Technology

• Oscar Nuñez will reprise his beloved The Office role in Peacock's new series
• The series doesn't have a name yet
• It is believed that producers wanted to keep Nuñez's appearance a surprise, but it was leaked

The Office was an absolute masterclass of comedy, and I consider it one of my favorite shows. I'll admit, I did feel a little apprehensive when I found out it was getting a Peacock spin-off show, but then again, it can't be as bad as the disastrous Prime Video attempt at The Office, right? Anyway, my interest in the new series has been piqued because it has just been revealed that Oscar Nuñez will be reprising his role as Oscar Martinez, and I'm so excited to see that character back again.

When it comes to Oscar's return, we know that he won't be the central character, so don't expect any Michael Scott style antics, we'll be seeing someone new in that role. Sadly, if you were hoping to see the likes of Pam, Jim, and Dwight, it seems Oscar is the only one returning to the office, and he'll be joining a new line-up of actors. As we understand it, this spin-off is set in the Office universe, but it's not a direct sequel.

According to Variety, producer Greg Daniels felt he "told a complete story for most of the characters — but if there was one character he’d like to revisit and explore, it would be Oscar.

Variety added: "It’s believed that producers were hoping to keep Nuñez’s involvement a surprise, but the news first leaked Wednesday via newsletter The Insneider."

What do we know about Peacock's The Office spin-off?

Right now, we don't know much! It doesn't even have a name and is currently known as the 'Untitled Daniels/Korman Project' (referencing the two producers Greg Daniels and Michael Korman). Hopefully we'll start to see some more details over the coming months though, as I am curious about this one.

We do have a loose plot, however, which is that we'll follow the staff of a dying Midwestern newspaper whose publisher is trying to revive it with volunteer reporters. So while it's not Dunder Mifflin, we've still got the paper connection, and it's going to be set up like a documentary using the same crew who filmed The Office. So this could explain why Oscar has found himself here, whether he's working for the new company or he's simply got experience being a documentary subject.

You might also like

• I'm obsessed with these fan-made Severance title sequences that parody The Office, Parks and Recreation, and Succession
• 4 horror movies on Peacock with 85% or higher on Rotten Tomatoes
• Wicked conjures up a March streaming release date on Peacock, and I'm sad that I won't be able to watch it before the 2025 Oscars

• Researchers find a new variant of the Vo1d botnet
• It seems to be designed to be an anonymous proxy
• At peak, it counted almost 1.6 million devices

If you are an Android TV user, take note - there is a new and dangerous botnet infecting endpoints left and right.

Cybersecurity researchers from Xlab have started tracking a new variant of the Vo1d malicious botnet which, in a span of just a couple of months, grew to roughly 1.6 million devices across 226 countries. The botnet’s size varies from day to day, and while it peaked in mid-January 2025, it currently counts around 800,000 devices, the researchers said.

The initial infection vector is unknown at the current time, but the majority of the victims are located in Brazil (25%), followed by South Africa (13.6%), Indonesia (10.5%), Argentina (5.3%), Thailand (3.4%) and China (3.1%).

Botnet for hire

A botnet can be used for many things, including Distributed Denial of Service (DDoS) attacks, residential proxies, ad manipulation, and more. In this case, Vo1d is being used as an anonymous proxy, redirecting criminal traffic and blending it in with legitimate, consumer traffic. It comes with advanced encryption, strong infrastructure powered by DGA, and state-of-the-art obfuscation techniques.

Since the number of infected devices varies greatly from day to day, the researchers believe the criminals are “renting” devices as proxy servers.

“We speculate that the phenomenon of "rapid surges followed by sharp declines" may be attributed to Vo1d leasing its botnet infrastructure in specific regions to other groups,” they said. So, during days when Vo1d had significantly fewer bots, they probably just “gave” the devices to someone else to use.

Android TV devices infected with malware will behave unusually. They will be sluggish, they will randomly display ads, or frequently crash seemingly without cause. To clean up the device, users should check their installed apps and remove anything unfamiliar or suspicious; scan with Google Play Protect, monitor their network’s activity and ultimately, if needed, perform a factory reset.

Via BleepingComputer

You might also like

• Security flaw in popular stalkerware apps is exposing phone data of millions
• We've rounded up the best password managers
• Take a look at our guide to the best authenticator app

• A privacy notice is appearing on some Samsung TVs every few hours
• It doesn't seem to be possible to fully disable from the TV
• But it might be possible if you log into Samsung's website

Do you have a Samsung TV whose pop-up won't stop? You might not be the only one. Samsung owners have been emailing us and posting on forums about one particularly persistent pop-up message that keeps on popping up – but the good news is that there may be a simple solution.

The message is about Samsung's privacy policy collecting user data, and many Samsung owners will have seen it; I'm one of them – though mine never came back after the first showing. But for some users the pop-up keeps returning; one of the affected owners told us that it pops up every few hours, which tracks with what people are saying online. But no one seems to be able to find an option on the TVs to disable it.

The particular customer who contacted us also contacted Samsung support, and was told by a rep that there wasn't a fix. But other Samsung owners beg to differ.

How to fix the Samsung privacy pop-up problem

The issue doesn't appear to be affecting huge numbers of people, which is perhaps why the Samsung support rep didn't know how to get rid of it. But it's clearly affecting some people across multiple models – we're aware of the issue happening on the Samsung S90C and QN90B – in certain countries, including the US, and they've been posting to sites such as Reddit and to the Samsung community website. And it's that latter site where we might have found the solution.

One pop-up-plagued poster said they got rid of the message by logging into their Samsung account on the company's website, and clicked on the privacy updates. "That stopped the notice from appearing on my TV," they say.

If that's not working for you, there's something else to try in the short term: this post on the Samsung community site talks about how annoying the message is: "For the past 2 days, I've been getting an EXTREMELY annoying popup notification with small ding sound. This pops up about every 2 hours and disappears in 5 seconds."

Their fix wasn't ideal, but it did seem to work: by blocking internet access from the TV so that it has local network access but not wide area network access, the pop-up stops appearing. But that's only convenient if you use a streaming device such as a Chromecast or Apple TV 4K to deliver things to your TV; if you're using the built-in smart apps, then of course they need to be online.

Based the fact that this appeared on my TV and disappeared forever, it looks very much like a software update gremlin has got this message stuck on a loop rather than finding its way into the ether once shown.

We contacted Samsung, which said it was looking into the problem on our reader's behalf, but wasn't able to provide more information by the time of publication. We'll update this story with official info as we get it.

You might also like

• The best Samsung TVs for all budgets
• Prefer a streaming box? All our streaming device stories are here
• Everything we know about Sonos's incoming TV streamer

With a pregnancy study and a smarter cycle-tracking approach, Oura is delivering more accurate health insights.

Want a portable cinema? How about something for a backyard movie night? The best portable projectors give you a huge screen for TV shows, movies and games just about anywhere.

This $330 smartwatch can outlast its competitors in battery life and fitness features, but may fall short on software updates.

• The wait for AMD's Radeon RX 9070 series is over, as it fully reveals the Radeon RX 9070 XT and RX 9070
• The Radeon RX 9070 XT will start at $599 and the RX 9070 will launch at $549
• Both GPUs will be available on March 6, with the RX 9060 series unveiling set for a later date

Finally, after months of waiting for AMD's full reveal of the Radeon RX 9000 series GPU lineup, we've now got a proper look at the next-gen cards via Team Red's performance showcase and pricing for the RX 9070 XT and RX 9070 - spoiler alert: fans may be in for an affordable treat.

AMD announced that the Radeon RX 9070 XT and RX 9070 will launch on March 6, starting at $599 / around £570 / around AU$960 and $549 / around £520 / around AU$880, respectively. Both GPUs will come with 16GB of VRAM using PCIe 5.0 connectivity (the same as Nvidia's Blackwell GPUs). There will be no reference cards direct from AMD, only partner cards that will utilize Team Red's new RDNA 4 architecture, which was built to provide much better ray tracing and raw performance in games.

As previously suggested, AMD claims the Radeon RX 9070 XT variant will be 42% faster than the RX 7900 GRE at native 4K graphics settings - it supposedly matches Nvidia's RTX 5070 Ti across several titles, but more importantly, AMD's new GPU is claimed to outperform Blackwell card in titles like Call of Duty Black Ops 6, which shows it's 24% faster at native 4K. Ray tracing is also a highlight in the graph (pictured below), with a suggested lead over the RTX 5070 Ti by 8%.

Of course, these stats should be taken with a pinch of salt since they come directly from AMD and don't factor in features like Nvidia's new Multi-Frame Generation, but it's still impressive stuff. The 9070 XT will most likely sit around 5-10% behind the 5070 Ti in real-world testing, but for 20% less cash at retail price.

We will await real-world benchmarks and stats once the RX 9070 series launches on March 6. For now, AMD's performance claims set its XT variant in pole position to become the budget GPU king at an affordable $599 price compared to the RTX 5070 Ti's $749 / £729 / AU$1,509 (if you can even find it at MSRP amid current stock shortages, that is).

That looks like budget GPU king right there... (Image credit: AMD) Now, I can gladly eat my words... Nvidia, you've got trouble

MicroCenter's listed prices before AMD's full reveal suggested the Radeon RX 9070 XT would launch at $699 - which I feared would ring true - but fortunately, it appears as though those were indeed placeholders (unless online retailers decide to inflate prices, that is).

I'm glad to say that Team Red has left me delighted with this reveal, as I've previously stated the GPU market desperately needs competition right now - Nvidia may still lead the high-end thanks to the powerhouse RTX 5090 and RTX 5080 GPUs, but what good is a premium GPU if you can't even find one for sale?

If the Radeon RX 9070 XT at $599 ends up outperforming - or even just coming close to matching - Team Green's RTX 5070 Ti as AMD claims, and manages to topple the previous gen's RX 7900 XTX, then in my eyes it's this generation's GPU king. RDNA 4 promises significant enhancements across the board, notably with ray tracing and FSR 4 (exclusive to RDNA 4 GPUs) - based on what Team Red presented, FSR 4's performance mode reminded me of what I saw over at Nvidia with its new transformer model in DLSS 4.

Barring any potential stock issues, the Radeon RX 9070 series (specifically the XT variant) is shaping up to be the go-to GPU this generation, especially considering Team Green's RTX 5000 series launch woes.

You may also like...

• Nvidia retiring PhysX for its RTX 5000 GPUs has made some gamers furious - but I don't think it's a complete dealbreaker
• New Nvidia drivers should fix a major RTX 50 series GPU issue
• Some lucky RTX 5080 buyers may not be so lucky after all, as Nvidia confirms ‘rare’ chip fault also applies to this GPU as well as the RTX 5090 and 5070 Ti

• Apple and the NHL are collaborating to give its officials Apple Watches
• They'll be used in an outdoor game for the first time on March 1
• A custom app can give officials countdowns and even tell them which players are in the penalty box

Apple and the National Hockey League (NHL) have announced the next step in a partnership that has seen officials across the league equipped with the best Apple Watches in order to assist them in officiating games.

Announced by the NHL, officials will wear Apple Watches and use a custom NHL Watch Comms App during an outdoor game for the first time, specifically the match between the Columbus Blue Jackets and the Detroit Red Wings at Ohio Stadium, March 1.

The NHL says that officials across the league have been using the technology on a widespread scale, with nearly all officials wearing a watch this season.

In fact, they've been using and refining it for a few seasons. The Apple Watches come with a custom app developed by Presidio, offering real-time in-game notifications.

Apple Watch on ice

(Image credit: Future)

The NHL says that communication between officials is difficult during games because of crowd noise and the electric pace of games. An outdoor environment makes that even more challenging, the biggest one being the lack of a scoreboard hanging above center ice, which limits how officials can keep track of game timing and situations.

Officials wearing the watch get haptic notifications during the game, including game clock and penalty clock countdowns, and even the jersey numbers of which players are in the penalty box at any given time.

NHL executive vice president of business development and innovation David Lehanski says the league has a 90 percent participation rate among officials using the tech. The NHL has been collaborating with Apple for a number of years, notably using iPad and Mac to deliver real-time videos to coaches and players in game for the past eight years.

The NHL has future plans to expand the tech, including adding support for direct communications between the four officials on the ice, other arena officials, and officials at the league HQ in Toronto. Apple Watch may even facilitate communication between officials and coaches in the future.

You may also like

• Apple’s celebrating the 2025 MLS season kicking off with a custom logo for each and every team
• iOS 18 is bringing live sports scores to your iPhone lock screen
• "If you’re a sports fan, I think you're going to be very happy with it," Apple's Eddy Cue explains how the tech giant made its new Sports app

• Sekoia spots hackers abusing a known flaw in Cisco devices
• This leads to the discovery of a botnet called PolarEdge
• Most victims are found in the US, but the botnet is "most prevalent" in Asia and South America

A previously-undocumented botnet has been expanding around the world for more than a year, targeting a range of Cisco, ASUS, QNAP, and Synology devices, experts have warned.

Cybersecurity researchers Sekoia observed the attacks on their honeypot, and used the information to detail the campaign, its infrastructure, and targets.

In its report, Sekoia said that as of late 2023, it spotted an unnamed threat actor targeting devices vulnerable to CVE-2023-20118 - an improper user input validation bug affecting different Cisco Small Business Routers. The flaw allowed them to execute arbitrary commands on the affected devices, pulling a malicious payload from a Huawei Cloud server located in Singapore. Digging deeper, Sekoia found traces of the campaign targeting devices from other manufacturers, as well. They named the botnet PolarEdge, and confirmed that at least 2,000 endpoints around the world were infected.

Endgame unknown

The botnet’s goal is unknown at this time, the researchers said.

“The purpose of this botnet has not yet been determined. Cross-checking the IP addresses with our telemetry has not revealed any specific activity,” the report reads.

Usually, cybercriminals would develop a network of infected devices to either run Distributed Denial of Service (DDoS) attacks, set up a residential proxy, run spam and phishing campaigns, spread malware, or engage in click fraud.

The majority of the victims are found in the US, but Sekoia says the botnet appears to be “particularly prevalent” in Asia and South America, although it cannot be certain if this was a deliberate move by the attackers, or just coincidence.

Despite infecting a relatively small amount of devices, Sekoia still deemed PolarEdge a dangerous threat.

“The botnet exploits multiple vulnerabilities across different types of equipment, highlighting its ability to target various systems,” the report concludes.

“The complexity of the payloads further underscores the sophistication of the operation, suggesting that it is being conducted by skilled operators. This indicates that PolarEdge is a well-coordinated and substantial cyber threat.”

You might also like

• Massive botnet is targeting Microsoft 365 accounts across the world
• We've rounded up the best password managers
• Take a look at our guide to the best authenticator app

Oh boy, inflammatory headline up there, you bet - but it's kinda true. Let me preface this with the fact that I am a professional system builder, at least in the world of tech journalism. At last count, I had completed something like 90 published build logs. Everywhere from online, for Techradar, Tom's Hardware, PC Gamer, and in print at Maximum PC magazine. I've done a lot with it. It's my bread and butter, encouraging good folk to forsake the shackles of corporate gaming PC system integrators and instead take the reigns in the pursuit of building their own bespoke rig perfect for them. It's a big part of what I do. It's literally the entire reason I have a career to begin with, and I'm very aware of that.

I still love system building, I do - and I'd encourage everyone to keep building their own rigs. That said, I'd be a liar if I didn't recognize the significant advantages that come from just buying one of the best gaming PCs you can get off the shelf, from the likes of iBuyPower, Origin, Corsair, CyberPower, or whoever your favorite system builder is.

The Corsair One is a phenomenal pre-built system - and far more compact than anything the average builder could assemble themselves.

The honest truth is that, unless you're an IT god who's spent years pruning over every little detail of system building with intimate knowledge of all of the latest and greatest hardware and its caveats, you're just not going to be able to compete with a consummate professional who's day job it is is to build gaming PCs for these companies. It's like comparing your dad's "famous" chili con carne recipe (because, let's face it, everyone's dad has a chili recipe that takes a full 12 hours of stewing in a crockpot for some unknown reason) to Gordon Ramsay's. Like yeah, it's good, but it's not going to blow up your palette and scream profanities at you while it does so.

The Cost

Enough fluff; why am I so enamored with system builders? Is it because I worked at Corsair years ago and secretly hate the job I've created for myself? No. It's not. That's sarcasm, by the way.

Nope, the biggest reason, by far, is cost and access. During the COVID-19 pandemic, I watched as the entire tech industry imploded on itself as the working-from-home boom erupted to the forefront of our little industry. Good people scrambled for hardware and components. Combine that with the global chip shortage and supply chain issues, plus the crypto craze and unscrupulous scalpers sucking up every graphics card they could get their grubby little mitts on, and the outcome was obvious. Building a custom rig was damn near impossible, at a reasonable price anyway. Even getting access to the hardware via traditional avenues was an insurmountable task.

At the time, the only people who could get a healthy supply of Nvidia and AMD graphics cards were the system builders and integrators. They had a direct line of access to Nvidia, AMD, Intel, and their AIB partners. That's a good enough reason to start with, particularly with the ongoing RTX 50 series shortages we're currently seeing, mimicking that same drama from 2020.

Finding a new GPU during the chip shortage was a nightmare - and it's still not easy. (Image credit: Aleksandr Grechanyuk / Shutterstock)

Then there's the cost. The thing is, most system builders aren't building three or four gaming PCs a week; they're building several hundred. That requires ungodly amounts of graphics cards, processors, RAM, motherboards, you name it. What that means is they buy them in by the pallet. We're talking 1,000 GPUs at a time, or more. Bought at a reduced price, sometimes several hundred bucks off what you'll see at Scan, Amazon, or Newegg, and of course usually tax-free. It's a similar reason as to why consoles are still so cost-effective by comparison (ever tried building a similarly priced PC to a PS5 Pro? It's nigh-impossible). So straight out of the gate, doing it yourself, you have to compete with a company that has access to combined savings across pretty much every component you can imagine, and can then write that tax off too.

Build costs are fairly minimal as well. Once you're experienced enough and have a template, any reputable PC builder can finish a fairly slick rig in under 60 minutes, ready for an OS install and more, so there's little cost on the labor front too - per unit, anyway.

Quality Control

And then there's the overall quality of the rig. Not only can you be sure that your pre-bought system is going to be built to spec and to a high standard (as long as you're buying from a trustworthy company that isn't cutting corners), but all of the compatibility issues, software problems, and things of that ilk will be sorted for you before it even lands on your doorstep.

You don't have to worry about BIOS updates, or accidentally buying the wrong PCIe SSD for your motherboard, or getting too small a power supply, or cooler installation, or memory frequency motherboard incompatibility, or anything along those lines.

Cable management, cooling, overclocking—all of it is going to be finely dialed from the jump. It'll just work straight out of the gate. Yes, you might have to go in and uninstall whatever aftermarket antivirus they've chucked on there once it arrives, but aside from that, it's almost immediately plug and play, and when time is a luxury, that's a big advantage.

CyberPower's pre-built PCs are good quality and promise a nice, tidy interior - no need for expert cable management skills on your part. (Image credit: CyberPower PC) Just Buy It?

So, chuck everything away then, close that PCPartPicker tab, and just buy a prebuilt, then? Well, yes and no. It's like your car, right? You're always going to get the folk who go around and build their own, tweak the ones they've got, and get ridiculous horsepower out of an engine in ways that shouldn't be possible. Usually for some ridiculous investment of time or sum of money along the way.

Or, alternatively, you could save the drama and just buy a 400-horsepower EV, or whatever your engine of choice is, straight off the shop floor, with few issues if any, and do 100 MPH in less than 5 seconds. More than enough for the average Joe. And you don't have to deal with any of the building, or the worrying, or the waiting.

But - and here's the caveat that keeps me in custom PC territory - you'll be driving down the road with your brand new car, ripping up the tarmac, and within ten minutes, you'll see the exact same model drive by, and they'll flash their lights at you, give you a wave, and you'll be like, "Damn, I could've had something special instead".

I love building PCs; I do. I love tinkering, choosing the parts, the theme, the colors, and even the RGB, but if I didn't do what I do as a day job and I needed something that just worked, without a doubt I'd just grab one off the shelf instead.

You might also like...

• The best all-in-one computer for 2025: top home and office AIO PCs
• My favorite laptop maker just unveiled its first desktop - and it's the cutest little PC I've ever seen
• Best mini PC of 2025: small desktop computers with big specs for every budget

Whether you’re on a road trip or a plane ride, these are our favorite travel strollers that fold as flat as possible.

Apple Intelligence is the Cupertino company’s take on artificial intelligence. It’s a suite of tools deeply integrated into compatible iPhone, iPad and Mac devices.

The idea behind Apple Intelligence is to make life easier for Apple users, using AI to add value through suggestions, automations and personalization. That includes everything from Writing Tools to message summaries.

New AI-powered features are being released in stages. With more set to come in 2025, here’s what you need to know about Apple Intelligence.

This article was correct as of February 2025. AI tools are updated regularly and it is possible that some features have changed since this article was written. Some features may also only be available in certain countries.

Apple Intelligence works across a range of Apple devices. (Image credit: Apple) What is Apple Intelligence?

Announced in 2024, Apple Intelligence is designed to enhance the everyday user experience with iPhone, iPad and Mac devices. It’s a suite of AI-powered features integrated into Apple’s core apps and services, including generative tools for editing writing, summarizing conversations and organizing photos. It also promises a context-aware Siri. Rollout has been piecemeal, with Apple drip-feeding features through a series of software updates.

Privacy is a key component of Apple Intelligence. The toolkit’s contextual awareness depends on unprecedented access to your data, which is why Apple is ensuring that as little as possible actually leaves your device. It runs all the tasks it can on device. When they’re not, Apple uses Private Cloud Compute, a security protocol designed to set a new privacy standard for AI tools.

What can you use Apple Intelligence for?

Apple Intelligence enhances the user experience in several ways. Here are some of the features which have already rolled out to Apple devices:

• Writing Tools is an editing assistant which can proofread text, summarize passages and rewrite your words using a different tone of voice.
• Summaries work across everything from group chats and emails to web pages and notifications, condensing information into digestible overviews.
• Siri gets smarter, allowing you type commands and ask follow-up queries, making it closer to an AI chatbot. It also gains ChatGPT support.
• Photos are easier to search for with natural language prompts; Clean Up works like Google’s Magic Eraser; and you can create Memories from prompts.
• Mail gains support for smart replies based on the contents of emails. Apple Intelligence also automatically categorizes, prioritizes and summarizes emails.
• Transcription is enhanced by Apple Intelligence, allowing you to record phone calls or voice notes and have them automatically transcribed.
• Visual Intelligence works with the Camera Control button on the iPhone 16, allowing you to get search results based on what you point the camera at.
• Genmoji is a generative emoji tool, which allows you to type in a description of the emoji that you want to send to generate it.
• Image Playground is a place to generate AI images, which can be based on your real friends and family.

We’ve put together a full run-down of all the Apple Intelligence intelligence features here, including when you can use them.

Image Playground on iPhone. (Image credit: Future/Jacob Krol) What can’t you use Apple Intelligence for?

As above, several Apple Intelligence features haven’t arrived yet. Chief among them is a more intelligent Siri. This is expected in the near future, introducing awareness of what you’re looking at on screen, as well as deep personalization based on information from your device.

Because it prioritizes privacy, there are also ways in which Apple Intelligence can’t offer the same real-time, cloud-based performance as other AI tools. It’s not a fully fledged chatbot like ChatGPT or Google Gemini, for example, and it can’t do photorealistic image generation. In fact, if you ask Siri a question that's too complex it will hand you over to ChatGPT for an answer.

Unlike Google Gemini, you can’t use Apple Intelligence on non-Apple devices. You need to own one of the following compatible Apple devices to use at least some of the AI features: iPhone 15 Pro, iPhone 15 Pro Max, iPhone 16, iPhone 16 Plus, iPhone 16 Pro, iPhone 16 Pro Max, iPad with M-series chip, Mac with M-series chip.

How much does Apple Intelligence cost?

At present, Apple Intelligence is completely free for users of compatible Apple devices. The AI features are being rolled out in a series of iOS 18 software updates at no additional cost.

This could change in future. Some analysts have claimed that Apple may start charging for some Apple Intelligence features. Luckily, other industry insiders have suggested that paid AI tools are several years away, with the first wave unlikely to be introduced before 2027.

Where can you use Apple Intelligence?

Apple Intelligence initially rolled out with language support for US English only. In December 2024, this expanded to include localized English for the UK, Australia, New Zealand and Canada. Support for more languages is coming in April 2025. For now, your device needs to be set to one of the supported languages to use Apple Intelligence.

Access is limited for users in the European Union and China. Apple Intelligence is available with macOS Sequoia 15.1 in these regions, but users will need to wait until April to use the AI features on iPhone or iPad.

If you are able to access Apple Intelligence, you’ll find its features deeply integrated into your device at a system level. Siri might be the voice of Apple Intelligence, but there’s no single ‘hub’ to access the full suite of apps. Instead, you’ll find that the tools become available in different apps when relevant to the context, whether that’s in Photos, Mail or Notifications. The toolkit is cross-platform, designed to work seamlessly across apps on iPhone, iPad and macOS.

Writings Tools from Apple Intelligence (Image credit: Future / Apple) Is Apple Intelligence any good?

Based on our experience, Apple Intelligence meaningfully improves the user experience across iPhone, iPad and Mac. Several members of our team have been using the toolkit on their devices, with positive feedback. Writing Tools work like a supercharged autocorrect, significantly improving the Notes app experience. We also rate the usefulness of Reduce Interruptions, which only allows key alerts through. Other Apple Intelligence features we keep coming back to include article summaries in Safari and the Clean Up tool in Photos.

While typing to Siri is also a useful addition, we can’t help but feel that the best is yet to come, including situational awareness. Although on-device processing enhances privacy, it could also be one of the reasons behind the slow rollout of Apple Intelligence features. The result is that users can get more advanced AI support from other chatbot apps, such as Gemini, or by accessing ChatGPT through Siri.

Use Apple Intelligence if...

You want native AI on your Apple device

Designed for Apple devices and integrated deeply with their operating systems, Apple Intelligence builds AI assistance right into core apps and services, offering contextual assistance within Mail, Photos and more.

You want secure, on-device AI

Built into OpenAI’s chatbot, ChatGPT search allows you to find detailed answers from the web through conversational dialogue. That includes the ability to submit follow-up queries within the same thread, written in natural language.

Don't use Apple Intelligence if...

You don’t have an Apple device

Unlike Google’s Gemini chatbot, Apple Intelligence can only be used on compatible iPhone, iPad and Mac devices, which means you’ll need to find a different solution if you’re using an Android device or Windows PC.

You need an full AI chatbot

Updates mean you can type and respond to Siri, but Apple’s smart assistant is a long way off the cloud-powered capabilities of AI chatbots such as ChatGPT and Gemini. Advancements are coming soon, though.

Also consider

• Gemini is Google’s AI multimodal chatbot. Available as an app for iOS users, it’s also increasingly integrated into the Android operating system. Hundreds of devices are supported, as are more than 40 languages. Features include spoken search and image generation.
• Galaxy AI is a suite of AI-powered features baked in to Samsung Galaxy mobile devices. Like Apple Intelligence, it’s designed to streamline the user experience through automation and personalization. Tools include real-time translation, automatic transcription and note organisation.

You might also like...

• What is Apple Intelligence? The new Apple AI for your iPhone, iPad and Mac explained
• How to use Apple Intelligence: Get started with Apple AI
• Apple Intelligence features explained - everything you need to know about Apple AI and when you can use it

March's biggest titles include The Electric State, Revelations and more.

• EV2 will be practical and accessible, opening up EVs to a wider audience
• Set to cost €30,000 (around $31,400/AU$49,400) it will be Kia’s best-value EV
• Kia’s PV5 electric van will arrive later this year

Although not quite as bombastic as its inaugural EV Day in 2023, where Kia mapped out its plans to accelerate its EV offering with the reveal of EV2 through to EV5, its most recent showcase confirmed its commitment to electrification… and revealed more about its progress.

On stage was the production-ready Kia EV4, both in saloon (or sedan) and hatchback guise, as well as an EV2 show car that gives us our closest look yet at what the smallest and most affordable Kia EV will look like.

Boasting chunky crossover exterior styling and an equally funky interior, it measures around the same size as the upcoming Renault 4 E Tech and the current Mini Aceman, with sales expected to begin in South Korea and Europe next year.

Whether the EV2 makes its way to the US is still to be confirmed, although the compact B-segment crossover isn’t particularly popular Stateside and Kia revealed its most affordable EV will be manufactured in Slovakia, making it very unlikely.

Speaking on the interior, Kia’s head of design Karim Habib explained that his team played on a “picnic in the city” theme, whereby owners can slide either the front seats all the way rearwards and stretch out, or slide them forwards and fold up the rears seats to take advantage of on the vehicle’s completely flat floor.

Image 1 of 6

(Image credit: Kia)Image 2 of 6

(Image credit: Kia)Image 3 of 6

(Image credit: Kia)Image 4 of 6

(Image credit: Kia)Image 5 of 6

(Image credit: Kia)Image 6 of 6

(Image credit: Kia)

Those rear seats also slide rearwards, almost into the boot or trunk, while the seat cushions flip upwards to create a cavernous chill-out zone inside.

That said, Habib revealed that while he was fighting for the flip up seats in the rear to feature in the production vehicle, having the front two seats on such a lengthy sliding rail would likely not be seen on the finished product.

Similarly, the neat ‘suicide doors’, which are hinged on the A and C-pillars to create a massive aperture when opened, also won’t make it to production, with Habib stating they are “too heavy and too expensive” at the moment, but that he hoped Kia would introduce them to a future model soon.

What's more, the concept showcased a projector system that can beam messages onto a dedicated area on the windows.

Built on the same E-GMP platform as the remainder of Kia’s line-up, the EV2 will feature the same 400V electrical architecture as the slightly larger EV3, which means a 10-80% charging time of around 30 minutes.

However, it will be among the first Kia products to introduce a new Lithium-ion Phosphate (LFP) battery pack when it arrives next year.

Full details on the exact battery range is due to be revealed at a later date, but it will likely sport the same 43.3kW and 51.5kW variants as those confirmed for Kia’s PBV van line-up.

Kia bets big on that #vanlife

(Image credit: Kia)

While we still await on news of a potential EV1, tipped to be the smallest and cheapest of Kia’s full electric vehicle line-up, as well as an EV7 and EV8 to plug the remaining gaps, the Korean marque has confirmed it is heading full speed into the world of electrified Car Derived Vans – or CDVs to coin one of marque’s many acronyms.

Predominantly aimed at the European, Canadian and Korean domestic commercial vehicle markets, where Cargo, Crew and Chassis Cab options will be available later this year, the platform has been designed to be as modular as possible.

This means it can be optioned as anything from wheelchair accessible vehicle to a freezer box van.

Better still, the PV5 is also designed for private customers wanting masses of load-lugging space, with a Passenger variant featuring three-row seating for up to seven, as well as the ability to easily drop seats for bulky lifestyle cargo.

Kia confirmed that a “Light Camper” variant of this model will also be available at launch, with a more serious camper van conversion in the pipeline. Expect that to come with folding beds, kitchens and other amenities.

Not ones to go lightly when entering a new market, Kia has also created a range of 'Add Gear' accessories, which include cup holders, coat hooks, trash bins and cool boxes, all of which either clip into place or slide onto dedicated rails in the cabin.

Image 1 of 5

(Image credit: Kia)Image 2 of 5

(Image credit: Kia)Image 3 of 5

(Image credit: Kia)Image 4 of 5

(Image credit: Kia)Image 5 of 5

(Image credit: Kia)

This allows potential buyers to instantly customize the interior of their PV5 to suit their needs.

Interestingly, the infotainment system found in Kia’s passenger cars has been ditched for an Android Automotive-based system that the company says helps open up the possibility of hosting third-party applications, such as Samsung’s IoT Pro, which will help small business and fleet managers work more efficiently on the move.

It is still unclear how this will work in the Passenger versions, but it could open up the possibility of more Android Automotive integration into future Kia cars, vastly increasing the number of available entertainment apps that are available for free download.

With an electric range of up to 250 miles, 30-minute fast charging and a hugely competitive starting price of €35,000 (around $36,600 / AU$58,00), Kia hopes it can make an impact in the commercial vehicle sector.

You might also like

• Kia's promising EV4 electric saloon breaks cover – and it stays true to its concept car roots
• Kia shows off new off-road EV concepts and they look like the perfect apocalypse escape rides
• I’ve driven the Kia EV3 and it's one of the best electric vehicles for most people right now

Donald Trump has the distinction of being the first U.S. President elected in the AI era. While he has rightly recognized the importance of AI with his decisions to appoint a White House AI & Crypto Czar and jumpstart infrastructure investment through the Stargate venture, his administration faces novel threats to U.S. AI dominance. As Trump settles into office, we’re seeing accelerating competition from new AI models like DeepSeek and increasing security threats as businesses and critical infrastructure systems integrate AI.

Despite security dangers, it is no exaggeration to say that AI has the potential to be humanity’s greatest creation, offering the means to go beyond the bounds of human capability and unlock discoveries that have been out of reach. Whether the U.S. will lead the world in these advancements will be determined within the next four years.

As the leader of the world’s largest economy, Trump will have the ability to influence and guide all of AI’s potential into a practical reality. He must focus on two forms of AI security to realize this, addressing both the security threats AI poses and the need to secure America’s dominance in the AI domain.

Securing AI-driven systems

Sustainable AI innovation cannot occur without balanced protections for its users. Removing barriers to entry to AI is achievable once the burden of safe usage is on the companies creating and implementing AI, not on its everyday users. Organizations should be allowed to freely distribute AI applications and systems, but only when they also make the associated testing and security reports freely available.

Consider the parallels with access to automobiles: they contain very advanced technology but everyone can see their safety rating and crash-test results before they get behind the wheel. Accessible safety ratings will similarly speed the safe democratization of AI.

In practical terms, smartphone operating systems are the real edge devices for AI, where most people will access the technology on an hour-to-hour, minute-to-minute basis. Putting the correct legal safeguards in place for how the owners of these operating systems implement AI will have the most dramatic and immediate impact on AI safety and security.

Meanwhile, the rise of AI agents represents a new vulnerability for U.S. businesses and national infrastructure, as agents can gain access to sensitive information and run mission-critical tasks such as building and resource control systems. It is all but certain that, in tandem with their push to innovate their own AI models, countries like China are making significant efforts to hack those of their adversaries. Indeed, there is already evidence of China-based threat actors attempting to exfiltrate information from OpenAI.

With the advent of advanced autonomous agents, the cutting edge will be agentic warfare, in which these agents will be tasked with finding ways to hack corporate or national defenses. To win on AI, our security must be as good – or better – than our enemies’ hacking innovations.

Building AI infrastructure

Securing physical infrastructure and the resources required for innovation is also essential to copper-fasten America’s position as the global leader in AI. We can have the smartest companies in the world but we can’t win the AI arms race if we don’t have the computing and physical infrastructure to stay ahead. DeepSeek has upped the stakes for model cost and performance but the bigger picture is about infrastructure and implementation.

That means reinvigorating our chip manufacturers, investing in high-performance computing, and making long-term investments in energy infrastructure. Smart trade deals with new and existing partners will ensure America’s access to the GPUs and technologies that underpin AI advances.

Running AI systems requires huge amounts of power: the IEA estimates the electricity use of data centers, AI, and crypto could more than double to over 1,000 terawatt-hours in 2026 – roughly equivalent to the electricity consumption of Japan. U.S. energy supply and grid infrastructure must be up to the task. Expanding wind, solar, and tidal energy can both power the AI revolution and create new jobs – a win-win for America.

Securing U.S. AI dominance

We must also plan for long-term AI innovation. America once led the world in technological research, spearheaded by our 17 National Laboratories that sprung from federal investment during World War II. The time is right to revitalize those labs, giving them access to abundant AI technology and funding to staff up and explore this new frontier. The new partnership between the National Labs and OpenAI is a step in the right direction.

The Trump administration must empower our labs, designating them as places where things can be created, not just homes for long-form research. Instead of spurring job losses, AI can make jobs safer, more skilled, and higher-value – all good things for our economy. Finding new opportunities for job-specific efficiency is one area where our National Labs can act as hothouses for AI innovation.

Of course, AI can be a powerful tool for government efficiency, with its ability to pull vast amounts of data from various sources and identify patterns and opportunities for improvements. All this data must be handled delicately, with an extreme emphasis on safety and the security of personally identifiable information. Consider the Department of Veterans Affairs, which has sensitive personal and health information on our 18 million veterans and their families. A data breach there would upset and alienate people and set back AI progress.

Establishing an AI task force

It is clear that the scale of the AI opportunity requires an unprecedented approach that balances appropriate safeguards with speed of action. This can be achieved by establishing a Department of AI Acceleration, which will allow the U.S. to rip through red tape, apply funding appropriately, and propel American-built AI. Sticking with President Trump’s knack for naming things, perhaps ‘AI Force’ is appropriate?

No matter its title, a dedicated agency that allocates funding to those developing AI in responsible ways will achieve better results than time-consuming regulation and legislation. If we want the US to accelerate and dominate in the AI space, let’s assemble the greatest minds in a single place, with the sole purpose of ensuring safe AI that benefits everyone.

Quickly establishing this landmark new department, backed by targeted investment in research and infrastructure and an agreed approach to AI security, will make America the global authority on safe AI – not just leading the way but lighting a path forward for the rest of the world. The opportunity is ours for the taking.

We've featured the best privacy tool and anonymous browser.

This article was produced as part of TechRadarPro's Expert Insights channel where we feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/news/submit-your-story-to-techradar-pro

• Researchers found tens of thousands of vulnerable AMS around the world
• 49,000 misconfigured AMS could represent a major problem
• Vendors are working on a fix

Tens of thousands of Access Management Systems (AMS), built by different vendors and spread across different countries, were found connected to the wider internet, misconfigured, and thus - exposed to cyberattacks.

A report from cybersecurity researchers Modat noted Access Management Systems are security frameworks that control and monitor who can access digital or physical resources within an organization. They authenticate users through methods like passwords, biometrics, or multi-factor authentication and authorize their level of access based on predefined policies.

Modat said they found 49,000 misconfigured AMS’, in different organizations around the globe. “Widespread internet exposure of AMS across multiple countries indicates a worldwide problem,” it said. The devices were found in key industries such as construction, healthcare, education, manufacturing, the oil industry, and government organizations.

Botnet for hire

Arguably the biggest problem with these misconfigurations is the compromised physical security of the affected organizations, as criminals could bypass physical security and access buildings which should otherwise be off limits.

But aside from that, another important takeaway is that cybercriminals could steal sensitive employee data this way. “Personal identification information, employee photographs, biometric data, work schedules, payslips, and complete facility control and access were all found,” Modat stressed.

This could open the floodgates to phishing, identity theft, social engineering, and other forms of fraud that could see sensitive government information exfiltrated from the servers.

Different AMS’ were affected differently, the researchers further explained. They said they detected a “high concentration” of vulnerabilities, mostly in European countries, the US, and the MENA region (Middle East and North Africa).

The majority of flawed devices were found in Italy (16,678), Mexico (5,940), and Vietnam (5,035).

Modat notified all of the affected organizations, but according to BleepingComputer, none responded so we don’t know how many mitigated the risk by now. The researchers also reached out to vendors, some of which confirmed to be working on a fix.

Via BleepingComputer

You might also like

• What Is IAM? Understanding Identity & Access Management
• We've rounded up the best password managers
• Take a look at our guide to the best authenticator app

• Microsoft argues against current restrictions on semiconductors
• Tech giant wants Trump to loosen export constraints
• The US and China are in the middle of an AI race

Microsoft has called on President Trump to ease export restrictions on crucial computer chips needed for AI technology.

The previous Biden administration implemented a cap on the export of American AI components to many foreign markets, amid concerns for national security, and while these restrictions are as yet unchanged by the Trump administration, Microsoft has warned the current President that this could give China a ‘strategic advantage’ in spreading its own AI technology, as emerging markets will be forced to turn away from the states for advanced technologies.

The limits, Microsoft argues, undermines two key priorities for US foreign policy; strengthening the US’s leadership in AI technologies, and reducing the country’s trade deficit.

Semiconductor stipulations

It says allies of the US were also affected by the restrictions, like Taiwan, South Korea, India, and Switzerland. The export restrictions “go beyond what’s needed”, and puts these allies in a “Tier Two category and imposes quantitative limits on the ability of American tech companies to build and expand AI datacenters in their countries”.

“The unintended consequence of this approach is to encourage Tier Two countries to look elsewhere for AI infrastructure and services,” Microsoft noted in its statement.

“And it’s obvious where they will be forced to turn. If left unchanged, the Diffusion Rule will become a gift to China’s rapidly expanding AI sector.”

The US and China are engaged in a ‘chip war’, with China declaring US chips unsafe, urging companies to use domestic semiconductors, with both nations racing to develop and build high-powered chips domestically.

Earlier in 2025, Microsoft revealed plans to spend $80 billion on AI data centers, with a large component of this plan focusing on export power.

In 2024, Microsoft also pledged over $35 billion in investments across 14 countries to build AI and data center infrastructure, and has reiterated that the US needs to cooperate with “allies and friends” to maintain its position at the top.

You might also like

• Take a look at our picks for the best cloud storage around
• Check out our recommendations for the best AI tools
• Microsoft is spending $700 million to ramp up security and computing power in Poland

Wicked, The Brutalist, Conclave and Anora are among the films with multiple nominations.

Make sure you're getting enough omega-3s in your diet to keep your body in tip-top shape.

Happy Oscars weekend, everyone! The world's glitziest event of the year is upon us and, no matter where you live, you'll be able to watch it on one of the world's best streaming services or one of your nation's local networks.

If you've had your fill of the 2025 awards season and want to watch something else, we've also got you covered on that front. Indeed, there are a number of exciting new movies and shows to stream over the next few days, so read on to see what you can enjoy ahead of the first week of March. Enjoy! – Tom Power, senior entertainment reporter

Running Point (Netflix)

It looks like Netflix is hoping to follow the success of the hit Apple TV+ sports comedy Ted Lasso with its new series Running Point. Swapping soccer for basketball, Kate Hudson takes on the role of a female Ted Lasso-style sports leader in Isla Gordon, who's unexpectedly put in charge of the Los Angeles Waves, one of the most famous professional basketball franchises in the US. In short: she must prove that she’s the right person for the job.

Going off the trailer alone, the Mindy Kaling-created series shares a similar humor and cheerfulness to Ted Lasso. We'll see if it's an ideal watch while we wait for a Ted Lasso's rumored fourth season but, based on its 69% critical score on Rotten Tomatoes, it doesn't live up its rival' critical acclaim. – Grace Morris, entertainment writer

• Watch Running Point on Netflix

The Lord of the Rings: War of the Rohirrim (Max)

An exciting and late addition to our everything new on Max in February 2025 article is The Lord of the Rings: War of the Rohirrim. Set nearly 200 years before The Lord of the Rings, one of February's final new Max movies tells the story of Helm Hammerhand (voiced by Brian Cox), the legendary King of Rohan. The War of the Rohirrim focuses on the Rohan's last stand against the Dunlendings after he unintentionally kills their leader and unleashes a full-blown war.

With its 49% Rotten Tomatoes score from the critics, The Lord of the Rings: War of the Rohirrim won't make it onto our best Max movies list. However, I think it’s still a unique and enjoyable way to bring to life the fantastical world of Middle-earth, especially in its anime format. – GM

• Watch The Lord of the Rings: War of the Rohirrim on Max

House of David (Prime Video)

House of David is a new Prime Video show that chronicles the rise of biblical figure David (Michael Iskander), aka the renowned king of Israel. The historical drama follows the once-powerful King Saul (Ali Suliman) as he falls victim to his own pride and, with God’s direction, the prophet Samuel (Stephen Lang) anoints a teenage outcast as the new king. Cue a battle for the throne.

Show creator Jon Erwin told ABC Entertainment News that House of David is similar to Game of Thrones as it explores the struggle for power, so this biblical epic certainly has the potential to be one of the best Prime Video shows. Whether Amazon's newest TV Original will, though, is up for debate. – GM

• Watch House of David on Prime Video

Berlin ER (Apple TV Plus)

It’ll take a lot to knock Severance season 2 off top spot, both in terms of Apple TV+ rankings and my own personal one, but I’m ready to give Berlin ER a go. There’s tough competition when it comes to our best Apple TV+ shows, and only time will tell if this new series will make it onto that list. Fans of popular medical dramas like Grey’s Anatomy, House, and new Max series The Pitt will likely enjoy this, but I hope it’s got enough for newcomers to the genre, too.

The eight-episode program follows a young doctor seeking a fresh start after her personal life falls apart. The treatment? Throwing herself into the emergency room in Berlin’s busiest hospital. It’s no easy task, of course, and she soon finds herself battling staff resistance and a run-down system, which is plenty to distract yourself with, in all fairness. You can’t think about your personal problems when you’re out there saving lives, right? – LB

• Watch Berlin ER on Apple TV+

Toxic Town (Netflix)

The Corby toxic waste case wasn't something I'd heard of until this Netflix show was announced, but, after reading about it, I know this dramatic retelling will make me cry. The four-part series follows three mothers who fought for justice after a growing number of babies were born with birth defects. The reported number was three times higher than those observed in children born in the surrounding regions, so it was up to ordinary citizens to get to the bottom of it. This case bears an upsetting resemblance to the Teflon scandal, as dramatized in the brilliant movie Dark Waters, which I also recommend you check out.

If Toxic Town's plot wasn't enough to hook me, the cast certainly is, with huge names including Doctor Who alumnus Jodie Whittaker, Bank of Dave star Rory Kinnear, and Sex Education's Aimee Lou Wood all involved. Will we be adding this to our best Netflix shows list? You'll have to wait and see. Based on its early glowing reviews, it seems a no-brainer though. – Lucy Buglass, senior entertainment writer

• Watch Toxic Town on Netflix

Doubters to Believers Liverpool: Klopp's Era (Prime Video)

As a huge Everton fan, I feel particularly dirty writing about our biggest (not the mention local!) soccer rivals in Liverpool FC. But, considering nobody else on the entertainment team is interested in the so-called beautiful game, I've got to suck it up. I want the record to show that I'm really not happy about this, though!

Anyway, Doubters to Believers Liverpool FC: Klopp's Era is a must-see Prime Video docuseries for anyone who religiously follows the current English Premier League leaders. The four-part show will deliver unprecedented, fly-on-the-wall access to Jurgen Klopp's final six months in charge of the Anfield-based soccer outfit, and will look back at his trophy-laden spell in charge of one of the world's most-supported sides.

Amazon has become the go-to place for the best sports docuseries around, but this is one I definitely won't be tuning in to watch. Yes, I'm just that bitter! – TP

• Doubters to Believers Liverpool: Klopp's Era on Prime Video

The 2025 Oscars (Hulu)

It’s that time of year again! We’ll be celebrating the biggest and best in the movie world this weekend, and we’ve all got strong thoughts about who should and shouldn’t win. Unsurprisingly, I'm rooting for The Substance to do well after hailing it as the best body horror of 2024. Meanwhile, my colleague Rowan Davies said he couldn’t believe that Netflix's Emilia Perez is this year's most-nominated movie or TV show at the time of the Golden Globes nominees, so the controversial movie is still frustrating cinema lovers as we go into Oscar season.

There are plenty of big releases worthy of praise, though, and I’m expecting big things from Conclave, The Brutalist, and Anora. Sometimes, these ceremonies are full of surprises, so we’ll have to wait patiently for the results. Don’t miss the entertainment industry’s biggest award ceremony when it’s streaming on Hulu (US), ITV1 (UK), and Channel 7 (Australia). – LB

• Watch the 2025 Oscars on Hulu (US), ITV1 (UK), and Channel 7 (Australia)

For more streaming coverage, read our guides on the best Netflix movies, best Disney Plus shows, best Paramount Plus movies, and best Prime Video movies.