British employees are sounding the alarm over the rise of AI agents in the workplace, warning that the tools are unreliable, unresponsive to feedback, and ...

OpenAI and Apollo Research tried to stop models from lying - and discovered something else altogether.

While Spotify has finally entered the hi-res streaming game, it technically still lags behind its rivals. But should you care?

Researchers just got very specific about what a language model's 'chain of thought' is actually doing.

Full Fact analysis of the government's elections strategy

Some Facebook users have noticed new settings that let Meta analyze and retain your phone's photos. Yes, you read that right.

The Online Safety Act does not require age verification to use a virtual private network.

The emergence of AI is spurring some firms to make their HR and IT departments work closer together.

Following the rollout of the UK Online Safety Act, Britons are now being asked to confirm their age across the internet. Here’s everything you need to know.

AI can boost productivity, but it can also derail your entire operation. From fake legal advice to customer service nightmares, here are nine places AI doesn't belong at work.

Ask more complex, nuanced questions on Google Search with AI Mode, now available in the UK.

Freely is releasing a streaming box of its own

Vertiports are going to be future of mobility, says Dubai Airports CEO Paul Griffiths

With governments demanding actual age verification on websites with adult content, and platforms like social media and Roblox introducing restrictions based on a user’s age, the controversy about different types of age verification and their implications is growing. Last week, Roblox announced new age estimation technology which, it says, should help to confirm users’ ages and unlock a feature called Trusted Connections for those aged 13 and older. Trusted Connections allows teens aged between 13 and 17 to add adult users (18+) they know in real life. It’s billed as an option to keep out predators, which is good. But the age estimation technology raises concerns and questions. While Roblox didn’t release any details about how its new technology works, the age estimation processes we know are based on Artificial Intelligence (AI) tools that scan selfies or short videos and compare them to a database to estimate the user’s age. Needless to say, they are not always right and it opens up the system to deepfakes, and spoofing. This kind of technology is more effective than asking the user to provide their birthday or check a box that they are over 18, but it’s not waterproof. We see similar concerns when it comes to age verification for sites that host adult content. As of this Friday, websites operating in the UK with pornographic content must “robustly” age-check users. The regulator, Ofcom, lists a number of allowed methods which all have their pros and cons: **Facial age estimation** Show your face, get a guess. You take a selfie or a short video, and an algorithm tries to figure out if you look over 18. The tech claims to keep data private, but facial scans are sensitive. And, as we pointed out, accuracy is far from perfect. If you’ve ever been asked to provide an ID in real life because you “look young,” you can expect a digital déjà vu. **Open banking** Banks know your age, so why not let them confirm it? Here, you allow the age-check service to peek at your bank account. No bank statements get handed over, just a yes/no to the question: “Is this customer an adult?” It’s easy, but convincing users to link their bank to a porn site might be a different story. **Digital identity services** This is the world of digital wallets for your ID. Think of it as carrying your driver’s license in your phone, but only showing the “over 18” part when needed. Sounds great and it is, but you’ll need yet another app in your digital life just to vouch for your adulthood everywhere you go. **Credit card age checks** Simple logic: you need to be 18+ to have a credit card, so showing a valid one counts as proof. The age-checker pings the payment processor to see if your card is legit. It’s quick and familiar, but not everyone over 18 has a credit card, so it’s not for everyone. Plus your purchase trail grows with every verification. **Email-based age estimation** Enter your email address. The system tries to deduce your age using records of that email in other “adult” places, like financial or utility services. Basically, you’re allowing digital snooping, and the effectiveness depends on your online life elsewhere having already tipped off your age somewhere along the line. **Mobile network operator checks** The system queries your phone provider to see if you have any age restrictions on your account. No parental controls? Looks like you’re an adult. Fast, but only as reliable as the information stored at your carrier, and not an option for users on pay-as-you-go or burner numbers. **Photo-ID matching** You upload your ID and a fresh selfie. The system checks if the faces and ages match up. Classic, effective, and widely used, but you’re giving away a lot of personal information, and trusting that it’ll be kept safe. ## Privacy concerns None of these options is perfect or without risk. Many of these options have privacy implications for the user, or as a commenter told BBC News: > “Sure, I will give out my sensitive information to some random, unproven company or… I will use a VPN. Difficult choice.” A VPN is a popular option to circumvent regulations that only apply in certain countries or states. VPNs offer a secure connection when you’re using the internet and they have a variety of uses, but one is getting around blocks based on your location. Work is being done on “double-anonymity” solutions, but implementation seems to be hard. Double anonymity basically separates the information of two providers from each other. The first provider (website asking for age confirmation) will only get the requester’s age and no other information. The second provider (the age verifier) will not receive information about the service or website the age verification is needed for. In essence, the system answers only the question “Is this user of the required age?” to the site, and the third party never knows for what purpose or where this answer is used. This approach is becoming a regulatory standard in places like France to balance protecting minors online with adult users’ privacy. We feel “double-anonymity” sounds a whole lot better than “age estimation.” But the real question is whether age verification is an effective method to protect children, or is it just another threat to our privacy? Let us know your opinion in the comments. * * * **We don’t just report on privacy—we offer you the option to use it.** Privacy risks should never spread beyond a headline. Keep your online privacy yours by using Malwarebytes Privacy VPN.

The smartphone wars have a winner, and it’s Android. No, this isn’t about which device has the best camera, the snappiest processor, or the flashiest AI features—this is about which device _owners_ are safer online, and in many ways, it is Android users who take the crown. According to a new analysis from Malwarebytes, when compared to iPhone users, Android users share less of their personal information for promotional deals, more frequently use security tools, and more regularly create and manage unique passwords for their many online accounts. They also, it turns out, fall victim to fewer scams. This is the latest investigation into research conducted earlier this year by Malwarebytes that surveyed 1,300 people over the age of 18 in the US, the UK, Austria, Germany, and Switzerland. In the original report released in June, Malwarebytes revealed how mobile scams have become a part of everyday life for most everyone across the globe—and how far too many individuals have essentially given up on trying to fight back. Now, Malwarebytes can reveal how iPhone and Android users differ when scrolling, shopping, and sending messages online. This secondary analysis has controlled for age, meaning that, while iPhone users did tend skew younger in the original data set, the differences identified here are more directly attributed to device type. Here are some of the key takeaways: * Apple users are more likely to engage in risky behavior. * 47% of iPhone users purchased an item from an unknown source because it offered the best price, compared to 40% of Android users. * 41% of iPhone users sent a Direct Message (DM) on social media to a company or seller account to get a discount or discount code, compared to 33% of Android users. * Apple users take fewer precautions online. * 21% of iPhone users said they use security software on their mobile phones, compared to 29% of Android users. * 35% of iPhone users choose unique passwords for their online accounts, compared to 41% of Android users. * Apple users are more likely to be the victims of scams. * 53% of iPhone users have fallen victim to a scam compared to 48% of Android users. Importantly, the behavioral splits here are largely device agnostic. Android users are not scanning fewer QR codes and iPhone users are not failing to make unique passwords because their respective devices are somehow incapable. Instead, iPhone users are making worse decisions about buying things online and about staying safe from all types of cyberthreats—whether that includes phishing attempts, social engineering scams, or malware infections. The reasons for this are complex and hard to identify, but Malwarebytes’ original research can provide a clue. Namely, iPhone users were slightly more likely than Android users (55% compared to 50%) to agree with the following statement: > “I trust the security measures on my mobile/phone to keep me safe.” That trust could have an adverse effect, in that iPhone users do not feel the need to change their behavior when making online purchases, and they have less interest in (or may simply not know about) using additional cybersecurity measures, like antivirus. Whatever the reasons, there is room for improvement. As explained by Mark Beare, general manager of consumer business for Malwarebytes, staying safe online today cannot rely on any single platform, device, or operating system. “Devices and operating systems are just gateways to apps and websites, and it’s often those online spaces that present cyber risks,” Beare said. “When those websites or apps serve malicious or deceptive content, it’s up to the user to decide what’s real, what’s a scam, and where they should or shouldn’t click.” Here is where iPhone users should most pay attention when using the internet. ## **Unsafe shopping** It’s getting harder to shop safely online. For years, the cybersecurity industry warned people about the most obvious red flags when making a purchase or offering a donation online: Don’t click on unknown links, don’t share personal information, don’t send messages directly to strangers, and don’t scan QR codes that can lead to unknown locations. Behind all of these could lie malware, data theft, and even the slow start of a social engineering scam. And yet, in the past few years, even legitimate businesses have asked everyday consumers to do these same, reckless things. Online stores ask that people send a Direct Message (DM) on social media for a discount code, or that they sign up their email or phone number for a promotional offer, or that they complete their payment by scanning a QR code, or that they track an upcoming delivery by clicking on a link sent via text. Just because established businesses are leaning into these tactics does not make the tactics inherently safe, and unfortunately, iPhone users are pushing back the least. According to Malwarebytes’ recent analysis, 63% of iPhone users signed up their phone number for text messages so they could get a coupon, discount, free trial, or other promotional offer, compared to the 55% of Android users who did the same. Similarly, 41% of iPhone users “sent a DM on social media to a company or seller account to get a discount or discount code,” compared to 33% of Android users. Malwarebytes also found that 47% of iPhone users “purchased an item from an unknown website or supplier because it offered the best price,” compared to 40% of Android users. In looking at the data, however, it is important to recognize that some of the behavior from iPhone users has been thrust upon them. For example, 70% of iPhone users have “scanned a QR code to begin or complete a purchase.” Beginning in 2020, scanning a QR code became commonplace as restaurants across the world implemented several strategies to limit the spread of COVID-19. This practice isn’t the fault of iPhone users (or the 63% of Android users who have done the same), and they shouldn’t be “blamed” for what the world asked of them. However, sharing a phone number, sending a DM to a stranger, and buying from unknown websites are decidedly not requirements today for making an online purchase. As Malwarebytes discussed on the Lock and Code podcast earlier this year, “data deals” in which consumers are asked to give up some of their privacy for a one-time discount are rarely, if ever, worth the cost. Separately, the most common start to a romance scam, job scam, or investment scam is through a DM sent on social media. Though legitimate companies have co-opted these strategies to boost engagement and revenue, the public still have an opportunity to push back. If they do not, there is a real risk that these marketing tactics become so normalized that online scammers will find it easier to send malicious messages, disguise their intentions, and steal from innocent people. ## **Not so pro(active)** Ever since a devastatingly effective commercial was unveiled to the public some 20 years ago, there’s been a persistent belief that Apple devices are somehow impervious to viruses, malware, and all other nasty cyber infections. The marketing ploy was wrong back then and it is still wrong today—Macs get plenty of viruses—but the damage is already done, and the consequences might be most visible in how iPhone users feel about traditional cybersecurity tools: In short, they don’t use them. According to Malwarebytes’ new analysis, just 21% of iPhone users said they use security software on their mobile phone, compared to 29% of Android users. iPhone users were also less likely than Android users to use an ad blocker (19% of iPhone users compared to 27% of Android users). The data gaps here are sometimes benign. The low use of “ad blockers,” in particular, should come as no surprise. These tools are mostly understood as add-ons for desktop and laptop versions of popular web browsers—such as Google Chrome, Microsoft Edge, and Mozilla Firefox. While many mobile browsers have ad blockers built in by default, this may not be known to the average user. Also remember that, as smartphone ownership increases across the globe, so do the numbers on smartphone “dependency.” According to Pew Research Center, 15% of adults in the US _only_ have a smartphone to connect to the internet, meaning, perhaps, that 15% of people simply cannot access the same security and privacy tools that are developed predominantly for computers. That said, the justifications for iPhone users start to fade when looking at one last number. Only 35% of iPhone users “choose unique and strong passwords for accounts,” compared to 41% of Android users. Creating strong, unique passwords for online accounts is foundational to staying safe online, and it has only been made easier and more accessible over time. For users who cannot remember a unique password for every account (which is every person alive), password managers are available—some for free—to help create, store, and recall as many strong passwords as needed. For users who do not trust a third-party password manager (understandably so), Apple released its “Passwords” app on iOS 18 nearly one year ago, making password management easier by default. And for users who don’t trust password managers (of which there are many), the antiquated practice of physically writing usernames and passwords in a private journal isn’t _that_ outlandish. In short, there is little excuse for failing to create and use unique passwords for every online account, and that goes for Android users, too. The technology can be intimidating, but it’s worth the work. ## **Security for all** The measurably unsafe behavior of some iPhone users online comes with unfortunate, measurable consequences. The poor password hygiene, risky buying behavior, and limited antivirus protection are all paired with a higher overall rate of victimization—53% of iPhone users have fallen victim to a scam compared to 48% of Android users. In the worst circumstances, these disparate rates could invite blame, but it’s the wrong conclusion to make. As any scam victim knows, the statistical analysis of victimization means absolutely nothing when you are personally trying to recover your money, your reputation, your private photos, and your sense of trust in the world around you. Every person, no matter their device, should create unique passwords for individual accounts, use security products (which can also detect malicious websites and phishing schemes), and rely on friends and family when something doesn’t feel right online. And for those who want 24/7 guidance on strange messages, phone numbers, and more, there is always Malwarebytes Scam Guard to lead the way. Try it today.

Concerns about hackers accessing quantum computers are being raised by cybersecurity experts, despite post-quantum cryptography encryption.