Feed aggregator

AWS CEO Matt Garman has apparently hit out at workers unhappy with the company's enforced return to office policy.

Earlier this year, Amazon Web Service (AWS) announced staff would be required to be in-office for 5 days per week, ending its hybrid working policy. Workers were given until January 2 to return to full-time in person work.

Somewhat predictably, this didn’t go down well. Reports emerged that claimed as many as 90% of workers were unhappy with this, and 73% were considering moving jobs - which Garman has now given his blessing to.

But not in a bad way

"If there are people who just don't work well in that environment and don't want to, that's okay, there are other companies around," Garman said, adding the decision was made to allow workers to "invent, collaborate, and be connected."

But compliance to the hybrid work order was fiercely enforced, with some employees who did not adhere to the policy told they were "voluntarily resigning" and were locked out of company systems.

The policy has certainly proved unpopular with some staff who will now lose the benefits of working from home, and will spend more time commuting. But AWS want to make it clear that this is a positive change.

"By the way, I don't mean that in a bad way," he said, adding, "we want to be in an environment where we're working together."

"When we want to really, really innovate on interesting products, I have not seen an ability for us to do that when we're not in-person," said Garman.

Whilst a few major companies are bringing in return to work orders, Spotify have bucked the trend. In a recent statement, Spotify is sticking by its hybrid work policy, as Chief HR Officer said the company trusts its workers, adding that they don't want to "treat them like children."

Via Reuters

More from TechRadar Pro

• Check out our pick of the best productivity tools
• More and more workers are being forced back to the office - and they aren't happy
• Take a look at our best work from home laptop choices

While I could sum up this review in six words – like the PX13 opposite, only bigger – that doesn’t do the ProArt P16 justice. Yes, it shares many of the same design hallmarks and specs as its little brother, but this 16in laptop can justly be called a graphics workstation.

The bigger chassis means more room for fans and cooling systems than the PX13, and the CPU benefits from a maximum thermal design power (TDP) of 70W rather than 65W. Asus evidently can’t make room for a GeForce RTX 4080 within this still-slender chassis, instead sticking to the RTX 4070 found in the top-end PX13. What you do get, if you select the 2TB model, is 64GB of RAM; the 1TB/32GB version costs £2,600 inc VAT.

I tested the more expensive unit, and it was a few per cent ahead of the PX13 throughout my tests. In Cyberpunk 2077 and Dirt 5 it averaged 102fps and 122fps at 1080p High, compared to 91fps and 99fps. It was a similar story in the demanding SPECperfview 2020 viewsets: taking Catia, 3ds Max and Creo as examples, the P16 returned 67, 99 and 110, while the PX13 scored 65, 96 and 109. These scores are admittedly humbled by the towering monsters that are the best workstations.

(Image credit: Future)

Honors were even in our rundown tests, where the P16’s larger 90Wh battery evened out the greater power consumption required by its 16in panel. Here, though, a battery life of around nine hours under light use seems eminently reasonable compared to laptops with similar graphical power – usually gaming laptop machines – and I can’t imagine that many people will want to lug this 1.9kg beast around with them despite its slim dimensions. If you want to use that CPU at full power, you’ll need to find space in your bag for the 200W power supply, too.

This uses a proprietary connector, so both USB-C ports are free for connecting peripherals. The faster USB 4 port sits on the left, alongside HDMI, USB-A and 3.5mm connectors, while a slower USB-C 3.2 Gen 2 port is on the right alongside a second USB-A port and full-size SD Express 7 card reader. There’s no RJ45 port, which is a shame when there’s so much spare space, but Wi-Fi 7 offers some compensation.

Asus doesn’t squeeze in a number pad, but don’t expect a roomier keyboard than the PX13 as a result. So far as I could tell, they’re identical from the size of the keys to the half-height cursors and the single-height Enter key. That’s fine when you’re tight for space, as is the case for the PX13, but I would like separate Home, End, PgUp and PgDn keys on a laptop of this size, and for the cursors to be full height and separated out. At least Asus has increased the size of the glass-coated touchpad, which comes complete with the DialPad I describe opposite.

(Image credit: Future)

The main beneficiary of Asus’ decision not to include a number pad is the audio. Speakers sit on either side of the keyboard, with bass, depth and subtlety to match standalone speakers. Then we come to the expanse of screen, with a 3,840 x 2,400 resolution to play with – a fraction beyond 4K. It’s Pantone calibrated while an average Delta E of 0.43 speaks to its colour accuracy. Gamers should note that it peaks at 60Hz while HDR video editors won’t be impressed by a 500cd/m2 peak when viewing HDR content. Asus states a 400cd/m2 maximum for SDR, but I measured a more modest 377cd/m2.

I’ve no complaints about color coverage (100% of DCI-P3) or the level of user control, as the MyAsus app lets you pick from four colour profiles: native, sRGB, DCI-P3 and Display P3. You can also use MyAsus to play around with microphone settings, but I found the mics worked fine in the default mode. As did the 1080p webcam, although streamers might have hoped for a higher resolution.

More surprising is that Asus supplies Windows 11 Home rather than Pro, while a single-year C&R warranty is basic. Thankfully, the P16 is built like a tank, while the matte black metal chassis proved resistant to my fingerprints and scratches.

You can also access this laptop’s internals if you want to upgrade the SSD, or even add a second one in the spare slot. However, all the RAM is embedded on the motherboard, so if you want 64GB, take the hit at the time of purchase. And it is a hit; the main reason this machine doesn’t earn five stars is that I expect near-perfection for this price and, while I can forgive sacrifices in an ultraportable such as the PX13, I’m less forgiving when there’s this much space to play with. All that said, the ProArt P16 is a remarkably well-made and powerful laptop and, niggles aside, it will be a superb workhorse for years to come.

We've rated the best business computers.

San Benito has a dominant cable provider in Spectrum, but fiber is on the way, and fixed wireless options are competing to provide home internet to the area.

Modern businesses are rapidly evolving as Artificial Intelligence (AI) moves from pilot projects to large-scale implementations. This 'AI at scale' revolution is transforming industries, driving innovation, and enhancing operational efficiencies with AI being deployed at scale across various business functions.

The transition from PoCs to large-scale AI marks a pivotal moment for businesses. Initial projects showcase potential but usually fall short of full-scale impact. It is often observed when AI is implemented organization-wide, it greatly enhances efficiency, decision-making, and competitive advantage. For example, a European electronics retailer uses AI to predict online shopper conversions and prompt interventions during a browsing session. Similarly, a European publication leverages Gen AI to summarize reader sentiment from comments and reviews to score and improve their articles.

However, according to the "Global AI Adoption Index" report, European businesses have been reluctant to adopt AI due to regulatory and safety concerns. The European Court of Auditors confirmed that EU AI investments lag behind global leaders. To address this, the EU implemented the AI Act, which mandates specific AI uses and reduces administrative and financial burdens, and places compliance responsibility on AI solution providers, recognizing that many businesses may not have the AI expertise. The AI Innovation Package and the Coordinated Plan on AI also aim to boost AI investment across the EU.

As businesses in Europe find it lucrative to increasingly adopt AI technologies at scale, they can expect significant advantages in internal efficiency and innovation without a large regulatory risk.

Internal efficiency

Scaling AI improves internal efficiencies by automating low-risk tasks, allowing more focus on strategic activities. For instance, AI can manage customer service inquiries, complex supply chain logistics or credit risk scoring. A leading European bank uses AI to analyze client documents and conduct credit risk assessments, even in “low resource” languages (languages with low digital data to learn from). Mortgage officers retain decision-making authority, the AI solution an untiring assistant.

Innovation and success

Scaling AI fosters innovation and enables uncovering of insights and making smarter decisions. For example, a European consumer goods firm uses AI to boost digital engagement by analyzing search terms and product descriptions in comparison to competition on third-party marketplaces. This helps them highlight key features and use apt keywords in the description. This insight is now also informing on product development.

Successful examples of AI at scale

To grasp AI's impact at scale, let’s explore its successful applications across various business functions:

Customer service: A large European bank's contact center uses an award-winning AI solution to assist agents by suggesting resolutions, re-routing calls, predicting intentions, and recommending response language using Gen AI. Integrated with the contact center application, the solution reduces change management complexity. Now widely used, it has improved customer satisfaction scores. And thereby reduced churn.

Human resources: Gen AI usage in HR is risky because it propagates inherent bias existing in historical selections. However, with guardrails that avoid PII data feed at the entry point itself (avoiding post codes, names, genders), and style-masking through summarizing rather than the original text being fed, these risks are being mitigated. The bias removal and improved accuracy are leading to usage at scale.

Supply chain management: A leading European CPG company is implementing an advanced AI forecasting system that autonomously manages demand forecasting. It automatically groups products into demand-forecasting-units (DFUs), selects optimal algorithms, self-corrects based on recent errors, and also uses Gen AI chat to explain exceptional forecast movements for over 20,000 store-SKU combinations, all without manual customization.

Marketing: Your website is often considered your digital billboard. Large apparel retailers are leveraging Gen AI to streamline web content creation by drafting content from various data sources, such as product master data, vendor information, PDFs/PPTs, and marketing digital asset libraries, thereby reducing content creation timelines. This Gen AI-generated content reduces creation timelines and undergoes human review and approval before publication.

Overcoming hurdles and key insights

While the benefits of AI at scale are compelling, businesses must also address several challenges:

Ethical considerations: AI systems must be designed and deployed ethically to avoid biases and ensure fairness as we saw in the HR example above. The AI Act in Europe will further assist companies in adopting AI ethically by providing a regulatory framework.

Technological infrastructure: Businesses must invest in the necessary technology to handle the demands of large-scale AI applications. Responsible AI is not only about regulations but also costs. For example, if one can automatically generate marketing content, it does not mean one should do that every day.

Change management: Effective change management strategies are essential to ensure a smooth transition and gain employee buy-in. When a contact center agent gets a prompt from an AI tool instead of a previous rules-based one, this embedding reduces any pushback.

The 'AI at scale' revolution is not just a technological advancement; it is a strategic imperative. While Gen AI is expanding AI possibilities, it is also bringing risks such as black box biases, prompt injections, jailbreaks, and potential runaway infrastructure costs. Without fail safes, a Gen AI system having an outage similar to a Crowd-strike event could be much more damaging. Responsible AI is about costs, employees, society and regulations. Think and thrive responsibly!

We've featured the best AI chatbot for business.

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

The Peace Garden State has a variety of internet service providers. Check out our top picks for the best home internet in North Dakota.

The Generative AI (gen AI) boom of the last few years sparked a push to establish regulatory frameworks for AI governance. This is a natural development, as the rise of AI seems to pose issues in data privacy and protection, bias and discrimination, safety, intellectual property, and other legal areas, as well as ethics that need to be addressed.

Highlights of specific outcomes include China regulating the use of certain AI models back in 2021. We also have the AI Act of the EU, which just came into force and will become fully effective by 2026. Other jurisdictions are following up with their own measures regulating the AI ecosystem.

AI regulation in the EU vs other jurisdictions

The comparative quickness with which the EU has released a uniform regulation to govern the use of all types of AI makes it stand out. Other large jurisdictions, such as China and the US, appear to have taken different approaches. China is regulating specific areas of AI step-by-step, addressing what is recognized as risks. In 2021, they introduced regulation on recommendation algorithms, which have by then increased their capabilities in digital advertising. It was followed by regulations on deep synthesis models or, in common terms, deepfakes and content generation in 2022. And then, in 2023, regulation on generative AI models was introduced as these models were making a splash in commercial usage.

The US has not yet enacted any federal-level AI regulations. There are quickly advancing regulations at the state level, such as the California AI Act. However, it may still take some time before they take effect, even after going through all the legislative hurdles. Is regulation halting innovation?

More rigid regulatory frameworks may impose compliance costs for businesses in the AI field and stifle competitiveness and innovation. On the other hand, they bring the benefits of protecting consumers and adhering to certain ethical norms.

Businesses are pushing back, and there is still plenty of controversy surrounding the appropriateness of specific bills, such as the California bill. However, there was also pushback to the EU AI Act, which was nevertheless introduced. Thus, it is not clear whether the ongoing delay in the US is only due to lobbyism or other obstacles in the legislation enactment process. It might also be because some still see AI as a futuristic concern, not fully appreciating the extent to which it is already a legal issue of today.

The impact of AI regulation

AI regulations do not only target AI development, as they may cover several issues such as data privacy and protection, intellectual property, consumer protection, AI deployment and use, liabilities for not following the regulations, and so on. For example, the EU AI Act also covers the usage of AI in physical devices, such as elevators. Additionally, all businesses that collect data for advertisement are potentially affected as AI regulation can also cover algorithmic bias in targeted advertising.

AI has significantly impacted the web scraping industry and is likely to continue to do so. From data collection, validation, analysis, or overcoming anti-scraping measures, there is a lot of potential for AI to massively improve the efficiency, accuracy, and adaptability of web scraping operations. While it is too soon to say now, any regulation on AI may accordingly impact the above-mentioned web scraping areas that involve AI as well.

AI regulations may also spotlight certain areas of law that were always very relevant to the web scraping industry, such as privacy or copyright laws. At the end of the day, scraping content protected by such laws without proper authorization could always lead to legal issues, and now so can using AI this way.

With the incoming regulated landscape, businesses must evaluate the specific data they wish to collect with the help of a legal expert in the field when embarking on web scraping activities. As for AI governance in general, it is important to recognize that the AI legal landscape is very new and rapidly evolving, with not many precedents in place to refer to yet. Hence, continuous monitoring and adaptation of your AI usage are crucial.

We've featured the best AI website builder.

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

Microsoft has admitted it lost more than two weeks of security logs for some of its cloud products, raising potentially concerning security risks.

Microsoft reportedly users about the problem, saying the loss was not due to a security incident or an attack, but rather came as a result of a software flaw.

“A bug in one of Microsoft’s internal monitoring agents resulted in a malfunction in some of the agents when uploading log data to our internal logging platform,” Microsoft was cited saying. The malfunction took place between September 2 and September 19.

Entra, Sentinel, and others

Logs are important because they help IT teams keep track of possible intrusions and other cyberattacks, so not having this information for more than two weeks puts the users at risk.

As per the reports, the malfunction affected a couple of products: Microsoft Entra, Sentinel, Defender for Cloud, and Purview. Affected customers “may have experienced potential gaps in security related logs or events, possibly affecting customers’ ability to analyze data, detect threats, or generate security alerts,” the company said in the notification.

TechCrunch reached out to John Sheehan, a Microsoft corporate vice president, who did not share more details about the bug, but did say that Microsoft fixed it: “We have mitigated the issue by rolling back a service change. We have communicated to all impacted customers and will provide support as needed,” he told the publication.

Logs are records of events and actions generated by applications or systems. They are used for debugging issues, monitoring performance, and auditing security. By capturing information about the system's operation, logs help developers troubleshoot problems, track system health, and identify potential security threats. That makes them a crucial tool in spotting and tackling cyberattacks.

Via TechCrunch

More from TechRadar Pro

• The difference between no-logs and zero-logs VPNs – and why it matters
• Here's a list of the best firewalls today
• These are the best endpoint protection tools right now

See what our CNET sleep experts think of Helix's latest mattress, the Helix Plus.

A critical vulnerability in the Kubernetes Image Builder has been detected allowing threat actors to access different Virtual Machine (VM) images with ease. A patch is already available, so if you’re using the image building tool, make sure to update it to the latest version as soon as possible.

Kubernetes Image Builder is a tool that helps build and maintain container images for Kubernetes environments. It simplifies the building, packaging, and deployment of containerized applications by generating optimized and reproducible images ready for Kubernetes clusters.

However, when one builds a Kubernetes VM image, it comes with a set of default credentials, which are the same for every user. As a result, crooks can easily access virtual machines with root privileges.

Randomly generated password

According to The Register, VM images built with the Proxmox provider are most at risk. The flaw on this platform is tracked as CVE-2024-9486, and carries a severity rating of 9.8/10, meaning it’s critical. Image Builder version 0.1.37, or earlier, are flawed, and it is recommended users migrate to Image Builder v0.1.38, or later, as soon as possible.

In this version, every new image build will be given a randomly generated password, with the builder account being terminated at the end of the build process.

Users that end up upgrading Image Builder should also re-deploy new images to any affected VMs, the publication stressed.

Besides Proxmox, there are other providers who are at risk, too - including Nutanix, OVA, QEMU, and others: However, in these instances, the severity rating is 6.3, since they disable the default credentials at the end of the image build process, and thus give the threat actor a much smaller window of opportunity.

Those that are unable to apply the patch at the moment should disable the builder account and thus mitigate the risk.

Via The Register

More from TechRadar Pro

• AWS has patched a rather embarrassing Kubernetes bug
• Here's a list of the best firewalls today
• These are the best endpoint protection tools right now

What was once one of the best Netflix movies is now about to become one of the new best Hulu movies in November 2024. Damien Chazelle's almost-best picture Oscar winner La La Land (2016) is making the switch from one of the best streaming services to the other, meaning you'll be able to stream it on Hulu from November 1 to witness the heart-wrenching Emma Stone and Ryan Gosling love story over and over again at your convenience.

Out of all the new titles new on Hulu in November 2024, La La Land is just one of the few movies that will be leaving Netflix in October 2024 for a new spot in Hulu's November schedule. Additionally, Hellboy (2019) and Sonic the Hedgehog (2020) are set to leave Netflix on October 31 and, like La La Land, will become part of Hulu's reshuffled catalog of movies on November 1. While we'd wish that they'd have a Rotten Tomatoes score as respectable as La La Land's 91% (Sonic the Hedgehog has 60%, and Hellboy a catastrophic 17%), I'm kind of living for the subtle shade on Hulu's part.

However that's not to say that Netflix won't see some of the best Hulu movies join its catalog, as it has everything in it to snag some of the titles that are leaving Hulu in November 2024. Netflix's schedule for next month has yet to be unveiled, and while we don't know what's coming to Netflix in November 2024 just yet, I can imagine that the platform will be preparing to welcome the holiday season – just as Hulu has planned.

Part of me has my fingers crossed behind my back in hopes that Netflix will recycle Hulu's rejected titles next month. But I have my eyes solely on Hustlers (2019) because, as I've mentioned in our roundup of titles leaving Hulu, Hustlers is one of my comfort movies and one of my favourites from that year. Bless the days when it was available on Prime Video.

You might also like

• Netflix teases Virgin River season 6's wedding of the year and I desperately need an invite
• Netflix’s Sweet Bobby: My Catfish Nightmare isn’t the only catfishing documentary worth streaming this week
• Hulu vs Netflix: which streaming platform do you choose in 2024?

We’ve recently seen the release of a major Windows 11 feature update, version 24H2, and now developer NTDEV, responsible for the development of projects like “Tiny11” and “Tiny10” (smaller slimmed down versions of Windows 11 and 10) has released an impressively lightweight, compact version of 24H2. NDTEV has been highlighted by publications for its efforts to make versions of Windows that use minimal resources on your device - both in terms of storage and system memory, which allows the operating system to run on older or underpowered machines.

Following 24H2’s release, Microsoft has announced details about Windows 11 Enterprise LTSC 2024 (also known as Windows 11 LTSC 24H2), a special version of the operating system (OS) designed for long-term use by businesses and organizations that don’t receive updates as frequently as other kinds of users. NDTEV’s newest update has modified this version of Windows to shrink it down to Tiny11 form.

NDTEV provided screenshots of their feat on X, claiming that you could make the Tiny11 version based on Windows 11 LTSC 24H2 fit on a DVD disc (if you remember those!). For context, a DVD can store about 4.7 GB of data, and NTDEV’s Tiny11 version only takes up 3.54 GB — leaving some extra space on the disc.

The team explains that they were able to achieve this using a method called LZX delta compression. Compression is a common computing process that makes files smaller while preserving essential information, and LZX is a method that relies on advanced compression techniques built into Windows 11’s NTFS file system.

Well ,the good news is that you can still make the entire tiny11 fit on a DVD! (with LZX compression, of course).Behold, tiny11 based on Windows 11 LTSC 24H2! pic.twitter.com/tXS3Z8KoSwOctober 12, 2024

Food for thought about hardware requirements

Versions of Windows like Tiny11 have become a favorite among some users because aside from shrinking it down to a size that can fit on a DVD, it also needs very little system memory (RAM) to run, leaving the rest of the device’s memory free for other tasks, which can lead to big performance boosts, even with underpowered devices. Previously, Neowin had reported about NDTEV demonstrating that they can run Tiny11 with just 176MB of RAM, a magnitude less than the amount of RAM required for the officially versions of Windows 11 (which is 4 GB).

In another impressive move, NDTEV made a “text-only” version of Windows 11 that used just 100MB of RAM by stripping away the OS’s graphical interface (the visuals you see that makes it easier to use, such as the Start menu and even the iconic windows that gave the operating system its name).

If you’re a little more confident when it comes to tinkering with your devices, you could try installing Tiny11 on your PC, but even if you’d rather stick with full-fat Windows 11, you can still appreciate NDTEV’s accomplishments. They continue to demonstrate the very real potential to run a modern OS with minimal hardware requirements, and Tiny11’s modest hardware requirements allow older and less powerful machines to run modern operating systems like Windows 10 and 11, which they might otherwise struggle with.

As hardware performance demands continue to grow, projects like these show both the initiative and creativity of developers like NDTEV, as well as how efficient software design can extend the lifespan of devices. In a world where we are having to begin reckoning with the environmental cost of our consumer demands, it’s a hopeful glimpse of how we can navigate it while continuing to innovate our technologies.

YOU MIGHT ALSO LIKE...

• Tiny11 is now even smaller, giving you Windows 11 23H2 but without the clutter
• Ditch Microsoft’s data-gathering and take control of your Windows 11 experience with the new Tiny11 Builder tool
• Windows 11 on an iPhone 15? Tiny11 developer gets cats and dogs living together!

Elon Musk’s latest changes for X are driving more users away – not exactly a surprise, granted – and many of them are flocking to rival social media outlet Bluesky. So many made the switch, in fact, it led to Bluesky briefly going down due to the volume of incoming new users.

The central move initiated by X that made the headlines for driving migration away from Musk’s platform is a change to the way the ‘Block’ button works. This was actually announced back in September, but is officially being implemented now (well, it’ll be in place ‘soon’ we’re told).

It means that going forward, X users who you have blocked will still be able to view your (public) posts – though they won’t be able to engage with them in any way (from replies to liking and so forth).

This is problematic for obvious reasons, in terms of enabling stalkers and trolls who will still be able to view the posts of an account that has blocked them, when previously this wasn’t the case. In the past, blocking meant that the blocked user couldn’t see any posts (or anything at all, save for a message telling them that they’ve been blocked), but soon, this will change.

As The Independent reports, this confirmation of the move to relax what blocking means led to a flood of departures from X to Bluesky, a decentralized take on X that was the brainchild of former co-founder and CEO of Twitter, Jack Dorsey.

Bluesky posted to say it had in excess of 100,000 new users inside 12 hours following the announcement by X, after the rival network highlighted the fact that its block function stops those who are blocked from viewing any posts.

Such was the volume of disgruntled defectors that Bluesky actually went down for a couple of hours, with those fleeing X noting that they couldn’t sign up (and for that matter, users couldn’t even login). Down Detector recorded a server outage lasting for a couple of hours, though this has now calmed down to a few scattered reports, and normal service appears to have resumed for Bluesky.

In an update, Bluesky noted that it has now gained half a million new users in the past day.

There’s another reason that some folks are rapidly exiting from X stage left (and right, and indeed center, clambering over the audience, it would seem), and that’s a change to X’s privacy policy.

As TechCrunch reports, the new policy includes an update that allows third-party collaborators to use content on X to train their AI models – unless the user opts out. This is a notable extension of the reach of AI training on X, which has so far only been used to train Musk’s own Grok AI (unless users opt out, again).

(Image credit: Getty Images) Analysis: You mystify me

There’s been a strong reaction to both moves, particularly the blocking change.

Let’s start with the AI side of the equation, though, and what’s going on there. Artists are not at all happy about the possibility of their content being used in AI training – for which they read as being ‘stolen’ effectively – but there is a choice to opt out (although we don’t yet know where that option will lie).

Still, clearly many folks feel this is an ominous switch in direction for X, and are voting with their feet to make a point. Certainly, it should be an opt in rather than opt out decision, but often this is a concern glossed over in the tech industry for obvious reasons.

To be fair to Musk, X isn’t the only social media platform heading down this path, of course – Reddit is at the head of the pack in terms of monetizing in the AI sphere.

The switch in blocking policy is a truly mystifying one, though, and seemingly part of Musk’s self-destructive side when it comes to his social media property. We can’t think of a good reason why this has been enacted, save for… nope. We just can’t think of one.

For X’s part, the main argument in favor of the change is that “block can be used by users to share and hide harmful or private information about those they’ve blocked,” and so the ability to still view the posts of a blocked account means greater transparency in terms of detecting such instances.

But that does not square up at all, not even remotely, with wider and far more pressing concerns about blocked people now being able to act in harmful ways towards those who have blocked them.

Diluting blocking in this way is undeniably going to make life better for stalkers, unhinged trolls and folks who might harass other denizens of X for kicks, and as many have rightly pointed out, there are genuine online safety concerns here.

While a blocked person may not be able to repost in theory, there is nothing to stop them from taking a screenshot of a post, and sharing that on X (or elsewhere). Or more broadly, there’s nothing preventing the blocked person from using info from the visible posts of the person they might be harassing or stalking.

Surely when you weigh up these two factors – the rights of the blocked, and rights of the blocker – on the scales of worry, the tipping that’s going on is severe.

Another reason raised for doing away with full-blocking is that it’s easy to get around by creating another dummy account – but not everyone will go to those lengths (far from it, we’d imagine), and so this is only a partial point of argument at best.

All in all, at the very least, isn’t some sort of compromise called for here? Perhaps in the form of some sort of full block option that could be left in.

The reaction to the move has been entirely negative, and hopefully, X and Musk are taking that on board.

Bluesky is certainly capitalizing on this, not just reinforcing that it offers full blocking capabilities as mentioned, but also more besides (the ability to “subscribe to block lists maintained by users you trust” for example).

Currently, Bluesky has half a million new recruits for its platform, as mentioned, and it’s currently trending on X with over 200K posts. The talk is now turning to how the new philosophy on blocking might end up with the X app being blocked itself from Google and Apple’s stores.

You might also like...

• Jack Dorsey's Bluesky is like Twitter without Musk – and that's good
• Elon Musk's year-long quest to destroy Twitter is the most successful inside job in tech history
• How to use ChatGPT – everything beginners need to know about the AI chatbot

Asus has announced its new ROG Thor III series PSUs that have been built to meet the demands of upcoming power-hungry components like Nvidia’s expected RTX 5090 GPUs, which are rumored to consume up to 600W of power. With PCIe 5.0 connectors, ATX 3.0 compliance, and wattage ranging between 1000 and 1600, it’s clear the Thor III series should be more than up to the task of powering the next generation of computer components.

A standout feature is Asus’ “GPU-FIRST” voltage-sensing technology. As the name suggests, it means the PSU will prioritize the GPU over the CPU. This seems sensible, considering it’s the new GPUs that may eat up the watts. The feature is paired with a patented “intelligent voltage stabilizer”, which Asus says enhances voltage delivery by up to 45%.

The Thor III series promises high energy efficiency and minimal power loss, with the 1000W and 1200W  models being  rated at 80 Plus at Platinum, and the 1600 model is rated at Titanium. It also boasts Lambda A+ acoustic ratings (the highest) – which should hopefully mean that the PSUs are near silent even when under load The PSUs are cooled  by a 135 mm axial-tech fan, and they also feature a 0dB mode, which optimizes cooling without unnecessary noise. If you hate the sound of gaming PCs whirring up their fans when the action heats up, the ROG Thor III series PSUs could be a great investment.

For easy, smoother installation, the ROG Thor series III includes native 12VHPWR connectors, so you won’t need to worry about adapters or keeping your build neat and tidy on the inside. The PSU comes with an integrated OLED display, with which you can monitor power usage in real-time. The PSU also has ARGB lighting, and you can customize it to your heart’s content through Aura Sync, ASUS’ lighting control software.

Following leaks of Asus’ flagship ROG Maximus Z890 Extreme motherboards (which will cost upwards of $1,000), designed to target Intel’s flagship Arrow Lake CPUs and Nvidia’s RTX 5090 GPU, it was obvious that ASUS would release an equally powerful PSU to compliment these builds. Judging by its specs, the Thor III series delivers.

PSU prices and lightning, very very frightening

How much this will set you back we don’t yet know, but what we do know is that it won’t be great for your energy bills, as these are designed to pump out the kinds of wattage that would impress the God of Thunder himself.

While that’s great for keeping powerful components running at their full strength, it could lead to a spike in your energy consumption, and at a time when many people’s bills are already rising, you’ll need to think carefully about spending money when they release, which should be soon.

You might also like

• What is a PSU? We explain all
• Nvidia’s RTX 5090 now rumored to have superfast clock speeds – as well as being super-slim – could this GPU be too good to be true?
• Top gaming PCs: great rigs for serious PC gaming

At nearly twice the size of some competitors, Elehear’s new hearing aids certainly make a statement.

Permira has confirmed the completion of its acquisition of top website builder platform Squarespace in an all-cash deal valued at $7.2 billion.

As part of the deal, Squarespace will be taken private, meaning it will no longer be listed on the New York Stock Exchange (NYSE).

The buyout, initially announced in early summer of 2024, ended up costing Permira $300 million more than the $6.9 billion it had initially set out to pay.

Squarespace is now a private company

Although members of the public can no longer buy shares in the company, Squarespace founder and CEO Anthony Casalena has retained a substantial majority of his equity, remaining one of the largest shareholders.

Casalena will continue to lead the company as CEO and Board Chairman, with the company’s other leadership members also remaining in place, confirmed Permira.

"We are excited to embark on a new chapter with Permira, one focused on our long-term strategy and commitment to serving entrepreneurs globally," Casalena added.

Permira Partner David Erlong said the investment company’s decision to back Squarespace recognizes the company’s strong relationships with small businesses and entrepreneurs, making a vote of confidence in the platform.

“We look forward to leveraging our decades-long experience and track record backing internet platforms and technologies that enable SMBs, as we support Anthony and his team in growing Squarespace and expanding its product suite through its next chapter of success.”

The deal, valued at $7.2 billion, equates to $44.50 per share, up from the $44 per share that was initially suggested and subsequently rejected by shareholders under the advice of Institutional Shareholder Services on the basis that the valuation undervalued the company.

More from TechRadar Pro

• Cut the costs by using the best free website builders
• We’ve rounded up a list of all the best web hosting providers
• Squarespace says it will go private after $6.9bn Permira deal

Portland residents still have several ISP options available, from speedy fiber internet to more affordable broadband selections.

The company behind Trump Watches prominently features an iconic image of the presidential candidate on its timepieces. There’s one big problem: It’s not allowed to.

If you're looking for more control over your lights, then this deal is for you.

When Sonos unveiled the Sonos Arc Ultra soundbar earlier this week, we had a few questions. And now we have the answers.

In particular, we've confirmation about the Arc Ultra's Dolby Atmos channels, its current stance on DTS, and its HDMI ports – and we've had an update on the status of the app.

The first questions was around its channel count, which is officially given as 9.1.4. The last number there is the relevant part – it means four height channels for Dolby Atmos. However, the Arc Ultra's specs say it has two upfiring drivers, so we wanted to get Sonos' official line on how two turns into four.

Sonos confirmed that are indeed two up-firing drivers, but the way they work with the rest of the soundbar makes them act like there are actually four upwards firing speakers – at least in theory. "The rear height arrays utilize a very strategic combination of the upfiring, outward firing and forward firing transducers to position it deeper in the room than the front height channels," the company said – so basically, it'll offer virtual rear height channels. However effective this is remains to be seen.

Checking the disappointments

There were two potential disappointments about the Sonos Arc Ultra, but that weren't explicit from Sonos' information and specs, so we also asked to confirm that a) it only has one HDMI eARC port, and no passthrough, and b) that there is still no DTS support.

In both cases, what we expect was confirmed. Sonos' long-standing line on DTS hasn't changed – it will consider adding new formats in the future, but nothing is planned – and there is just a single HDMI port.

That's very unusual among the best soundbars now: all the major premium competition considers at least one HDMI passthrough as the bare minimum. A lot of people today need more flexibility in their connection, because they have a high number of boxes to connect to their TV. The Samsung HW-Q990D soundbar actually adds to the number of ports you have available. The Sonos Arc Ultra leaves you with one fewer.

How's the app going?

According to CEO Patrick Spence, the app is 90% back to normal since its troubled update. "We've worked tirelessly to address the most critical app issues and reintroduce the most requested legacy features. We've now reached a level of quality that gives us the confidence to release our exciting new products," he said in a statement. "As outlined in our commitments, we will continue to improve the software regularly and are determined to make the Sonos experience better than ever."

Sonos continues to provide app update timescales on its support page; the to-do list for late October includes:

• Improving system setup and reliability of adding new products
• Improved volume control & responsiveness (iOS)
• Music library performance improvements (iOS)
• Improving overall system stability and error handling
• Support for Arc Ultra and Sub 4

And the mid-November update will deliver:

• Playlist editing
• Support for Android users with multiple homes
• User Interface improvements (based on your feedback)
• Improved music playback error handling

Those plans are for both the iOS and Android versions of the Sonos app. The Android app will also get "improved volume control & responsiveness" in a December update.

You might also like

• 7 movies with stunning Dolby Atmos music and soundtracks to make your home theater sing
• I didn't care for Dolby Atmos music, but this one mind-blowing experience made me a believer
• I experienced Dolby's reference 34-speaker Dolby Atmos theater – and it made me appreciate the importance of movie theaters

Nothing's better than a crispy fry, but you can only achieve them in an air fryer by following these rules.