seen Samsung make with the Galaxy S25 Ultra's S Pen, but that's just speculation.

In any case, ET News also claims that the foldable display will be 8 inches, and the cover screen will be 6.5 inches, which would be a match for the Galaxy Z Fold Special Edition but bigger in both cases than the Samsung Galaxy Z Fold 6.

Mass production in May

The Samsung Galaxy Z Fold 6 launched in July 2024 (Image credit: Samsung)

Finally, the site states that the Samsung Galaxy Z Fold 7 will enter mass production in May, by which point Samsung will have finalized the thickness of the phone.

It’s always worth taking claims like these with a pinch of salt, but we’ve heard from multiple sources that the Galaxy Z Fold 7 might lack a digitizer, so there’s a good chance that that much, at least, is correct, and we've also heard those screen sizes before.

Whether the phone really will be less than 10mm thick is more questionable, since previous claims suggested it wouldn’t be any thinner than the Z Fold Special Edition. We’ll probably find out in July, as that’s likely when the Galaxy Z Fold 7 will launch – probably alongside the Samsung Galaxy Z Flip 7, and perhaps even a tri-fold phone.

You might also like

• Best foldable phones: the top folding smartphones right now
• Best Samsung phones: top Galaxy handsets ranked
• Samsung Galaxy S25 Ultra review: the ultimate Android

• Windows 11’s Copilot app has a new feature in testing
• It offers the ability to ‘auto start on login’ for the app
• This could be a handy timesaver for those who use Copilot regularly

Windows 11 has an incoming change for the Copilot app whereby it can be set to automatically load in the background when you start your PC.

PhantomOfEarth, who regularly posts bits and pieces of Windows-related observations and rumors on X, noticed the development.

New Copilot app update for Insiders: 1.25014.121.0, with a new auto start on login (runs in the background) feature. pic.twitter.com/0urRNzmQrWFebruary 10, 2025

As shown in the above post, there’s a new ‘auto start on login’ choice in the Settings for the Copilot app, which when enabled does just that – it automatically starts Copilot (in the background) when your system is fired up.

Right now, the option is still in testing (in version 1.25014.121.0 of the app), but providing there’s no pushback or problems, it should go live for all Windows 11 users before too long.

(Image credit: Microsoft) Analysis: The stumbling journey of the Copilot assistant

You might be thinking ‘who cares’ when it comes to this additional feature for Copilot, and that’s a fair enough point. I don’t imagine usage of the Copilot app is all that widespread, and indeed, I’d be surprised if it wasn’t a niche feature in Windows 11 – but for those people who do make use of the AI, this is still a handy little extra touch.

What it means is that they can invoke the Copilot app with the Alt+Space keyboard shortcut (assuming that’s also enabled), without having to wait for it to load up the first time this action is taken in a new computing session. (Because it will have already loaded up already, in the background).

The good news is that the option isn’t on by default, so Copilot isn’t being forcefully pushed into the background of everybody’s Windows 11 installation. You can either use this option, or just feel free to ignore it.

All in all, it’s a relatively minor change, and as with anything to do with Copilot, I’m waiting for Microsoft to justify its existence in a more convincing manner. There were some big promises of an AI that could make sweeping system-wide changes based on simple requests back at the launch of the Copilot assistant on the desktop. However, all that appears to have been, well, swept under the carpet as time passed by, and Copilot was decoupled from the internals of Windows and made a standalone app.

Maybe Copilot will be realized in this form eventually, but I can’t help but think that this destination feels a long, long, way off, given how things have progressed – or rather haven’t – with the desktop assistant thus far.

Via Windows Latest

You may also like...

• Fed up with Copilot in Windows 11? Microsoft pushing a new keyboard shortcut for the AI isn’t likely to help, and it could mess with your desktop mojo
• Are you unable to get security updates for Windows 11 24H2? Here’s the likely reason why, and the fix to get your PC safe and secure again
• A Chrome-killing browser will be OpenAI's next big shot at Google, according to a new report

• Report claims 80% of execs are worried AI innovation is outpacing governance
• Fewer than one in four have strong AI frameworks, NTT Data finds
• Workers also need to be upskilled to get the most from AI

The rapid pace of development could be the biggest thing we have to worry about when it comes to artificial intelligence, new research has claimed.

A report from NTT Data noted how governance, security and workforce readiness could all be lagging, despite the fact that AI adoption and investments continue to surge.

In fact, four in five (80%) executives agree that leadership and governance are failing to keep pace with AI advancements.

AI is developing too quickly

Already, eight in 10 (80%) business leaders are concerned about unclear regulations that are preventing them from scaling generative AI initiatives, with a further 89% surveyed concerned about AI security risks. That said, fewer than one-quarter (24%) of organizations have a strong AI governance framework, and nearly three-quarters (72%) don’t have a GenAI usage policy.

It’s not all about policies, either, because many business leaders believe their workers might not be ready for full-scale AI adoption. Two in three (67%) execs worry their workers lack the right skills.

Furthermore, three in four (75%) leaders say that their AI ambitions conflict with corporate sustainability goals, forcing them to rethink their strategies entirely.

“The enthusiasm for AI is undeniable, but our findings show that innovation without responsibility is a risk multiplier," noted NTT DATA CEO Abhijit Dubey.

Looking ahead, NTT DATA is calling for AI and generative AI models to be built responsibly from the ground up, integrating security, compliance and transparency at their cores.

The report also highlights the need for ethical guidance and governance to cover social elements, not just legal requirements, as well as the upskilling of employees.

Dubey concluded: “The business community must act now. By embedding responsibility into AI’s foundation – through design, governance, workforce readiness, and ethical frameworks – we unlock AI’s full potential while ensuring it serves businesses, employees, and society at large equally.”

You might also like

• Check out the best AI tools and best AI writers
• White-collar jobs are increasingly being exposed to AI
• Boost your upskilling efforts with the best online learning platforms

• Elden Ring Nightreign is launching on May 30, 2025
• There will be standard and deluxe editions as well as a Collector's Edition
• The Collector's Edition features a trove of physical goodies including a statue

It's great news for FromSoftware fans as the upcoming Elden Ring Nightreign - a multiplayer spin-off of the 2022 mega hit - finally has a release date. What's more, you can pre-order the game or one of its various special editions now.

Elden Ring Nightreign will officially launch on May 30, 2025 for PS5, PS4, Xbox Series X, Xbox Series S and PC via Steam. If you didn't get into the recent closed network tests for the game, at least you won't have to wait much longer to dive in.

You can also head over to the official Elden Ring Nightreign website to check out the various editions as well as pre-order the game for yourself. There are digital and physical editions available, as well as an eye-wateringly pricey Collector's Edition - as we've come to expect for Bandai Namco-published FromSoftware titles.

Here's a rundown of all the available editions, their pricing, and what they include:

• Standard Edition ($39.99 / £34.99): Includes the base game
• Deluxe Edition - Digital ($54.99 / £47.99): Includes the base game, a DLC code and digital art book and soundtrack
• Seekers Edition - Physical ($54.99 / £47.99): Includes the base game, all Deluxe Edition content and a steelbook
• Collector's Edition ($199.99 / £179.99): Includes the base game, all Deluxe Edition content, steelbook, eight Nightfarer cards, a hardcover art book and a 25cm statue of the Wylder Nightfarer character

Overall I'm extremely glad that Elden Ring Nightreign is confirmed to be a mid-budget title rather than a full-priced game. The relatively smaller scale of the game definitely fits that smaller price tag. Unless of course, you're particularly wedded to that Collector's Edition and the 10 inches of Wylder it provides.

If you're unfamiliar with Elden Ring Nightreign, it's a player-versus-environment (PvE) multiplayer titles that can be played solo or with a group of three players (two player co-op isn't supported, actually).

Each run consists of three days, and you'll need to rush through the map collecting loot and defeating enemies, all while avoiding being absorbed by an ever-shrinking circle (a la a battle royale game like PUBG Battlegrounds). The end of each night sees your squad face down a challenging boss before you can proceed to the next day. It sounds like manic fun and definitely a tick up from the slower pace of the main game.

You might also like...

• PlayStation State of Play – everything announced, and the show as it happened
• Sci-fi shooter Metal Eden might be the coolest new game from the latest State of Play
• Currys is the latest UK retailer to let you register interest in Nintendo Switch 2 news and potential pre-order information

In 1847, Hungarian physician Ignaz Semmelweis made a revolutionary yet simple observation: when doctors washed their hands between patients, mortality rates plummeted. Despite the clear evidence, his peers ridiculed his insistence on hand hygiene. It took decades for the medical community to accept what now seems obvious—that unexamined contaminants could have devastating consequences.

Today, we face a similar paradigm shift in artificial intelligence. Generative AI is transforming business operations, creating enormous potential for personalized service and productivity. However, as organizations embrace these systems, they face a critical truth: Generative AI is only as good as responsibility for the data it's built on—though in a more nuanced way than one might expect.

Like compost nurturing an apple tree, or a library of autobiographies nurturing a historian, even "messy" data can yield valuable results when properly processed and combined with the right foundational models. The key lies not in obsessing over perfectly pristine inputs, but in understanding how to cultivate and transform our data responsibly.

Just as invisible pathogens could compromise patient health in Semmelweis's era, hidden data quality issues can corrupt AI outputs, leading to outcomes that erode user trust and increase exposure to costly regulatory risks, known as in integrity breaches.

Inrupt's security technologist Bruce Schneier has argued that accountability must be embedded into AI systems from the ground up. Without secure foundations and a clear chain of accountability, AI risks amplifying existing vulnerabilities and eroding public trust in technology. These insights echo the need for strong data hygiene practices as the backbone of trustworthy AI systems.

Why Data Hygiene Matters for Generative AI

High-quality AI relies on thoughtful data curation, yet data hygiene is often misunderstood. It's not about achieving pristine, sanitized datasets—rather, like a well-maintained compost heap that transforms organic matter into rich soil, proper data hygiene is about creating the right conditions for AI to flourish. When data isn't properly processed and validated, it becomes an Achilles' heel, introducing biases and inaccuracies that compromise every decision an AI model makes. Schneier's focus on "security by design" underscores the importance of treating data hygiene as a foundational element of AI development—not just a compliance checkbox.

While organizations bear much of the responsibility for maintaining clean and reliable data, empowering users to take control of their own data introduces an equally critical layer of accuracy and trust. When users store, manage, and validate their data through personal "wallets"—secure, digital spaces governed by the W3C's Solid standards—data quality improves at its source.

This dual focus on organizational and individual accountability ensures that both enterprises and users contribute to cleaner, more transparent datasets. Schneier's call for systems that prioritize user agency resonates strongly with this approach, aligning user empowerment with the broader goals of data hygiene in AI.

Navigating Regulatory Compliance with the DSA and DMA Standards

With European regulations like the Digital Services Act (DSA) and Digital Markets Act (DMA), expectations for AI data management have heightened. These regulations emphasize transparency, accountability, and user rights, aiming to prevent data misuse and improve oversight. To comply, companies must adopt data hygiene strategies that go beyond basic checklists.

As Schneier pointed out, transparency without robust security measures is insufficient. Organizations need solutions that incorporate encryption, access controls, and explicit consent management to ensure data remains secure, transparent, and traceable. By addressing these regulatory requirements proactively, businesses can not only avoid compliance issues but also position themselves as trusted custodians of user data.

Moving Forward with Responsible Data Practices

Generative AI has tremendous potential, but only when its data foundation is built on trust, integrity, and responsibility. Just as Semmelweis's hand-washing protocols eventually became medical doctrine, proper data hygiene must become standard practice in AI development. Schneier's insights remind us that proactive accountability—where security and transparency are integrated into the system itself—is critical for AI systems to thrive.

By adopting tools like Solid, organizations can establish a practical, user-centric approach to managing data responsibly. Now is the time for companies to implement data practices that are not only effective but also ethically grounded, setting a course for AI that respects individuals and upholds the highest standards of integrity.

The future of generative AI lies in its ability to enhance trust, accountability, and innovation simultaneously. As Bruce Schneier and others have emphasized, embedding security and transparency into the very fabric of AI systems is no longer optional—it's imperative. Businesses that prioritize robust data hygiene practices, empower users with control over their data, and embrace regulations like the DSA and DMA, are not only mitigating risks but also leading the charge towards a more ethical AI landscape.

The stakes are high, but the rewards are even greater. By championing responsible data practices, organizations can harness the transformative power of generative AI while maintaining the trust of their users and the integrity of their operations. The time to act is now—building AI systems on a foundation of well-cultivated data is the key to unlocking AI's full potential in a way that benefits everyone.

Improve productivity with the best AI tools.

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

• Audio-Technica's new cartilage conduction headphones are here
• They feature 20 hours of battery life
• Cartilage conduction uses your ears to transmit sound, in a similar vein to bone conduction

Audio-Technica has today unveiled its brand new Audio-Technica ATH-CC500BT2 cartilage conduction headphones.

The second-generation of its cartilage conduction efforts, these headphones are a great alternative to the best bone conduction headphones on the market, trasmitting sound through the cartilage of the ear rather than the bones in your skull.

It's a nifty way to avoid some of the vibration complaints you might have had with the likes of Shokz, which we find to be an everpresent quirk of the existing bone conduction technology.

Cartilage conduction turns your ears themselves into speakers, letting you hear music and calls all while keeping your ear canals free and open. Not only is this more hygienic than your average pair of workout headphones, it also leaves you free to stay in tune to your surroundings, keeping those runs and bike rides safer.

Audio-Technica ATH-CC500BT2 : The details

(Image credit: Audio-Technica)

As you can see, they look a lot like bone conduction headphones, with a similar open-ear design and lightweight fit. They're made of titanium and silicone and weigh just 32 grams.

Audio comes via AT's PAT.P vibration drivers and its Acoustic Pure Sound Stabilizer, which should eliminate the degradation in sound quality vibration normally delivers.

A real highlight is the 20 hours of battery life from a single charge, with 10-minute quick charge offering up 120 minutes of playback.

You can use them to take calls with a built-in microphone, and there's AI-powered noise reduction technology to make your voice clearer during calls and meetings.

They're IPX4 water resistant, so immune to rain, splashes, and sweat, and can be paired to two Bluetooth devices simultaneously. You'll also get compatibility with Siri, Google Assistant, and Audio-Technica's Connect app. Charging is delivered via USB-C.

The Audio-Technica ATH-CC500BT2 headphones are available now from Audio-Technica.com, retailing at £119 / €139. Unfortunately, those of you in the US are out of luck, as the headphones won't be available there. AU prices are TBC.

You may also like

• The Powerbeats Pro 2 are my new favorite workout headphones, but not because of their heart rate monitor
• Apple just unveiled its most ambitious health study yet – here's how you can take part with just an iPhone
• A Redditor installed a PSP emulator on a Galaxy Watch 5, and now it runs GTA at 60fps and God of War

After waiting two years for The White Lotus season 3, I can't wait to finally make my reservation and find out who will be in the body bag this time around, as a colorful cast of wealthy socialites check in to the infamous luxury hotel chain once again.

Despite the hotel attracting a string of murders in previous seasons of the show, the allure of these extravagant resorts are clearly hard to resist for the rich as the third season of the hit anthology series takes us to a brand new location: Thailand.

The social satire series follows the exploits of a number of guests and employees over the space of a week, with the cast including Leslie Bibb, Carrie Coon, Michelle Monaghan, Walton Goggins, Aimee Lou Wood, Jason Isaacs, Parker Posey, Patrick Schwarzenegger, Blackpink's Lalisa Manobal and Natasha Rothwell reprising her role as salon manager Belinda from The White Lotus season 1.

Show creator Mike White teased that season 3 is going to be "longer, bigger, crazier", as well as being “much, much darker” than the two previous seasons, according to White's interview with Time.

I don't even want to skip the opening credits

Let's be honest here, we're all guilty of skipping the opening credits of a new TV show or movie we've been watching. It's often rare to see a title sequence capture us and make us refrain from pressing the 'Skip' button. However, the same can't be said for The White Lotus.

While I'm excited to see a slew of mysteries and dark secrets against the backdrop of murder and chaos, I've already got The White Lotus theme song blasting through my headphones. What makes the opening credits stand out from the rest of the best Max shows is the fact that each theme song is tailored to reflect the season's location, but always features a consistent melody.

Composer Cristobal Tapia de Veer's Aloha! and Renaissance, served as the show's theme tunes for the first and second season, which were set in Hawaii and Sicily respectively. White's expert storytelling doesn't stop there though, as the enticing digital artworks that accompany Cristobal's danceable beats at the beginning of the episodes offer up plot clues and weave in the show's key themes and characters.

Yes, the actual plot is important, but The White Lotus does the rare thing of making the drama's theme music as popular as its groundbreaking premise and I can't wait to add the third track to my playlist.

The White Lotus season 3 will officially launch on February 16 on Max where we'll once again be greeted by Cristobal's delightful composition and the show's suspenseful yet alluring commentary on wealth, class and exploitation.

You might also like

• We Live In Time is Max’s #1 most-watched movie – here are 7 more powerful A24 films with over 91% on Rotten Tomatoes
• Max unveils first look at Euphoria season 3, but I'm not excited about the hit show's long-awaited return
• I can’t wait to watch these 3 new movies and shows on Max with over 91% on Rotten Tomatoes

Across Europe we need to change our mindset around ‘closing the digital skills gap’. If we spend too much time lamenting it, we'll fall even further behind.

Instead of considering the challenges ahead as a gap that needs bridging in traditional ways, we need to plot an entirely new route - something we can do using tools now available to us like no-code technology. The skills gap no longer exists in the way it used to, because the solutions themselves have changed.

That’s not to deny that in Europe we face a talent drain. Recent years have seen us lose some of our brightest minds to US-based companies, and the reasons for this go beyond the pay packet.

Something we could learn from our cousins across the Atlantic is how to create a more positive culture around failure and the lessons this teaches us. Trying and failing isn’t automatically a bad thing as long as the same mistakes aren’t continually repeated. Crucially, failure indicates learning and is a key aspect of innovation and progress.

Fail fast and hard or fail anyway

The business culture in the US typically recognizes failing hard and fast as a sign of progress, in a way we don’t see in more risk-averse European companies, and this won’t serve us when it comes to finding the most innovative solutions to business challenges. Overall, this discourages entrepreneurship and investment and makes it harder to fund innovation, contributing to talent going elsewhere.

Businesses of all sizes are increasingly faced with the challenge of finding people with the right technical skill set. However, ‘closing’ this skills gap with a fresh cohort of newly trained programmers won’t solve all their problems.

Let’s be specific. In real terms, for most business leaders today concerned with fulfilling orders and keeping customers and shareholders happy, having access to the right data at the right time is key when making decisions. They need a digital skill set in their organizations to drive actionable, meaningful insights. Most realize they can use AI to do this - but that’s often as far as they get.

Leveraging AI is no longer just about understanding that specialty, but instead about a powerful combination of software development skills and understanding of data strategy and integration. And software skills themselves are changing as the tools available to us get ever more sophisticated.

Why innovation starts with IT

In IT we constantly look ahead. Part of our job is to anticipate the next big challenge and the next fix for a problem - but in nearly every role I’ve had in my career, a large part of the job has been dealing with legacy solutions. They’re just not going anywhere. Any new, exciting technology that comes in has to fit alongside whatever existing systems underpin a business’s processes.

No-code solutions are particularly good at doing this as they provide building blocks which work with both older and newer technologies and systems. They allow for seamless integration without extensive coding, making it easier to update and improve legacy systems.

The other thing that’s changed within business technology is that technical decisions are no longer the preserve of the technical experts. A technical foundation is still vital for some jobs, but not for all. These decisions are often now fundamental to the way a business is run, which means they have to be made with wider business understanding and an appreciation for what the customer needs, rather than a purely technical one.

In reality, a background in software development is becoming less of a requirement for solving technical problems - thanks in part to no-code solutions, as well as other AI automation tools. An understanding of how these solutions work in the wider context of the business is crucial for successful implementation and adoption - and for the best results.

Automation and AI tools

The kind of abstraction and automation approach that AI tools and no-code enable goes some way to solving the ‘problem’ of the digital skills gap. Abstracting away the need to program specific fields and providing the ability to make something visual and more easily understood helps open up the playing field to include those with less technical expertise.

For example, a logistics company might need their shipping processes to remain compliant with regulatory requirements but face complicated programming challenges. Through the use of a no-code solution, combined with the business expertise of the staff to produce the kind of business logic necessary to navigate the low margins often faced by mid-market businesses, they can avoid being stuck with problems that can’t be fixed with manual labor alone.

This kind of situation is where data integration is particularly valuable, because from a business point of view, data only means something if it is applied to an actionable challenge or problem in a business context. Abstraction and automation allow people to focus on what matters. IT professionals spend so much time on complicated problems and can forget that they’re helping drive business outcomes. They need to be given the tools to be able to focus in the right way.

Making the most of the innovation opportunity

However, we’re not talking about settling for something that’s second best. This is a brave new world, and making the most of no-code solutions will bring about a whole host of positive, and perhaps even unintended, secondary benefits.

The first is diversity. Computer science is still a field that’s often inaccessible to many, and while this is slowly changing, the simple fact is that no-code is easier to learn and therefore more accessible to people from all backgrounds. The ease of use of the tools also means more people start to explore their possibilities, bringing about more interesting and disruptive solutions used by a wider range of people.

A second overlooked benefit is helping solve challenges around the ageing workforce. Many employers face a knowledge gap when long-time employees retire, taking with them perhaps decades of experience and institutional knowledge. Financial institutions, supply chain and manufacturing companies often possess old systems still generating data. No-code solutions can help fill those gaps when the people who built some of the systems of yesterday retire but the software is still used.

Appreciating benefits like these that extend beyond efficiency helps us leapfrog barriers to adoption more easily. Often, technical experts hesitate when it comes to abstraction because they think it will mean less freedom and flexibility, and therefore they’re resistant. We see this all the time when it comes to finding new ways of doing things - you have to see it (and appreciate how much time it could save you) to believe it.

Our days spent discussing the digital skills gap are numbered. The IT professionals of the future will still, of course, include coders but we’ll see far more convergence of technical and business skills in the workforce of the future. This will, in part, happen organically - young people of today are far more interested in taking ownership of their digital journey - but it will no doubt be aided by more widespread use of AI tools and solutions that automate certain tasks and processes faster than you can say ‘abstraction’.

We've compiled a list of the best mobile app development 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

I've enjoyed pitting various AI chatbots against each other. After comparing DeepSeek to ChatGPT, ChatGPT to Mistral's Le Chat, ChatGPT to Gemini 2.0 Flash, and Gemini 2.0 Flash to its own earlier iteration, I've come back around to match DeepSeek R1 to Gemini 2.0 Flash.

DeepSeek R1 sparked a furor of interest and suspicion when it debuted in the U.S. earlier this year. Meanwhile, Gemini Flash 2.0 is a solid new layer of ability atop the widely deployed Google ecosystem. It is built for speed and efficiency and promises quick, practical answers without sacrificing accuracy.

Both claim to be cutting-edge AI assistants, so I decided to test them from the perspective of someone with a casual interest in using AI chatbots in their everyday lives. Both have shown themselves to be effective at a basic level, but I wanted to see which one felt more practical, insightful, and actually helpful in everyday use. Each test has a screenshot with DeepSeek on the left and Gemini 2.0 Flash on the right. Here’s how they did.

Local Guide

(Image credit: Screenshots of Google Gemini/DeepSeek)

I was keen to test the search abilities of the two AI models combined with insight into what is worthwhile as an activity. I asked both AI apps to "Find some fun events for me to attend in the Hudson Valley this month."

I live in the Hudson Valley and was aware of some things on the calendar, so it would be a good measure of accuracy and usefulness. Amazingly, both did quite well, coming up with a long list of ideas and organizing them thematically for the month. Many of the events were the same on both lists.

DeepSeek included links throughout its list, which I found helpful, but the descriptions were just quotes from those sources. Gemini Flash 2.0's descriptions were almost all unique and frankly more vivid and interesting, which I preferred. While Gemini didn't have the sources immediately available, I could get them by asking Gemini to double-check its answers.

Reading tutor

(Image credit: Screenshots of Google Gemini/DeepSeek)

I decided to expand on my usual test for AI's ability to offer advice on improving my life advice with something more complex and reliant on actual research. I asked Gemini and DeepSeek to "Help me devise a plan for teaching my child how to read."

My child isn't even a year old yet, so I know I have time before he's paging through Chaucer, but it's an aspect of parenthood I think about a lot. Based on their responses, the two AI models might as well have been identical advice columns. Both came up with detailed guides for different stages of teaching a child to read, including specific ideas for games, apps, and books to use.

While not identical, they were so close that I would have had trouble telling them apart without the formatting differences, like the recommended ages for the phases from DeepSeek. I'd say there's no difference if asked which AI to pick based purely on this test.

Vaccine superteam

(Image credit: Screenshots of Google Gemini/DeepSeek)

Something similar happened with a question on simplifying a complex subject. With kids on my mind, I explicitly went for a child-friendly form of answer by asking Gemini and DeepSeek to "Explain how vaccines train the immune system to fight diseases in a way a six-year-old could understand."

Gemini started with an analogy about a castle and guards that made a lot of sense. The AI oddly threw in a superhero training analogy in a line at the end for some reason. However, similarities in training to DeepSeek might explain it because DeepSeek went all in on the superhero analogy. The explanation fits with the metaphor, which is what matters.

Notably, DeepSeek's answer included emojis, which, while appropriate for where they were inserted, implied the AI expected the answer to be read from the screen by an actual six-year-old. I sincerely hope that young kids aren't getting unrestricted access to AI chatbots, no matter how precocious and responsible their questions about medical care might be.

Riddle key

(Image credit: Screenshots of Google Gemini/DeepSeek)

Asking AI chatbots to solve classic riddles is always an interesting experience since their reasoning can be off the wall even when their answer is correct. I ran an old standard by Gemini and DeepSeek, "I have keys, but open no locks. I have space but no room. You can enter, but you can’t go outside. What am I?"

As expected, both had no trouble answering the question. Gemini simply stated the answer, while DeepSeek broke down the riddle and the reasoning for the answer, along with more emojis. It even threw in an odd "bonus" about keyboards unlocking ideas, which falls flat as both a joke and insight into keyboards' value. The idea that DeepSeek was trying to be cute is impressive, but the actual attempt felt a little alienating.

DeepSeek outshines Gemini

Gemini 2.0 Flash is an impressive and useful AI model. I started this fully expecting it to outperform DeepSeek in every way. But, while Gemini did great in an absolute sense, DeepSeek either matched or beat it in most ways. Gemini seemed to veer between human-like language and more robotic syntax, while DeepSeek either had a warmer vibe or just quoted other sources.

This informal quiz is hardly a definitive study, and there is a lot to make me wary of DeepSeek. That includes, but is not limited to, DeepSeek's policy of collecting basically everything it can about you and storing it in China for unknown uses. Still, I can't deny that it apparently goes toe-to-toe with Gemini without any problems. And while, as the name implies, Gemini 2.0 Flash was usually faster, DeepSeek didn't take so much longer that I lost patience. That would change if I were in a hurry; I'd pick Gemini if I only had a few seconds to produce an answer. Otherwise, in spite of my skepticism, DeepSeek R1 is as good or better than Google Gemini 2.0 Flash.

You might also like

• I discovered a surprising difference between DeepSeek and ChatGPT search capabilities
• Yes, Google's new Gemini 2.0 Flash is much better than the old 1.5 model – here's what I found
• I matched Mistral AI's new Le Chat app against ChatGPT, and it did surprisingly well

• Google Gemini has added its Deep Research AI model to iPhones
• Deep Research searches the web, compiles, and reports back to Gemini Advanced subscribers
• The information is refined and organized into results viewable on Google Docs

iPhone users who love Google’s Gemini AI assistant have a new tool to help them condense information from the internet. The tech giant has added its Deep Research feature to iOS devices, at least if you’re a Gemini Advanced subscriber. Deep Research debuted on Gemini's web portal in December and rolled out to Android users earlier this month.

Deep Research is an AI-powered tool for compiling information. Essentially, it takes the classic Google search collecting of links and extends it several steps to read what's at those links and organize what it finds into something useful. It’s the first “agentic” feature in Gemini, meaning the AI is more proactive and doesn't just answer questions; it carries out an entire research project.

If you do subscribe to Gemini Advanced and have the app on your iPhone, you can switch to Deep Research by picking it from the model list. Select “1.5 Pro with Deep Research” near the newer experimental 2.0 Flash model. Then, just ask Gemini a research question, something big and messy and complicated if you want to test its limits. Gemini will respond with a step-by-step plan it will undertake on your behalf. You can tweak the approach if you don’t like it, deleting parts or adjusting the focus, and then tap the “start research” button. Gemini then heads off on a digital scavenger hunt, digging through sources, running multiple searches, and refining its findings in real-time.

Don't expect instant answers. It can take between five and ten minutes to complete, longer if it's an especially difficult topic. You don’t have to babysit it, though. Gemini will send you a notification when the work is done, and you can check your chat history later to review the results. Once it’s ready, you’ll get a structured report, complete with sections, tables if necessary, and a full list of sources. And if you want to make it look even more official, you can export it directly to Google Docs.

Of course, Google isn’t letting you run wild with infinite research requests. There are daily limits, and the app will politely remind you how many you have left, just in case you were planning to outsource your entire workload to AI. Right now, Deep Research runs on Gemini 1.5 Pro, but Google has hinted that it will eventually move to the more powerful 2.0 Pro once that model exits its experimental phase.

Ads AI

The launch of Deep Research on iPhones matters for more than just access reasons. The pitch to iOS users signals how aggressively Google is leaning into Gemini for all kinds of productivity demands. Unlike a standard chatbot response, which can sometimes feel like a slightly smarter autocomplete, Deep Research attempts to simulate how a human researcher breaks down a topic, refines their findings, and presents something meaningful.

It’s also a direct response to how rival AI companies are pursuing similar goals. OpenAI's version of this kind of feature even shares the Deep Research Name. But, with OpenAI, Meta, and Apple working on more advanced AI-driven assistants, Google is clearly betting that features like Deep Research will make Gemini a must-have.

Deep Research is, in some ways, just another tool in Google’s AI arsenal, but it's a powerful one. It's easy to see the appeal to people who feel like they are drowning in information when online. It may not remake the experience of looking things up online, but it could set Gemini up as the crowning jewel for future digital research projects.

You might also like

• Yes, Google's new Gemini 2.0 Flash is much better than the old 1.5 model – here's what I found
• Google is bragging about Gemini 2.0 Flash, but how does it compare to ChatGPT for casual AI chatbot users?
• Google Gemini's new model is the brainstorming AI partner you've been looking for

• Sci-fi shooter Metal Eden has been revealed
• It features FPS combat and fast-paced parkour
• It is set to launch on May 6, 2025

As part of today's State of Play, developer Reikon Games has unveiled Metal Eden, a new sci-fi first-person shooter (FPS) that looks like an incredible mix of Doom and Ghostrunner.

Set in a dark sci-fi universe where robots are powered by cybernetic Cores that contain digitized human souls, you play as the advanced android Aska as she fights to save a lost human colony. Throughout your journey, you will be accompanied by a sentient spaceship, Nexus, that gradually reveals more about the world and your purpose in it.

The combat system mixes powerful weapons with the ability to rip out enemy energy cores and use them to boost your own attacks. This is on top of plenty of fast-paced parkour, which sees you dashing, grappling, and wall-running your way around its eight futuristic levels.

You can see the trailer and get an early glimpse at the game for yourself below.

All things considered, it looks pretty impressive - especially in the visual department. It's definitely a massive step up from the studio's previous release, Ruiner, which, while very enjoyable, had a much more of an isometric, indie look.

I'm definitely keen to try Metal Eden when it launches for PC, PlayStation 5, Xbox Series X, and Xbox Series S on May 6, 2025.

You might also like...

• PlayStation State of Play live – the announcements as they happened
• Citizen Sleeper 2: Starward Vector is an immersive sci-fi tale about the strengths of community
• I’ve used the PS5 Slim for months and it’s an excellent evolution of the original PS5, offering the same power and performance in a slimmer, neater package

• Ultra-low GPU temps sound impressive until you consider the power bill
• Cooling a GPU with an AC unit may work, but the size and risks make it impractical
• A 12,000 BTU AC for GPU cooling: it works, but at what cost?

A Chinese techtuber has integrated a 12,000 BTU air conditioning unit into the liquid cooling system of the RTX 4090 and an Intel Core i9-13900K processor.

ESC’s setup replaces a conventional PC-mounted radiator with a household air conditioning unit. The modified Xiaomi KFR-35GW air conditioner serves as the primary cooler component while the system directs chilled liquid through a loop connected to the PC.

ESC plans to pair a Intel Core i9-14900K processor paired with the incredibly elusive Nvidia GeForce RTX 5090 GPU, but the lack of an available water block for the RTX 5090 has put a - ahem - dampener on that.

AC + CPU + GPU = R U OK?

In initial tests, the RTX 4090 maintained a temperature of just 20°C under stress testing. When the air conditioning unit was moved outdoors, GPU temperatures remained exceptionally low, reaching 2°C at idle and 20°C under load.

So, air-con cooling demonstrates impressive cooling performance, but there are several practical concerns.

First, integrating an HVAC system into a PC is an involved process, involving welding, refrigerant management, and custom piping. That's just not suitable for the average user, and introduces risks related to handling refrigerant gases and electrical components. On the off chance you even want to try this at home; don't.

Another concern is energy consumption. The air conditioning unit in this setup typically draws around 1.2 kW of power, which is significantly higher (and more expensive) than traditional PC cooling solutions.

Extreme cooling can lead to condensation, which poses a serious risk to the longevity of workstation components; without proper insulation and moisture control, the system could suffer from unintended damage due to a build-up of water in the chassis.

Consider, also, the size of the AC unit, which measures 765 x 268 x 550mm, significantly larger than a 360mm PC cooling radiator and even bigger than the PC tower it is cooling. It's incredibly impractical, and so I can't see this as anything other than a click-baiting hobby project.

Via Tom's Hardware

You may also like

• How data centers can use liquid above ground to stay cool
• Evolving data center cooling for AI workloads
• Absurdly powerful PC with 7 liquid-cooled Nvidia RTX 5090 GPUs has just gone on sale — and it is in stock

• Volkswagen has teased an image of its upcoming entry-level EV
• The new car will kick off the marque’s ID range
• A show car will be unveiled at the end of March

Electric vehicles are typically considered the preserve of early adopters and high earners, but the price of everyday, pure-battery EVs is tumbling – and Volkswagen is the latest to announce the introduction of its most affordable electrified model to date.

Scheduled to go on sale in 2027, we will see a preview of the so-called ID.One in a month or so, with the German marque confirming that it plans to sell the small hatchback for €20,000 in Europe, which is just £17,000 in the UK or around $20,800 / AU$33,000 at current exchange rates.

Solely aimed at European buyers (small cars rarely do well in the US), the new model will sit underneath the upcoming ID.2all (yes, that's its name) electric car that is planned for next year.

However, the official name hasn't been revealed yet and we could well see a return of the Up! badge, as this everyday EV has been designed to fill the void that the Up! left, offering “low-cost entry-level mobility” to the masses.

Alongside the ID.2 all, which is expected to cost around €2,000 more, Volkswagen hopes it can turn around some of its patchy EV fortunes with a handful of attractive small cars that the majority of buyers can afford.

The early teaser image looks good, revealing a squat, purposeful front end and clear Volkswagen design cues. The German marque will bring its reputation for quality but spare some of the additional niceties in order to keep production costs down.

As a result, we expect the battery range to be modest, designed primarily with city driving in mind, while the in-car technology could be kept to a minimum.

Speaking to Autocar, Volkswagen technical development boss Kai Grünitz hinted that customers might use their own devices, rather than having "a huge infotainment system", for example.

Analysis: VW makes EVs affordable, but is it soon enough?

(Image credit: Volkswagen)

There has been a noticeable shift in the sort of electric vehicles that are coming onto the market in recent months, with larger, more luxurious EV launches becoming far less frequent, as more manufacturers turn towards entry-level models.

In Europe, the Renault 5 E-Tech has captured the imagination of many, scooping numerous awards from industry panels, as well as outselling the Tesla Model Y mere weeks after it was launched in France.

Similarly, Citroen has just announced the new C3 Aircross, which will come with the option of a small petrol engine, but is also offered in a pure EV version that starts at just £22,990 (around $28,500 / AU$45,500).

Below this, the Dacia Spring has been making waves with its ridiculously low £14,240 asking price (around $18,000 / AU$28,600), although Dacia is a brand that isn’t afraid to offer the absolute barebones as it chases affordability.

On the other hand, Volkswagen has a long and successful history of offering entry-level cars that look and feel like more expensive propositions, getting buyers on board at an early age with a view that they will remain VW customers for life.

This €20,000 car has the potential to do what the Polo and Up! have done for the brand in previous decades, but 2027 feels like a lifetime away.

If Renault continues its retro-inspired EV success with the promise of even smaller and more affordable models to come, it could well beat the Germans to it.

You might also like

• I’ve driven the new Renault 5 – and it's finally the first small, fun EV for the masses
• Ferrari reveals 2025 launch date for its first fully electric car – here's everything we know
• Millions of cheap EVs in China will get advanced self-driving features this year – leaving the rest of the world far behind

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 Wednesday's puzzle instead then click here: Quordle hints and answers for Wednesday, February 12 (game #1115).

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 #1116) - hint #1 - Vowels How many different vowels are in Quordle today?

• The number of different vowels in Quordle today is 3*.

* 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 #1116) - 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 1.

Quordle today (game #1116) - 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 #1116) - 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 0.

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 #1116) - hint #5 - starting letters (2) What letters do today's Quordle answers start with?

• S

• D

• A

• T

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

Quordle today (game #1116) - the answers

(Image credit: Merriam-Webster)

The answers to today's Quordle, game #1116, are…

• SCALP
• DWELL
• AMPLE
• TRUNK

Saved by the bell today – the bell in my head going “Nooooooo!”. With one guess left I was typing out Dwelt, as that was the only word I could think of that began D-W-E-L when the alarm went off, I woke up and realised I couldn’t use T, and the word I was looking for was DWELL.

I’m sure some people find Quordle a doddle, but I’m not one of them.

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

Daily Sequence today (game #1116) - the answers

(Image credit: Merriam-Webster)

The answers to today's Quordle Daily Sequence, game #1116, are…

• GREET
• CONIC
• GUAVA
• BORNE

Quordle answers: The past 20

• Quordle #1115, Wednesday 12 February: SHOOK, GRADE, POLAR, SEEDY
• Quordle #1114, Tuesday 11 February: HEAVY, CIRCA, PESKY, SCION
• Quordle #1113, Monday 10 February: TWIXT, FRESH, GUISE, TABBY
• Quordle #1112, Sunday 9 February: TAPIR, SHAKE, TOKEN, SEVEN
• Quordle #1111, Saturday 8 February: AFOOT, ALIKE, HUMUS, TOWEL
• 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

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 Wednesday's puzzle instead then click here: NYT Strands hints and answers for Wednesday, February 12 (game #346).

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

• Today's NYT Strands theme is… It's in the cards

NYT Strands today (game #347) - hint #2 - clue words

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

• IRATE
• SOAPY
• TWIN
• GANG
• REAR
• TANG

NYT Strands today (game #347) - hint #3 - spangram What is a hint for today's spangram?

• Mark the occasion

NYT Strands today (game #347) - hint #4 - spangram position What are two sides of the board that today's spangram touches?

First side: top, 4th column

Last side: bottom, 3rd column

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

NYT Strands today (game #347) - the answers

(Image credit: New York Times)

The answers to today's Strands, game #347, are…

• BIRTHDAY
• BABY
• WEDDING
• SYMPATHY
• CONGRATS
• LOVE
• SPANGRAM: GREETINGS

• My rating: Easy
• My score: Perfect

Tomorrow is Valentine’s Day and one of the biggest days of the year for makers, creators and purveyors of GREETINGS cards.

I get why you would send a card to someone you wanted to be in a relationship with. However, in order to make the most from their biggest shot at turning a profit outside of Christmas, companies such as Hallmark have convinced us that we need to tell our partners we love them in card format – even if we see them every day and can tell them our feelings in a way that’s much more personal than sending them a folded piece of paper with a picture of a train beside the words “I CHO-CHOSE YOU”.

The alternative view is that greetings cards are a nice thing and who wouldn’t want a card telling them they’re loved and worth sticking with, despite it all. Signed The Romance Grinch.

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

Yesterday's NYT Strands answers (Wednesday, 12 February, game #346)

• TRAIN
• TOOK
• GIRL
• TOWN
• SMALL
• LONELY
• WORLD
• MIDNIGHT
• SPANGRAM: JOURNEY

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 Wednesday's puzzle instead then click here: NYT Connections hints and answers for Wednesday, February 12 (game #612).

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 #613) - today's words

(Image credit: New York Times)

Today's NYT Connections words are…

• GUILT
• MIGHT
• MOLD
• EGO
• POWER
• EXPRESSION
• TOKEN
• MUST
• HEAD
• WILL
• GUIDE
• GESTURE
• CAN
• MODEL
• SYMBOL
• BLUEPRINT

NYT Connections today (game #613) - hint #1 - group hints

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

• YELLOW: A pattern to follow
• GREEN: Giving regards 
• BLUE: A probability
• PURPLE: Add a word that rhymes with hip

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 #613) - hint #2 - group answers

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

• YELLOW: TEMPLATE 
• GREEN: DEMONSTRATION, AS OF APPRECIATION 
• BLUE: MODAL VERBS 
• PURPLE: TRIP 

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

NYT Connections today (game #613) - the answers

(Image credit: New York Times)

The answers to today's Connections, game #613, are…

• YELLOW: TEMPLATE BLUEPRINT, GUIDE, MODEL, MOLD
• GREEN: DEMONSTRATION, AS OF APPRECIATION EXPRESSION, GESTURE, SYMBOL, TOKEN
• BLUE: MODAL VERBS CAN, MIGHT, MUST, WILL
• PURPLE: __ TRIP EGO, GUILT, HEAD, POWER

• My rating: Moderate
• My score: 1 mistake

I struggled with Purple and Blue today. Thinking that the link was about drive and desire, I made up a group containing MIGHT, WILL, EGO, and POWER before seeing the much more obvious MODAL VERBS.

I wonder if the art of compiling Connections is to make up a list of five words and then take the easiest and most obvious word out. If __ TRIP had included Road, then I think I would have got it. Although I may just be on an EGO trip.

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

Yesterday's NYT Connections answers (Wednesday, 12 February, game #612)

• YELLOW: DOCUMENTS OF OWNERSHIP CERTIFICATE, DEED, RECEIPT, TITLE
• GREEN: BITS IN A VARIETY SHOW DANCE, MONOLOGUE, SKETCH, SONG
• BLUE: PROPS FOR CHARACTERS ON "THE SIMPSONS" DONUT, PACIFIER, SAXOPHONE, SKATEBOARD
• PURPLE: STARTING WITH "SHAKE, RATTLE AND ROLL" ANDROID, RATTLESNAKE, ROLLERBLADE, SHAKESPEARE

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.

• Apple has officially launched the Apple TV app on Android devices
• As of today, you can download the app from the Google Play Store
• MLS Season Pass will also be supported via Google Play's billing system

Starting today (February 12), Apple's TV app will be available to download off the Google Play Store, giving Android customers a very welcome surprise.

The app brings all the same great features that iOS users will be familiar with, like access to Apple TV+ and MLS Season Pass, and is also available in over 100 regions globally, including the US, UK, and Australia.

That means you'll be able to sign in or sign up for Apple TV+ and MLS Season Pass through the app just like you would on any iOS device. But be aware that your device (whether that's a phone, tablet, or foldable) has to support Android 10 or later to be able to use it, especially to ensure it works properly with the Google TV living room experience.

Don't have an Apple TV+ account? That's not a problem because you'll be able to sign up for an Apple TV+ free trial to try out the app for seven days before registering for a subscription.

What do we know so far about Apple TV's Android app?

The Apple TV+ homepage is coming to an Android phone near you. (Image credit: Apple)

Although the Apple TV app has been optimized for Android, a lot of the same great features are likely to still here, including the ability to see all the movies and shows you're currently watching across different devices in one place under the section 'Continue Watching'. You'll also still find your 'Watchlist' of saved movies and shows that you've saved to watch in the future.

The homepage largely looks the same as it does on any iOS device, with several rows of grouped content to create collections based on whether a movie or show is new, trending, the most-watched, or available to stream now. A dedicated section called 'Must See Hits' on Apple TV+ will likely replace the existing 'For You' section that most iOS users see. Most notably, the 'Store' tab at the bottom of the app is missing because iTunes store content will not be made available.

This slight change in the categorization of its content is one of our biggest indications that Apple is pursuing this Android rollout to attract new customers to its streaming service, and it makes a lot of sense when you consider that it has the smallest subscriber base of all the best streaming services.

Apple TV branches out

You'll be able to stream Friday Night Baseball via the Apple TV app on Android devices. (Image credit: Apple)

The Apple TV app's Android rollout comes at a great time for the company, as it gives more people the ability to finally stream hit shows like Severance season 2 as well as new Apple TV+ movies like The Gorge, which is releasing this Friday (February 14) on Valentine's Day.

It's also just in time for Major League Soccer's 30th season, which kicks off next weekend (on February 22), allowing those signed up for MLS Season Pass to get in on all the action with access to exclusive content, analysis, and coverage.

If you're a sports fan, you'll also be able to catch weekly games from Friday Night Baseball and Sunday Night Soccer, which is new to Apple TV+ subscribers in 2025, as well as all the best Apple TV+ shows and best Apple TV+ movies.

Make sure to check back in with TechRadar to find out how the Android version of the app performed compared to its iOS counterpart and what other differences we've spotted between the two apps.

In the meantime, let us know what you think about the Apple TV app on Android in the comments below.

You might also like

• There are two new ways to stream Apple's MLS Season Pass this year
• Viewers are flocking to Apple TV Plus – and it’s all thanks to Lionel Messi
• Why Apple TV Plus beat Netflix for our Streaming Service of the Year award

• Intel releases its sixth annual Product Security Report, outlining how it addresses security vulnerabilities
• Intel also slammed AMD and Nvidia, sharing unflattering stats
• Claims AMD has 78 vulnerabilities with no planned fixes, Nvidia has only high-severity security bugs

Intel has heavily criticized its two biggest competitors - AMD, and Nvidia - for alleged wide-ranging security troubles.

Its 2024 Product Security Report, the sixth such annual missive, discusses security challenges, how they were addressed, and why Intel has a “competitive edge” in security assurance. For its own solutions, Intel said almost all of the new vulnerabilities found in 2024 (96%) were discovered internally, adding that all of the hardware issues found were also discovered by its own research team.

Then it turned its attention to its rivals - claiming AMD reported 4.4 times more firmware vulnerabilities in its hardware root-of-trust, and 1.8 times more firmware vulnerabilities in its confidential computing technologies compared to Intel. It added that AMD discovered only around half (57%) of all the reported platform vulnerabilities, and that the company has 78 flaws with no fixes planned.

Protect yourself from identity theft online

Go Incogni and get 55% off using code TECHRADAR. Incogni erases you and your family from the sites that expose your personal information to identity thieves and robocalls.View Deal

Battle of the giants

For Nvidia, Intel said the firm reported 18 high-severity vulnerabilities, 13 of which were remote code execution flaws. At the same time, Intel’s GPUs only had 10 issues reported, one of which was a severe bug.

Both AMD and Nvidia have not yet commented on the report.

Reports like these should always be taken with a grain of salt since they’re one sided, but especially since Intel is competing with AMD and Nvidia in different fields.

When it comes to the CPU market, Intel is still the domineering force. However, AMD increased its market share by 5.7% in Q3 2024. On the GPU front, Nvidia might be the undisputed champion, but Intel has recently released Gaudi 3 as a competitor. Unfortunately for the company, it failed to hit its target and canceled its next-gen Falcon Shores lineup.

Via Tom's Hardware

You might also like

• This major firmware flaw is affecting Intel-powered PCs across the world
• We've rounded up the best password managers
• Take a look at our guide to the best authenticator app

• Researchers find a faster way to etch deep holes for 3D NAND
• Plasma-based cryo-etching technique doubles etch speed, improving efficiency
• Faster etching might mean cheaper storage, but real-world impact is TBD

3D NAND flash memory is different from traditional single-layer NAND because it vertically stacks memory cells to cram more storage into smaller spaces.

The process involves carving precise, deep holes into alternating layers of silicon oxide and silicon nitride, and this has always been a bit slow, until now.

A team of researchers from Lam Research, the University of Colorado Boulder, and the U.S. Department of Energy's Princeton Plasma Physics Laboratory (PPPL) has developed a plasma-based technique that can etch the deep, narrow holes required for 3D NAND memory at a much faster rate, a paper published in the Journal of Vacuum Science & Technology A claims.

Will it benefit end users?

The team’s approach uses a cryogenic etching process using hydrogen fluoride plasma rather than the traditional method.

"Cryo etch with the hydrogen fluoride plasma showed a significant increase in the etching rate compared to previous cryo-etch processes, where you are using separate fluorine and hydrogen sources," said Thorsten Lill of Lam Research. Using the new method saw etching rates for the layers soar from 310 nanometers per minute to 640 nanometers per minute - more than doubling the efficiency.

"The quality of the etch seems to have improved as well, and that's significant," Lill added.

The researchers also looked at the impact of phosphorus trifluoride. Adding it during the process quadrupled the etch rate for silicon dioxide, but it only had marginal impact on the silicon nitride layer. They also looked at ammonium fluorosilicate, a chemical which forms during the etching process when silicon nitride reacts with hydrogen fluoride. This slows down the etching process, but adding water was found to counteract this.

While the technical achievement is to be applauded, the practical implications are less clear. Faster, better etching rates may simplify and speed up production, but whether these savings trickle down to better or cheaper storage devices remains to be seen.

“Most people are familiar with NAND flash memory because it’s the kind that is in the memory cards for digital cameras and thumb drives. It is also used in computers and mobile phones. Making this type of memory denser still - so that more data can be packed into the same footprint - will be increasingly important as our data storage needs grow due to the use of artificial intelligence,” explained Igor Kaganovich, a principal research physicist at PPPL.

You might also like

• End of the NAND layers race: innovation across vectors
• Micron's budget DRAMless SSD could end low performance SATA drives
• Scientists inch closer to holy grail of memory breakthrough

• Microsoft and Carnegie Mellon University published research that says AI is causing “the deterioration of cognitive faculties”
• The study involved 319 people who use AI tools at work at least once per week
• The survey-based study opens the door to deeper research into AI side effects

Brain rot is usually associated with endless doom scrolling of low-quality online content, but a new Microsoft Research study suggests that overusing AI might be causing “the deterioration of cognitive faculties.” Oh no.

Performed in collaboration with Carnegie Mellon University, the study – titled The Impact of Generative AI on Critical Thinking: Self-Reported Reductions in Cognitive Effort and Confidence Effects From a Survey of Knowledge Workers – asked 319 participants who use “GenAI tools at work at least once per week” (such as Gemini, ChatGPT, CoPilot) to judge how AI usage at work was affecting their critical thinking skills.

It found that “while GenAI can improve worker efficiency, it can inhibit critical engagement with work and can potentially lead to long-term overreliance on the tool and diminished skill for independent problem-solving.” The study added that people's jobs are becoming less about solving problems and more about making sure the AI has correctly solved the problem – morphing our work from “task execution to task stewardship” – which led people to feel like their ability to think critically isn’t as sharp as it was before they started using AI.

Is it time to panic? Not quite yet

AI might still be a help not a hindrance (Image credit: Pixabay)

Seeing the headlines and reading the study it can feel like it’s time to pull the big AI lever and shut it all down for good in order to save our brains from being irreparably damaged by a tool that may be hurting us more than it’s helping us. But while the study certainly highlights challenges we desperately need to tackle in the AI space – ultimately the relationship between humans and AI we should be cultivating – it’s not quite as dire as it seems.

That’s because the study focuses on how AI users perceive AI usage is affecting their critical thinking. While the study uses a survey to quantify these feelings it comes down to how people feel and, more importantly, it doesn’t create a true quantitative comparison between frequent AI users and people who never or rarely use the AI.

While people who use AI a lot may feel they’re less capable of critical thinking it could be that because they’re not as distracted by less difficult tasks they can more easily tackle challenging ones thanks to AI’s help – or it could be there’s been no change in their critical thinking skills. But to measure actual changes in critical thinking we’d need a study with quantitative tests that compare different groups' ability (groups decided based on their AI usage) to complete tasks – including a control who never uses AI for work.

Until we have that kind of study it’s impossible to judge if AI really is making us dumber. However, that doesn’t mean we should just dismiss this Microsoft research.

Make AI work for you (Image credit: Shutterstock/SomYuZu)

Firstly, using this study as inspiration for a more quantitative study like the one I’ve described would be a fantastic next step. Secondly, while the participants' feelings may or may not represent factual reality, they represent reality for the people in this study – and they shouldn’t be dismissed.

If people feel like they’re simply AI stewards rather than performing meaningful work that could lead to dissatisfaction in their job. Some psychologists believe dissatisfaction with how we spend our time scrolling is the root cause of the social media brain rot experience (via the BBC). A similar sense of meaninglessness could be causing AI users’ similar brain deterioration feelings at work.

With AI becoming more prevalent we’ll need to work out how best to fight back feelings of deterioration, but a good first step will always be some self-care. Be that finding a new job that stimulates us, taking up a hobby that we can devote our freed-up brain power towards, or some other strategy that works best for you.

You might also like

• Why is AI in everything these days?
• OpenAI's ChatGPT Super Bowl ad was created entirely by humans
• Sam Altman says AI is progressing faster than Moore’s law as he predicts AGI is ‘coming into view’, and it's leaving me worried about the future