TechRadar News

A new study by OpenAI has identified that ChatGPT-4o does give different responses based on your name in a very small number of situations.

Developing an AI isn’t a simple programming job where you can set a number of rules, effectively telling the LLM what to say. An LLM (the large language model on which a chatbot like ChatGPT is based) needs to be trained on huge amounts of data, from which it can identify patterns and start to learn.

Of course, that data comes from the real world, so it often is full of human biases including gender and racial stereotypes. The more training you can do on your LLM the more you can weed out these stereotypes and biases, and also reduce harmful outputs, but it would be very hard to remove them completely.

What's in a name?

Writing about the study (called First-Person Fairness in Chatbots), OpenAI explains, “In this study, we explored how subtle cues about a user's identity—like their name—can influence ChatGPT's responses." It’s interesting to investigate if an LLM like ChatGPT treats you differently if it perceives you as a male or female, especially since you need to tell it your name for some applications.

AI fairness is typically associated with tasks like screening resumes or credit scoring, but this piece of research was more about the everyday stuff that people use ChatGPT for, like asking for entertainment tips. The research was carried out across a large number of real-life ChatGPT transcripts and looked at how identical requests were handled by users with different names.

AI fairness

“Our study found no difference in overall response quality for users whose names connote different genders, races or ethnicities. When names occasionally do spark differences in how ChatGPT answers the same prompt, our methodology found that less than 1% of those name-based differences reflected a harmful stereotype”, said OpenAI.

Less than 1% seems hardly significant at all, but it’s not 0%. While we’re dealing with responses that could be considered harmful at less than 0.2% for ChatGPT-4o, it’s still possible to ascertain trends in this data, and it turns out that that it's in the fields of entertainment and art where the largest harmful gender stereotyping responses could be found.

(Image credit: OpenAI) Gender bias in ChatGPT

There have certainly been other research studies into ChatGPT that have concluded bias. Ghosh and Caliskan (2023) focused on AI-moderated and automated language translation. They found that ChatGPT perpetuates gender stereotypes assigned to certain occupations or actions when converting gender-neutral pronouns to ‘he’ or ‘she.’ Again, Zhou and Sanfilippo (2023) conducted an analysis of gender bias in ChatGPT and concluded that ChatGPT tends to show implicit gender bias when it comes to allocating professional titles.

It should be noted that 2023 was before the current ChatGPT-4o model was released, but it could still be worth changing the name you give ChatGPT in your next session to see if the responses feel different to you. But remember responses representing harmful stereotypes in the most recent research by OpenAI were only found to be present in a tiny 0.1% of cases using its current model, ChatGPT-4o, while biases on older LLMs were found in up to 1% of cases.

You might also like...

• 4 ways you can use ChatGPT's Canvas mode to improve your daily life
• New ChatGPT prompt goes viral with Sam Altman’s approval
• ChatGPT’s Google search rival is set to go live this year – could SearchGPT break Google's hold on the web?

New research has revealed over a third (36%) of IT leaders admit to having disabled security measures on their systems, and a staggering 70% have admitted to reusing systems passwords.

IT leaders have a lot of confidence in their organisation, as 80% say their employees wouldn't fall for a phishing attack. Having said this, 64% of leaders themselves have clicked on phishing traps.

Despite this, a study by Arctic Wolf says that they can be quick to punish others for mistakes, with 27% having witnessed the termination of an employee for falling victim to a scam.

Do as we say, not as we do

The report outlines not just a disconnect between IT leaders and their employees, but also, if you’ll excuse the dramatics, between leaders and reality. Many overestimate their organizations ability to spot phishing attacks, and are overconfident about their cybersecurity.

According to the report, despite their assuredness, 83% of leaders have observed employees clicking on phishing simulation links, and 61% of leaders have reported one or more cybersecurity breaches in the last 12 months.

“Cybersecurity isn't just about technology—it’s about people. As threat actors grow more sophisticated, security leaders must move beyond traditional security training methods and adopt a comprehensive human risk management strategy that will not only help them to better identify and mitigate threats, but more importantly foster a more proactive and security-conscious workforce.” said Adam Marre, chief information security officer, Arctic Wolf.

The threat landscape is evolving fast, and phishing attacks are becoming more sophisticated and more frequent, so overconfidence could leave companies vulnerable. Firms need robust cybersecurity now more than ever, and that requires an honest assessment of risks and vulnerabilities.

More from TechRadar Pro

• Check out our pick of the best endpoint protection software
• C-level executives are a weak point for cybersecurity
• Take a look at our best malware removal software choices

It's no secret that I love my Kindle; not just the new ultra-fast Amazon Kindle Paperwhite 12th Gen, but virtually all the Kindles that have come before it. It's a fantastic purpose-built E Ink reading device that lets me read dozens of books without a recharge. And yet it has one feature that has caused me – and more than a few others – a fair bit of frustration: the power button.

Kindles are pervasive. During the recent Kindle launch event, Amazon Devices Lead Panos Panay said the company is this year seeing the highest sales of Kindles in over a decade. That means many of you are discovering the somewhat odd placement of the Kindle power button for the first time. If you have your Kindle nearby, pick it up and examine the bottom edge. Next to the USB-C port is the shiny little power button.

There is nothing wrong with this button. It has no special powers beyond turning your Kindle on or waking it Kindle from sleep; a long press will turn off the Kindle completely, but given its power-sipping capabilities, who does that?

The problem with this placement, though, is that it seems to not take into account how some of us read with these E Ink devices. Usually, I'm reading my Kindle in bed with it propped up on some pillows, but I've also read with it on my lap or balancing on my backpack or an airplane's tray table. This means most of the weight of the ereader is resting on that little power button. It's not uncommon for me to forget, and maybe rest the weight of my hand on the Kindle while I hold it; other times I'm using it hands-free, and the Kindle is wedged between the top of a table and the back of the seat in front of me, and the person in that seat leans back just a little. It's at these times that the power button gets pressed, and the Kindle goes to sleep or, even worse, shuts down. This invariably happens during the most important passage of a 400-page book. It never fails. I cry out in surprise as I realize my mistake.

Misplaced

The newest Amazon Kindle Paperwhite (left) and last-generation restarting (Image credit: Future / Lance Ulanoff)

I also feel that Amazon has made a mistake. Why is the power button there? I'm not the only one asking this question. It's quite easy to find a few Reddit threads addressing the question. In these threads are various plausible explanations for the placement of the button, such as the location of the motherboard (which, based on teardowns, does sit at the bottom of the device), the need to balance the weight of the devices, or the possible necessary proximity to the USB-C power port.

None, though, are quite right.

When I spoke to Amazon's Kindle hardware lead Kevin Keith about the issue, he offered a different explanation, but not before trying to solve my problem.

First Keith asked if I had a cover for my Kindle, because the thickness of the cover would immediately mitigate this issue by putting the power button level with the surface of the Kindle cover. No more accidental presses. But I hate covers. I'm a travel-light kind of guy, and I try to shed anything that adds any unnecessary weight. Kindles are plenty tough in my opinion (the Signature models are waterproof) and don't need that extra protection. Keith joked that they needed to get me a cork cover which would be a lot lighter.

And that's when Keith opened up about the real reason for the button placement.

"We've looked at that because we've gotten this feedback and a lot of it is to keep the dimensions really tight, and so if you remember what Panos was saying, the thing you don't want to mess with, with a Paperwhite, is how it feels in your hand for long reading sessions of one-handed reading, and so that's why when we increase the screen viewing size we always have to shrink the bezels. And the [fact that the] button has to be near the light guide is essentially what is driving this."

Okay, I know, that doesn't entirely clear this up, but the thinking here appears to be related to, for one thing, cutting down the distance between components, especially the power and the panel that guides the LED lighting (which is arrayed across the base of the Kindle) across the entire screen. It also means you're not reaching across the screen to power up the Kindle, and thereby not covering the light sensor on Paperwhite Signature editions that automatically adjust to light conditions (granted the sensor will catch up in short order and adjust the lighting as soon as you move your hand). Finally, Amazon isn't putting the power button on either side of the Kindle, or even the back, because it might change how the Kindle feels in your hand.

Put another way, Amazon obviously has no intention of moving that power button any time soon. Put yet another way: have you thought about buying a case?

You might also like

• Hands on: Amazon Kindle Scribe 2024: An excellent 2-in-1 eReader gets annotation and a healthy infusion of AI
• Hands on: Amazon Kindle Paperwhite (2024): Super fast and bigger, but refreshingly familiar
• Hands on: Amazon Kindle Colorsoft: The color Kindle you've been dreaming of is

Unity has announced that Unity 6, the latest version of its cross-platform game engine, is now available worldwide.

Unity 6 can be downloaded here, and is the company's "most stable and best-performing version of Unity" yet that has been built, tested, and refined in partnership with game developers around the world.

According to Unity, the updated software will allow developers to create games much faster and more efficiently than ever before and comes with new features like end-to-end multiplayer workflows, tools that target the mobile web, alongside new graphics capabilities that move workloads from the CPU to the GPU, improving CPU performance by up to 4X in internal.

Unity has also shared its plans for the engine post-launch, confirming that it will be "dedicating long-term product and engineering resources to Unity 6" to enhance features and add new functionality, while continuing stability.

"We are relentlessly focused on delivering tools to help game developers build games more quickly and efficiently while also facilitating innovation," said Matt Bromberg, President and CEO of Unity. "We’re going to do everything we can to ensure that Unity 6 is at the heart of game development for years to come."

Alongside the launch of Unity 6, the company has also introduced some brand-new learning resources for developers, like Time Ghost, the latest Unity Originals real-time cinematic demo which includes advancements in environment building and character design, as well as Megacity Metro, a demo that showcases how to build a 100+ player cross-platform multiplayer game.

"Unity 6 gives our global multi-functional team the stability and scalability we need. The improved performance combined with features like the new WebGPU graphics API and the seamless live services integration make our workflow much more efficient and our production quality top-tier,” said Josh Loveridge, managing director of Stratton Studios.

"It's been a game-changer while developing PGA TOUR Rise. We’ve truly been able to push the boundaries of creativity at every step of the development pipeline."

You might also like...

• Rare boss Craig Duncan to take over as Xbox Game Studios' head in November
• PlayStation: The Concert is a new video game music show coming in 2025, and will feature famous scores from Sony's most iconic games like The Last of Us and God of War
• Call of Duty: Black Ops 6, Modern Warfare 3, and Warzone are coming to Xbox Cloud Gaming next week

Marc Benioff, the CEO of Salesforce, a company that has been investing heavily in artificial intelligence tools for its portfolio, has called Microsoft’s Copilot AI a “tremendous disservice” to the industry.

In a recent interview on the Rapid Response podcast, Benioff likened Copilot to Microsoft’s early office assistant, Clippy, suggesting the service is disappointing and fails to deliver meaningful value.

The comments come despite Microsoft’s multibillion-dollar investment in ChatGPT maker OpenAI several months ago, and the subsequent injection of what is considered one of the best AI models available into its products.

Salesforce CEO criticizes Microsoft’s Copilot AI

Benioff went as far as suggesting Microsoft’s extensive AI efforts might not be around for long, predicting faded hype and a dwindling user base.

In the podcast, he added: “It doesn't work. It spews data all over our floors, it doesn't deliver value. I haven't found a customer who has transformational work with Copilot. Copilot is just the new Microsoft Clippy.”

Keen to separate Microsoft’s efforts with those of Salesforce, the CEO praised the company’s new Agentforce tool for helping to transform businesses by delivering tangible value. Agentforce is already reportedly handling “a couple of trillion AI transactions per week,” leading to satisfied customers.

Looking ahead, Benioff alluded to an AI landscape dominated by AI agents destined for enterprises, suggesting the demise of Microsoft Copilot, which promises to deliver productivity and efficiency improvements for workers.

Despite the CEO’s criticism, Microsoft’s influence in the AI space remains strong. Its relative early entry combined with its affiliation with OpenAI put it high on the scoreboard among companies choosing to employ AI. Subsequently, Microsoft has plans to continue investing in its data centers to support future expansion.

Microsoft became the second company to reach a market cap of $3 trillion, well into its AI journey, and analysts are predicting that it could be the world’s most valuable company in the years to come. Apple currently holds that title, but chipmaker Nvidia, which is responsible for providing those all-important data center chips, has also entered the race, and strongly.

More from TechRadar Pro

• Check out the best AI writers
• We’ve listed all of the best productivity tools to help workers get more done
• Nvidia could be about to knock Apple off the top spot for most valuable company

For many, watching Die Hard is a Christmas tradition. Believe it or not, the iconic 80s action movie can be as comforting as watching Home Alone or Elf. But now, your traditional holiday flicks are about to get an exciting addition with Netflix's new thriller Carry-On that's set to bring some Die Hard action this festive season.

Taron Egerton has become something of an action star himself due to his roles as secret agent Gary "Eggsy" Unwin in the Kingsman franchise and heroic outlaw Robin Hood in the modern retelling of the mythical legend. In Carry-On, Egerton packs a punch as a young TSA agent who battles against a villainous Bateman (Ozark, Arrested Development), a mysterious passenger who blackmails him into allowing a strange package onto a busy flight on Christmas Eve and endangering everyone's lives onboard.

If airport security wasn't hard enough, this new Netflix movie makes it even more challenging in the new trailer (see below). Bateman sets the sinister tone of Carry-On with the ominous voiceover: "You have ten minutes before everybody in the airport dies. All you have to do is nothing." Cue a montage of high-octane mayhem with car crashes and gun fights, then you have everything you could want from a holiday action flick.

What is Carry-On about?

Landing on the best streaming service on December 13, Carry-On follows TSA agent Ethan (Egerton) working on Christmas Eve. If a heaving airport filled with irritated passengers wasn't enough to contend with, his arduous shift soon turns into a full-blown nightmare when a terrorist (Bateman) blackmails him into letting a dangerous package slip through the scanner and onto a Christmas Eve flight.

Egerton told Entertainment Weekly of his character: "I think he's someone who lacks a sense of drive and direction but is ultimately really resourceful and capable. He's a bit of a hero in waiting. He's pretty noble and selfless. He has all the qualities of an archetypal hero underneath the lack of direction."

Just looking at the trailer, it has the potential to become one of the best Netflix movies or even land a spot on our best Netflix Christmas movies list alongside the streamer's upcoming festive rom-com Our Little Secret starring Lindsay Lohan.

Other cast starring in Carry-On are Sofia Carson as Ethan's girlfriend whose life is threatened by the terrorist, Danielle Deadwyler, Theo Rossi, Dean Norris, Logan Marshall-Green, and Josh Brener.

You might also like

• Lonely Planet is Netflix's #3 most-watched movie but it's not loved by critics – here are 3 better romantic dramas with over 85% on Rotten Tomatoes
• Netflix’s Sweet Bobby: My Catfish Nightmare isn’t the only catfishing documentary worth streaming this week
• Two of Netflix’s biggest shows have been renewed – maybe I can trust it not to cancel everything I like again

Samsung has released a new video teaser, which we think hints at an impending launch for the long-rumored Samsung Galaxy Z Fold 6 SE.

The video shows a person carrying an envelope through a hallway before revealing a tablet-sized device with the words “You’re Invited” displayed on the screen – we think this device has a solid chance of being the Galaxy Z Fold 6 SE, previously known as the Galaxy Z Fold Slim.

Confusingly, the translated press release that accompanies this video states a reveal date of October 21, while the video itself suggests an event on October 24. It’s possible that these are sequential reveal and release dates, but we can’t say for sure.

(Image credit: Samsung)

The main selling point of the Z Fold 6 SE is expected to be its design – even the earliest rumors referred to it as a thinner version of the flagship Galaxy Z Fold 6. We’ve seen estimates for folded thickness that vary from 7.7mm to 11.5mm, and the latest rumor from GSMArena points to an impressive unfolded thickness of 4.9mm.

More recently, we’ve seen suggestions that the Z Fold 6 SE will feature a less prominent crease through the center of the inner display, which would allay one of our last remaining complaints with Samsung’s Z Fold devices.

And yet more rumors claim the Z Fold SE will have larger displays than its predecessor, with Korean news outlet Financial News (via 9to5Google) suggesting a 6.5-inch cover display and 8-inch inner display.

The Galaxy Z Fold 6 was released on July 24, meaning it’s far too soon for a fully-fledged sequel. As for other clues in the teaser, the envelope could hint at the device’s thinness, and the “You’re Invited” wording certainly feels exclusive enough for a ‘special edition’.

The 11-second teaser was uploaded to Samsung’s Korean-language Samsung Newsroom YouTube channel, but is conspicuously absent from the equivalent US or UK channels.

This tracks with prior rumors: as we previously reported, the Galaxy Z Fold 6 SE will likely be exclusive to Korea and China when it finally launches.

For now, much of the above is based on rumors, but it seems like we won’t have to wait long for official details. It’s very unlikely that the Samsung Galaxy Fold 6 SE will launch worldwide, but this is otherwise poised to be one of the best folding phones on release.

For the latest official updates as we hear them, be sure to keep up with our Samsung phones coverage.

You might also like

• Another company just beat Samsung to Android 15 – and it’s not who you’d expect
• Android 15 has broken Instagram for many – here's how to fix it
• Is it time to switch from Apple to Samsung? iPhones reportedly aren't holding their value as well as they used to

Google is finally letting the sun go down on Manifest V2 as it switches to V3, meaning that many apps reliant on V2 will no longer work - including one of the most popular ad blockers, uBlock Origin.

Manifest is part of the Google Chrome browser that governs how browser extensions function, as well as the permissions they are granted. The V3 version of Manifest is not as liberal with its permission grants, rendering many ad blockers and other extensions effectively useless.

Ad blockers have grown in popularity for users looking to speed up the performance of their browsing experience, reduce the amount of data websites collect on them, or simply get rid of clutter displayed on articles. Unfortunately, Chrome generates a hefty sum through advertising which is undercut by ad blockers.

Grin and bear it or time to switch?

uBlock Origin Lite, a barebones version of the main version, will still work with Manifest V3 but will not block every single ad. Those looking to continue using uBlock Origin may want to consider using a different web browser.

Google says the move to V3 brings additional benefits such as more capabilities, new APIs, and enhanced security.

“Manifest V3 removes the ability for an extension to use remotely hosted code, which presents security risks by allowing unreviewed code to be executed in extensions. With this change, an extension can only execute JavaScript that is included within its package and subject to review by the Chrome Web Store,” the Manifest V3 page states.

It is important to note that Manifest V3 is a change to the Chromium engine, but so far only Chrome has removed the ability to use V2, with Microsoft Edge and Opera continuing to offer V2 support with no confirmed dates for a V2 switch off.

More from TechRadar Pro

• These are the best privacy tools and anonymous browsers
• More and more users are set to opt out of third-party cookies on Google Chrome
• Take a look at our guide to the best VPN

Blizzard Entertainment and iam8bit have announced a Diablo 4 vinyl soundtrack release.

Pre-orders for the Diablo IV 4xLP box set are now live at the iam8bit official online store and will set you back $100 (around £77).

The four LPs will feature the iconic music from the game from Leo Kaliski, Ted Reedy, Ryan Amon, and Derek Duke. The discs included showcase designs that pay homage to some of the game's classes and named have colors named after classic moves: Grizzly Rage Gold, Grim Harvest Green, Frost Nova Blue, & Rapture Red.

Artist Ryan Brinkerhoff has crafted the album art, celebrating Diablo 4's fighters, with each sleeve paying tribute to a different class’s strength and iconography. The entire set is wrapped up in a gorgeous outer album box.

In addition to four LPs, the box set will also feature an exclusive code for an in-game Mount Trophy that can be redeemed via Battle.net and can be used by all classes.

"Getting around Sanctuary’s no easy task, but it’s a hell of a lot better when you’re looking your best - and nothing’s swankier than a precious ornament for your ride," iam8bit teases. "Whether you’re playing co-op with a friend or fighting online foes, you’ll be able to show off your devotion in style.

The exclusive Diablo IV 4xLP is expected to ship in Q2 2025, meaning buyers should expect their pre-orders to arrive between April and June next year.

Diablo IV launched last year for PC, PlayStation, and Xbox. TechRadar Gaming rated the game five out of five stars, with writer Cat Bussell calling it a " meticulously crafted action RPG" that "sets a new standard for isometric, open-world adventures."

You might also like...

• Best PC games 2024: the must-play titles you don’t want to miss
• Rare boss Craig Duncan to take over as Xbox Game Studios' head in November
• Xbox Cloud Gaming will reportedly let players stream their own games from next month

Arcane season 2 continues to deliver exciting new teasers and the latest from Netflix was set to a new song, which is enough to get anyone pumped for upcoming scenes when it arrives on the best streaming service in November.

We were already very excited about the Arcane season 2 soundtrack with our senior entertainment reporter Tom Power suggesting it would hit the number one spot on his Spotify Wrapped this year. And I'm inclined to agree, considering the epic new track they've just debuted.

Check out the Come Play trailer for one of the best Netflix shows below, featuring the titular track by Stray Kids, Young Miko, and Tom Morello. Yes, you will want to listen to it on repeat. No judgement from me!

What do we know about Arcane season 2 so far?

Arcane fans have been eating well recently as Netflix has done a lot to promote the upcoming season, with plenty of teasers and clips to get us excited for what's to come. We also know that Arcane season 2 is the final season so it's been a short and not always sweet journey for the League of Legends series.

In an interview with TechRadar, star Ella Purnell (who plays Jinx) teased that season 2 "will emotionally devastate fans", so it sounds like we've got a big storm coming. You might want to stock up on tissues before the series debuts, as it sounds like it'll pull some big emotional punches.

Plot wise, we know that there'll be an even greater focus on Jinx and Vi's relationship as the latter tries to stop her sister from leading a revolution, and if Purnell's comment is anything to go by we know it's not going to be an easy watch. All will be revealed next month and we can hardly wait.

You might also like

• The Night Agent season 2: Netflix release date prediction, likely cast, plot synopsis, and more news and rumors
• Netflix unleashes the trailer for its new werewolf series Wolf King but that’s not the only animated show I’m excited for
• 7 shows canceled by Netflix in 2024 so far that you should still watch

A company was hacked after hiring a fake IT professional from North Korea. It has not been clarified whether this was a deliberate cyberattack against the organization, a disgruntled former employee, or a “simple” scam.

The company, which was not named, operates either in the US, UK, or Australia. It sought to add an IT professional to the team, and tapped into the global talent pool. There, it found a suitable candidate, who obviously went through the hiring process and got the job.

The person that was hired, however, faked their entire identity, including knowledge and previous experience. After being hired, the scammer accessed the company’s infrastructure and downloaded as much sensitive information as they could.

Simple scam, or something more?

The miscreant worked for four months with the company, before allegedly being fired for poor performance. After that happened, the crook threatened to release all of the stolen data on the internet, or sell it to the highest bidder. He demanded a six-figure ransom in exchange for keeping the data private.

According to the BBC, it is not known whether the company paid the ransom or not.

This could either be a simple scam, or a disgruntled former employee taking revenge upon their former employers. However, it could be something more.

Lazarus Group, a North Korean state-sponsored threat actor, is known in the cybersecurity community for its “fake job” attacks. Usually, they would set up a fake job ad on social media and try to “hire” software developers working in high-profile organizations. During the interview process, they would trick the candidate into installing malware, gaining access to their company’s IT infrastructure.

The attack works both ways, too, since the crooks were seen targeting organizations directly, by trying to get hired. Lazarus apparently goes for people’s cryptocurrency and uses the money to fund the state’s weapons program.

Via BBC

More from TechRadar Pro

• North Korean Lazarus hackers are using a fake coding test to steal passwords
• Here's a list of the best firewalls today
• These are the best endpoint protection tools right now

The Xiaomi Smart Band 8 Pro might be the best cheap fitness tracker on the market, but a new patent has revealed the company may also be working on a smart ring that could come with a breakthrough new feature.

The market is currently the domain of giants like Oura and Samsung, whose Galaxy Ring and Oura Ring 4 make up the best smart ring options right now.

Every smart ring today comes in various different sizes, and you usually need to buy a sizing kit to get the right fit. But this annoying and time-consuming process could soon be a thing of the past if Xiaomi's new patent is anything to go by.

As spotted by 91Mobiles, the patent, filed in China, reveals that Xiaomi is working on a smart ring with an auto-adjustable design. It features an elastic component and a spring mechanism that allows the ring to adjust its size automatically.

One smart ring (size) to rule them all?

(Image credit: Oura)

If Xiaomi has indeed cracked a one-size-fits-all smart ring, it could be a massive shakeup for the industry. Not only would it reduce the time and cost involved in choosing the right size of a smart ring, but as the report notes it could pave the way for a ring that can be worn on any finger or can be shared by more than one person.

Behind the scenes, such a smart ring would also be cheaper to manufacture and replace in case of issues, possibly reducing the price for consumers.

This ground-breaking advance aside, Xiaomi's smart ring includes support for Wi-Fi, 5G, and an array of sensors including a gyroscope and temperature. Interestingly, there are also references to a speaker module, a microphone, and a touch panel built into the ring. As the report notes, there are hints too of using the ring to control smartphone features like a camera, suggesting that this could be a companion device like an Apple Watch.

A patent is of course just an idea committed to paper, and not hard evidence of Xiaomi's upcoming product plans. But it's safe to say that the company is definitely looking hard into smart rings.

You might also like

• I spent 24 hours with the Samsung Galaxy Ring, and it's scarily good so far
• I tried wearing an Oura smart ring and might never go back – but it can’t replace my Apple Watch
• Sorry, there's no way Apple is canceling its rumored smart ring plans

Apple’s MacBook Pro models with the M4 chip are at this point strongly rumored for an imminent launch, and another piece of evidence has surfaced to reinforce this idea.

This is chatter from the supply chain disclosed by DigiTimes (in report about the notebook market in general, flagged up by MacRumors).

DigiTimes informs us: “Apple is expected to unveil new MacBook Pro models equipped with the latest M4 chip in October, which could boost related component shipments. This trend is evident in the revenue reports from component manufacturers such as Jarllytec and Shin Zu Shing for September.”

The most recent rumors align with the late October launch mentioned here – which should mean we get the press invites being sent out by Apple very soon indeed – and this would come ahead of a November 1 on-sale date, theoretically.

If the info from the grapevine is correct, we can expect MacBook Pro, Mac mini and iMac models with M4 chips to be launched at the end of the month. Next year, it’ll be the turn of the MacBook Air to get an M4 makeover, plus the Mac Pro and Mac Studio – in theory, anyway.

(Image credit: Future) Analysis: A veritable avalanche of leaks

Who are the companies mentioned in this most recent supply chain leak, then? Jarllytec is a Taiwan-based outfit that makes notebook hinge mechanisms (fancy ones), and has supplied for MacBooks before (for a long time in fact). Shin Zu Shing also provides components for Apple’s laptops (and reportedly recently benefited from a shift in Apple’s ordering where the firm ditched a previous supplier).

So, apparently the respective revenue sheets for those companies make for some telling hints regarding the imminent arrival of MacBook Pro M4 models, presumably due to beefy shipments of various related components.

There are plenty of other signs that a MacBook Pro launch is imminent, including a rather staggering flood of M4 MacBook Pro leaks which include a whole bunch of spilled pictures and videos and even a report of the laptop being sold already on the black market. Also, current stock levels of the MacBook Pro (and other Macs) are reportedly ebbing away, as you’d expect just before the release of new models.

At this point, it’d be a serious surprise – a shock even – if the next-gen MacBook Pro didn’t turn up. After all, this is the device that’s been a lot more heavily rumored than the other M4 Macs Apple is expected to launch later in October. The not-so-great news for MacBook Pro fans is that not that many changes are expected in the new take on Apple’s laptop, aside from the addition of the M4 SoC, that is.

You might also like

• I used to laugh at the Mac Mini but today I bought one
• These are the top MacBook deals right now
• Apple just declared a bunch of Macs obsolete – these are the affected MacBook and iMac models

Getting your hand son the brand-new DJI Air 3S – which we gave four-and-a-half-stars in our DJI Air 3S review – might be trickier than anticipated as the company is having issues importing its latest drone to the US. For the time being, it might only be possible to purchase the drone from DJI.com directly, not through any third-party stores.

This isn’t anything to do with the ban bill which passed in the US House of Representatives earlier this year. Instead, DJI claims it’s because the Department of Homeland Security incorrectly believes the drones were produced in forced labor camps and is blocking the drones using the Uyghur Forced Labor Prevention Act (via The Verge)

In a letter it sent to distributors – which has been shared on social media – DJI refutes these accusations stating “DJI strongly affirms that no forced labor is involved at any stage of our manufacturing process.

“Our company does not and has never operated manufacturing facilities in the Xinjiang province of China, nor do we obtain any materials from that region. All of our manufacturing is based in Shenzhen, where our company is headquartered, or Malaysia.”

Following up with a post on its official blog DJI says it has all of the evidence it needs to clean up this “misunderstanding” and prove it doesn’t make its drones in the Xinjiang region of China.

Until the issue is resolved however it’ll be harder for US businesses to get their hands on DJI’s enterprise and agricultural drones, as well as regular folks from snagging the DJI Air 3S – at least from retailers other than DJI itself.

(Image credit: DJI) The first ban of many?

Even if this existing issue is resolved, DJI’s future in the US will remain uncertain.

The wider US import ban is still looming, and while it has yet to pass through the Senate it feels like it may be a matter of when, and not if, unless there’s a major attitude shift in the US government.

Though the ban looks set to only affect imports as it is currently written, so if a DJI drone catches your eye, you should still consider grabbing it – as already-approved models won’t be permanently grounded even if a ban does pass.

So take this DJI Air 3S issue as a taste of what’s unfortunately likely to come in the coming months and years from the world's best drone manufacturer.

You might also like

• DJI’s new dual-camera Air 3S drone could be the ultimate all-rounder
• The best drone 2024: top flying cameras for all budgets
• DJI Air 3S vs Air 3: worth the upgrade?

A new report jointly created by MIT Technology Review Insights and data company Snowflake has confirmed that many businesses are simply failing to harness the power of artificial intelligence.

As many as four in five (78%) organizations are unable to fully capitalize on their AI investments according to the study, largely due to inadequate data management.

Data governance, security and privacy concerns emerged as the top challenge (59%), followed closely by data quality and timeliness (53%) and the costs associated with AI resources (48%), highlighting the need for a simpler and clearer foundation.

Data is preventing companies from realizing AI’s benefits

It’s clear that there are challenges between initial investments and tangible rewards, but companies still appear to be optimistic. Nearly three-quarters (72%) agree that AI would boost efficiency and productivity, with around half believing it will enhance market competitiveness (55%). However almost all (95%) reported facing hurdles when implementing AI.

While more than half (53%) of business leaders describing themselves as ‘somewhat ready’ for artificial intelligence and everything that it entails, only one in five (22%) state that they are ‘very ready,’ signifying that they have fewer challenges related to accessing scalable computing power, data silos and integration issues, and data governance.

Speaking about the “big ambitions” that companies have surrounding GenAI, Snowflake Head of AI Baris Gultekin said: “A strong data foundation is at the core of generative AI capabilities, and business leaders need to move quickly to deal with concerns such as data security and cost, and establish the foundation they need to deliver on the promise of AI.”

Moving forward, the report calls for modern cloud data platforms that allow companies to better manage and access vast amounts of unstructured data.

More from TechRadar Pro

• Ready to get on board with artificial intelligence? Here are the best AI writers
• Businesses are spending more and more on AI, but keep hitting the same roadblocks
• Check out our roundup of the best cloud hosting providers

Polar, makers of some of the best heart rate monitors and best running watches, experienced a cybersecurity attack this week. On 11 October, Polar posted a statement on the “updates” section of its website, informing users and customers that it was subject to ‘various levels of disruption’, and has shut down the login and signup pages as a result.

If you’re a Polar user, you might be worried your sensitive health data is at risk, but speaking to TechRadar, Polar says your data ‘has not been compromised’.

On its website, Polar states: “Polar’s online store in the United States has been subjected to various levels of disruption, resulting in the suspected compromise of individual users’ order information… According to preliminary information, only a few customers' account details have been accessed."

Polar went on to state: “The data breach has affected only a small portion of customers in Polar’s US online store. The breach does not apply to any other customer data stored by Polar. We want to emphasize that the data of Polar Flow users has not been compromised during the attack.” Polar echoed these sentiments in a statement made to TechRadar after we reached out for comment.

Polar emphasizes that users don’t need to do anything at this stage, although if you want to buy anything on the Polar website, you’ll have to do so from a guest account.

(Image credit: Sora Shimazaki / Pexels) Analysis: Health data makes a tempting target

While in this instance, Polar states that order information rather than personal health data was compromised, Polar Flow makes a tempting target. Health data is incredibly valuable to hackers, advertisers, and even certain governments due to its intensely personal nature: imagine taking your medical records and GPS location, and giving that information to known criminals to sell to invasive advertising services and identity thieves.

Polar was at the center of a data leaking scandal back in 2018 due to flaws in user privacy settings, but things have tightened up since then. Garmin has also experienced its own cybersecurity problems, as Russian hacking group Evil Corp attacked Garmin in 2020, taking its services offline for more than three days.

How to protect your data

In this day and age, it’s hard not to put our trust in placing everything online. As TechRadar’s Senior Fitness and Wearables Editor, it’s my job to test the latest data-harvesting health gadgets: my information is probably scattered over a dozen different databases. So how can you protect yourself?

Generally, if you’re still receiving value from a service like Strava or Fitbit Premium, you have to continue supplying the service with your health data. It’s worth taking a look at fitness tracker privacy policies if you’re concerned about it, which should list if the data is shared with any third parties. If you’ve left data on a service you no longer use, and want to minimize the risk of being involved in a data breach, you may be able to ask the company to delete the data it has on you.

In the US, whether you have a right to delete data depends on what state you live in: according to the personal information removal service DeleteMe, there are only 12 US states with comprehensive privacy laws. In the UK, things are a bit simpler: you can use the Information Commissioner’s Office guidelines to withdraw consent to use your personal data in writing. Australians will also find it difficult to get their data removed, although the government has previously mooted enshrining a “right to be forgotten” into Australian law.

You might also like...

• Where is all your health data going? The Google and Fitbit scandal explained
• Best cheap fitness trackers 2024: Inexpensive wearables to monitor your health
• Exclusive: Polar's CEO talks about the Sennheiser collaboration and a potential future smart ring

Having given us four new Kindle models yesterday, Amazon has taken one away: the Kindle Oasis, first launched in 2016 and last refreshed in 2019, is now discontinued – marking the end for Kindle devices with page-turn buttons.

"Once current inventory of Kindle Oasis sells out online and in stores, we will not restock the device," Amazon's Devon Corvasce told The Verge. The device is currently showing as unavailable on the Amazon US site, with used models available on Amazon UK.

According to Corvasce, Kindle users are more comfortable with tapping the screen to go forwards and backwards through their ebooks, and that's the approach all Kindle models will be taking from now on.

In our Amazon Kindle Oasis review, we described it as a "luxury ereader" that just about justified its higher price. The button system did add some extra bulk, but it worked well, and it was also easy to hold in one hand thanks to a ridged-back design.

Past and future

The new Amazon Kindle Colorsoft (Image credit: Future / Lance Ulanoff)

With no updates since 2019, it's perhaps no surprise that Amazon has decided to put an end to the Kindle Oasis. Between the 2016 launch and the final 2019 version, there was one other update, which came in 2017.

The display on the most recent Kindle Oasis was 7 inches, corner to corner, and both brightness and warmness could be adjusted. Brightness could also be set to change automatically, based on the ambient light conditions.

Looking to the future, you can check out our hands-on Amazon Kindle Colorsoft review for our first impressions of the first Kindle to come with a color screen. It'll set you back $279.99 / £269.99, with pricing for Australia yet to be announced.

The other new models just announced by Amazon are refreshes for the Kindle Paperwhite, the Kindle Scribe (with stylus support), and the entry-level Amazon Kindle, so there's plenty of choice if you're looking to replace your aging Kindle Oasis.

You might also like

• Here's what Amazon gets right about the new Kindle Scribe
• These are the best ereaders you can buy
• Kindle vs Kobo: how to choose

The exploitation of zero-day vulnerabilities, flaws that were abused before the developers built a fix, is growing faster than the exploitation of n-day vulnerabilities (those for which a patch is already available).

This is according to a new report from Google’s cybersecurity researchers Mandiant, who describe it as a “worrying trend”.

The researchers recently analyzed 138 exploited vulnerabilities that were disclosed in 2023, and concluded that 70% were abused as zero-days, while 30% were n-days. In previous years, the ratio was closer to 60% for zero-days, and 40% for n-days, meaning the crooks are growing increasingly reliant on zero-day vulnerabilities.

Social engineering

“While we have previously seen and continue to expect a growing use of zero-days over time, 2023 saw an even larger discrepancy grow between zero-day and n-day exploitation as zero-day exploitation outpaced n-day exploitation more heavily than we have previously observed,” the researchers explained.

Besides the increase in the numbers, the average time-to-exploit (TTE) has also decreased, suggesting that the attackers are exploiting these flaws faster than ever before. Two years ago, the average TTE was 32 days. Last year, it was merely five days, meaning the flaws are getting abused almost immediately.

But there is a silver lining to the research. Mandiant says organizations have gotten better at detecting zero-days, which also resulted in higher numbers in the report. It is quite possible that in previous years, a larger portion of these attacks went unnoticed. Companies have also gotten better at patching. They do it faster, and more frequently nowadays, forcing the hackers to move faster themselves. Hence the shorter TTE.

Looking into the future, Mandiant says the trend of zero-day exploitation is expected to grow, especially with improved detection tools. Zero-days are likely to remain a highly coveted approach for threat actors because they offer a critical window of attack before patches can be applied.

If this trend continues, Mandiant anticipates time-to-exploit will fall even further.

More from TechRadar Pro

• What are zero-day vulnerabilities?
• Here's a list of the best firewalls today
• These are the best endpoint protection tools right now

Motorola has begun rolling out a beta version of Android 15 to select Edge 50 devices, becoming one of the first phone makers to give users a taste of the latest Android operating.

Currently, only some Motorola Edge 50 Fusion devices are eligible for the beta, and we aren’t quite sure of what the criteria for selection is – nonetheless, it’s very encouraging to see this type of movement early on.

Such early deployment of Android 15, even in beta form, comes as something of a surprise from Motorola given that the company isn’t exactly known for rapid upgrades. However, Motorola launched the Edge 50 series with the promise of five years of OS updates, so speedier updates could be a product of this new commitment.

Google remains the sole phone maker to deliver a stable Android 15 update thus far, having pushed out a Pixel Launcher update on October 15. Google’s haste makes sense considering that the company develops the Android source code, which was itself released on September 4.

We’d usually expect Samsung to be hot on the heels of Google when it comes to adopting major Android updates, but we’ve heard nothing official about the expected Samsung One UI 7 update yet.

Samsung has promised the Android 15-powered update to a wide array of phones, from the ultra-budget Galaxy A14 to the flagship Galaxy S24 Ultra.

As YTechB reports, the list of Motorola devices expected to get Android 15 updates includes the Edge 50 series, Razr phones launched in 2023 or 2024, and several G series handsets.

In global markets, Xiaomi and Honor have both enabled an Android 15 beta for select devices, but like Motorola neither of these companies have delivered a stable Android 15 update.

The fight for second place in the Android 15 stable release race is still on – for the latest updates as we hear them, be sure to keep up with our Android coverage.

You might also like

• Android 15 lets you turn your phone into a handy smart-home dashboard – here’s how
• The latest Pixel Drop is here, bringing upgrades to cameras, Call Screen, and more
• Android phone makers are wasting no time in copying the iPhone 16 camera button

As more and more companies switch their apps and services to subscription models, you might have found it increasingly difficult to cancel your subscription and end the payments. Most of us have struggled with a company that makes it near-impossible to stop the charges, but that could soon become a thing of the past thanks to new consumer-friendly rules that are about to come into effect.

In a press release, the FTC has announced that businesses will have to make it as easy for consumers to end a subscription as it is to sign up for one. The new 'click-to-cancel' rule is designed to address complaints about 'negative option' billing – that is, subscriptions and recurring payments where you must take action to have them canceled.

Right now, some companies make it as difficult as possible to end a subscription. Instead of allowing you to end it online, for example, they might make you call a phone number where a sales representative will cajole you with offers and other tactics designed to keep you spending. While signing up for a recurring payment can take seconds, canceling one can take hours – or longer.

(Image credit: Shutterstock / Ivan Marc)

Once the click-to-cancel rule comes into effect in roughly 180 days’ time, that should all change. The FTC says its new guidelines will prohibit companies from engaging in the following practices:

• Misrepresenting any material fact made while marketing goods or services with a negative option feature
• Failing to clearly and conspicuously disclose material terms prior to obtaining a consumer’s billing information in connection with a negative option feature
• Failing to obtain a consumer’s express informed consent to the negative option feature before charging the consumer
• Failing to provide a simple mechanism to cancel the negative option feature and immediately halt charges.

The FTC says it opted against enacting other rules. These include the requirement for companies to send annual subscription reminders to customers, as well as a ban on firms suggesting reasons for you to keep paying when you attempt to cancel a subscription if you haven’t previously asked to hear about these reasons.

Commenting on the news, FTC chair Lina M. Khan said, “Too often, businesses make people jump through endless hoops just to cancel a subscription. The FTC’s rule will end these tricks and traps, saving Americans time and money. Nobody should be stuck paying for a service they no longer want.”

With subscriptions becoming an increasingly common way for companies to bring in revenue, complaints have been rising at an equally steep rate, with the FTC saying it now receives an average of 70 complaints a day about shady recurring payment practices. Hopefully the new rules will put a stop to some of the most egregious examples and make it far easier to get more control over your money.

You might also like

• I wasted megabucks on tech subscriptions – here's how to avoid my mistakes
• I’m cutting my TV streaming bills by 75% in 2024 with subscription hopping – here’s how
• Can't decide which streaming service to cancel? This app makes subscription hopping between Netflix, Disney Plus and more easy