Technology

• Tenable urges users to update their Nessus instances to avoid a potential plugin security issue
• A previous plugin update saw agents going offline
• The earliest clean version is 10.8.2, so users should update now

Tenable has urged users to update their Nessus instances to avoid a potential plugin security issue.

Tenable Nessus is a widely used vulnerability scanner that helps identify and assess security vulnerabilities, misconfigurations, and compliance issues in networks, applications, and systems.

However, in the final hours of December 2024, the company said it was “aware of and actively investigating” an issue with Nessus agents going offline after plugin updates for certain users on all sites - and as a result, the company temporarily stopped plugin updates.

Resetting plugins

The incident apparently affected Nessus Agent versions 10.8.0 and 10.8.1, for users in North and Latin America, Europe, and Asia. To address the issue, Tenable released Nessus Agent version 10.8.2.

"There is a known issue which can cause Tenable Nessus Agent 10.8.0 and 10.8.1 to go offline when a differential plugin update is triggered. To prevent such an issue, Tenable has disabled plugin feed updates for these two agent versions. Additionally, Tenable has disabled the 10.8.0 and 10.8.1 versions to prevent further issues," the release notes detailed.

Now, users are called to either upgrade to 10.8.2, or downgrade to 10.7.3 to bring their Nessus agents online. However, they also need to reset their plugins.

“If you are using agent profiles for agent upgrades or downgrades, you must perform a separate plugin reset to recover any offline agents," the company concluded.

To adress the bugs, users first need to reset agent plugins via a script or a nessuscli reset command, and then manually upgrade the Tenable Nessus Agent using the 10.8.2 install package.

Tenable claims to have more 44,000 customers worldwide, including 65% of the Fortune 500. While the exact number of Nessus users isn't publicly disclosed, it is safe to assume that Nessus is quite popular in the cybersecurity community.

Via BleepingComputer

You might also like

• Fluent Bit vulnerability threatens almost all popular cloud platforms
• Here's a list of the best antivirus tools on offer
• These are the best endpoint protection tools right now

Shifting your sleep schedule can be tough, but using these nine tips will make sure the habit sticks.

• 4K at up to 100 inches from a shoebox-sized projector
• Its sibling, the N3 Ultra Max, delivers 4K at up to 180 inches
• Not available until late 2025

Fancy a tiny ultra-short-throw projector that can project a 100-inch image while sitting under six inches from the wall? Then you might want to check out the new O2S Ultra from JMGO. It's one of two advanced projectors the firm is unveiling at CES, and it claims to be the world's smallest 'laser TV' (which basically means an ultra-short throw projector that has TV streaming and/or tuning tech built-in, often supplied with a matching screen).

The specs are impressive for such a small device. It delivers an estimated 2,500 ISO Lumens – the brightness figures haven't been certified yet – with 4K resolution from a box just 12.3 x 11.4 x 5.5 inches in size, so it's not going to take up your entire room or dislocate your shoulder if you're moving it around.

It has a throw ratio of 0.18:1, enabling it to deliver 100 diagonal inches from just 5.75 inches away, and it has 110% of the BT.2020 color gamut. The full on/full off contrast ratio is 1,800:1 and there's Google TV built in for streaming smarts.

The bad news? You'll have to wait for it. The O2S Ultra is currently planned for a global launch in the final quarter of 2025. Pricing has yet to be disclosed, but if JMGO can get it right, this definitely has potential to bother some of the options on our list of the best ultra short throw projectors.

The shoebox-sized O2S Ultra can project 4K at up to 100 inches from less than six inches away (Image credit: JMGO) This new projector takes it to the Max

The firm's second new projector, the N3 Ultra Max, uses the same MALC 3.0 technology as the OS2 Ultra. According to JMGO it "achieves the elusive balance of brightness, contrast, and color accuracy – a feat unmatched in the market."

More specifically, the N3 Ultra Max delivers 4,000 ISO Lumens, a 5,000:1 contrast ratio and an AI-powered gimbal. That enables you to point the remote control at any spot on a wall and the projector will automatically align and adjust itself. JMGO calls this "point and play".

The N3 Ultra Max delivers 4K resolution at up to 180 inches, and it too comes with Google TV on board. Like its sibling, it's due to launch in the final quarter of 2025. Are we looking at two of the best projectors of next year? JMGO has been making really good-value products for a while, so if these can nail the price-to-performance balance, it's very possible.

Read more from CES 2025

TechRadar will be extensively covering this year's CES, and will bring you all of the big announcements as they happen. Head over to our CES 2025 news page for the latest stories and our hands-on verdicts on everything from 8K TVs and foldable displays to new phones, laptops, smart home gadgets, and the latest in AI.

And don’t forget to follow us on TikTok and WhatsApp for the latest from the CES show floor!

• The best 4K projectors for home theater heaven
• The best portable projector, according to our experts
• The best projector sales and deals

• Brisk It annouces grilling AI update
• Vera 2.0 can understand images
• And it annoucnes a new more affordable AI grill

Generative AI-enhanced gadgets were all the rage in 2024, but something that stood out from last year’s more run of the mill announcements was Brisk It's Origin grills and its Vera AI which promised to help you grill food perfectly. Now, at CES 2025, Brisk It is back with new hardware and an AI update.

To catch you up on Vera 1.0, Brisk It explained previously its AI is ideal for helping you plan meals – by asking it for meal advice based on what you want to eat, your allergies, and what you have in the fridge – and it can also take control of your compatible grill. You simply tell the AI what you’re cooking and it’ll handle the temperature.

Now Vera 2.0 is set to deliver a serious update, thanks to image recognition. Instead of simply listing ingredients you can snap a picture of them, or snap a picture of a dish or recipe, and Vera can help you plan meals based on the images you submit. It also promises to be a better grilling companion.

(Image credit: Brisk It) AI grilling gets more accessible

The big downside of Vera, however, is it's only compatible with Brisk It’s AI grills which aren’t cheap. The Origin 940 is usually $1,099.99 and the Origin 580 is $849.99 – though we have seen them on sale recently for $200-off and $150-off respectively.

Even then that’s still pretty pricey, but with Brisk It's new Zelos-450 smart grill you should be able to try AI grilling out at “a much more accessible price” – at the time of writing we don’t have a precise price for this grill, but we’ll update this article when we know more.

The Zelos-450 doesn’t just boast AI assistance – which promises to make real time cooking adjustments if you make mistakes, and “reliable, hands-free grilling from start to finish” – but also the same wood-chip grilling experience as other Brisk It grills for a wood-fired flavor finish on your food. In addition it offers 450-square inches of cooking space, and cooking temperatures of 180F to 500F.

We’re hoping to test the new Zelos-450 smart grill out for ourselves soon. We’ve heard good things about Brisk It’s other AI grills so we’re excited to see this one in action, and on that note we’re planning to demo the tech at the company’s CES booth – so look out for our thoughts on that demo when we publish them and be sure to follow our Tiktok which will have the latest tech from CES.

You might also like

• The best smart home devices 2025
• I tried using Alexa and smart home devices to help fix my routine
• Amazon Echo vs Apple HomePod vs Google Nest

This year’s crop of robot vacuums can climb stairs and pick up dirty socks that are in their way. Soon they will be able to play with your cat. Good luck to us all.

We take a first look inside the Flow electric travel trailer, which the company says will ship in the first half of 2025. And yes, you can tow it with your Cybertruck.

• The SwitchBot Lock Pro is compatible with 99.9% of front door locks
• It can be opened in 16 different ways, and supports the Matter standard
• SwitchBot has also launched a smart doorbell designed for seniors

SwitchBot has unveiled a new smart lock that can be retrofitted to just about any door without replacing your existing hardware, and can be unlocked in 16 different ways. The SwitchBot Lock Ultra is a successor to the SwitchBot Lock Pro, and is compatible with even more types of locks – great news if you rent and don't want to incur the wrath of your landlord.

In the unlikely event that the SwitchBot Lock Ultra doesn't play nicely with your existing door hardware (SwitchBot claims it should work fine with 99.9% of locks), the company can help you out by 3D-printing a custom solution for you.

Once it's fitted, you can unlock it using your fingerprint, a smartphone app, an NFC card, or a voice command. It's even Matter-compatible if you also pick up one of SwitchBot's Matter hubs, letting it work as part of a mix-and-match smart home ecosystem.

Who's there?

SwitchBot is showing off the new lock at CES 2025 in Las Vegas, alongside a set of new robot vacuums and its first video doorbell. The device, simply named the SwitchBot Video Doorbell, is designed with senior users and children in mind, and includes a portable display with a 4.3in display so you don't need a smartphone to see who's at the door and speak to visitors.

This seems like an excellent idea that solves a problem with most 'accessible' video doorbells. For example, the Doro Hemma is an excellent doorbell that's designed with older homeowners in mind, but requires a mobile app, which excludes anyone who uses a simple feature phone (or doesn't have a smartphone at all).

The SwitchBot Video Doorbell comes with a portable display so you don't need a smartphone to see who's at the door (Image credit: SwitchBot)

The SwitchBot Video Doorbell has an extra loud 100dB chime (second-generation Ring doorbells typically max out at 80dB) to make it easier to hear, and can be mounted with either screws or adhesive. According to SwitchBot, it boasts a very impressive 20-month battery life (most wireless video doorbells need to be charged at least twice a month), and supports up to 512GB local storage (sold separately).

We're hoping to put both devices to the test soon, so we can see how they compare with the best smart locks and best video doorbells from big brands like Philips and Ring.

We’re covering all of the latest CES news from the show as it happens. Stick with us for the big stories on everything from 8K TVs and foldable displays to new phones, laptops, smart home gadgets, and the latest in AI.

And don’t forget to follow us on TikTok for the latest from the CES show floor!

You might also like

• Smart locks: everything you need to know about intelligent security
• 7 things I learned after getting a home security camera
• Are smart home security systems more of a security risk than a benefit?

LG has revealed its 2025 OLED TV lineup and among the models announced is the LG G5, one of its flagship 4K OLED models. The LG G5 is successor to 2024’s LG G4, one of the best OLED TVs of last year, and looks to provide several upgrades over the G4 – but there’s one upgrade suggested by LG that’s caught my eye in particular, and it’s not one of the biggest and flashiest features.

First, let’s talk about the flashy features. The LG G5 will be the world’s first AMD FreeSync and Nvidia G-Sync certified TV at 165Hz, a step up over the G4’s top 144Hz refresh rate, which will put the G5 in new territory compared to the best gaming TVs (although 144Hz and above is only relevant for PC gamers). It’s worth noting though that the G5 won’t support Dolby Vision and 165Hz at the same time.

The G5 will feature LG's new Brightness Booster Ultimate tech, which also appears in the LG M5, the high-end wireless 4K OLED TV in LG’s 2025 TV lineup, and LG says the G5 (and M5) are up to three times brighter than the entry-level B-series of OLED TVs (but didn't offer any more comparisons to more directly comparable sets).

One interesting absence this year though is the use of micro-lens-array (MLA) tech within LG's OLED range (including the G5), which helped provide the brightness upgrade over W-OLED panels in previous models, such as the LG G4 from 2024. Yet LG still claims the G5 will be the brightest the G series has been, suggesting this new Brightness Booster tech and a recently revealed 'four stack' OLED panel will be doing a lot of the leg work.

2024’s LG G4 hit 1,489 nits peak brightness (measured on a 10% HDR window) compared to the LG B4’s 656 nits, so hopefully we’ll see a big brightness boost if it's going to get close to being fully three times brighter.

However, while this all sounds great, it’s actually another area of brightness that makes the G5 an exciting prospect, and the one I’m most looking forward to seeing in action – and that’s its fullscreen brightness. LG claims that the G5 will have up to 40% higher fullscreen brightness than its predecessor, the G4, and that could be a gamechanger.

Fullscreen brightness in a nutshell

While not the flashiest feature to talk about, the LG G5 will hopefully have noticeably higher fullscreen brightness than its predecessor, the LG G4 (pictured). (Image credit: Future)

Brands often quote peak brightness numbers when discussing a TV’s brightness levels, with the likes of Hisense and TCL, which make some of the best mini-LED TVs on the market, quoting numbers up to 10,000 nits. While OLEDs are a ways off this number, there are signs that OLEDs could start to hit 3,000 nits, with Philips among the first to say its sets can hit this figure.

Measuring peak brightness, however, is done by shining a bright light in a very small section of the screen, between 2-10% of the screen’s total area. But when it comes to watching TVs, fullscreen brightness (ie, how bright the whole thing can be, uniformly) is just as important.

Fullscreen brightness is important for many types of movies and TV shows, but the really big one is sport, where you want to see clarity and rich colors across the whole screen at the same time, to get the full viewing experience. Just as importantly, fullscreen brightness is vital in helping combat the reflections you might see in brighter viewing environments – say a room filled with a lot of natural light – because the brighter the image, the more likely it is to overwhelm any reflections, so you can focus just on what you’re supposed to be watching.

Fullscreen brightness has never been OLED’s strong suit, often hitting between 200-300 nits on the most flagship of OLEDs, compared with the 600-800 nits you’ll get on a high-end mini-LED. It takes inventive measures like the Samsung S95D’s, 2024’s TechRadar Choice Awards winnner for TV of the Year, OLED Glare Free anti-reflection tech to effectively beat reflections, and that’s just using a matte screen rather than increasing brightness.

The LG G5 – the ‘brightest’ OLED yet?

While the LG G5 won't hit the brightness levels of a mini-LED like the TCL QM851G (pictured), reports suggest we could be seeing the brightest OLED models yet. (Image credit: Future)

When we measured the LG G4’s fullscreen brightness, it yielded a result of 257 nits in Filmmaker mode, which is impressive for an OLED, but was still far less than the 318 nits achieved by the QD-OLED Samsung S95D. And both models were significantly dimmer than the likes of the Hisense U8N and Samsung QN90D, mini-LED sets that achieved results of 805 and 659 nits in Filmmaker mode respectively.

While OLED is unlikely to hit fullscreen brightness at mini-LED level anytime soon, a 40% fullscreen brightness increase in the G5 is nonetheless intriguing. If correct, we should expect the LG G5 to hit over 350 nits in Filmmaker Mode, and while that’s still not mini-LED levels, it’s a significant increase from mid-range OLEDs and flagship OLEDs from previous years, and should be very visible to the naked eye.

It’s worth noting that this is also estimated based on figures for Filmmaker Mode, one of the dimmer preset picture modes available, but one we like here at TechRadar because of its accuracy. Brighter picture modes such as Standard and Cinema are likely to be even brighter still, and I find that during my own testing, these can be essential under our bright testing room lights.

We were highly impressed with the LG G4’s overall picture quality, awarding it five out of five in our LG G4 review and while there were reflections present, overhead lighting wasn’t much of an issue. So if the LG G5 can improve on this even further with higher overall brightness, it could add itself to the rare group of OLEDs that can handle any brighter rooms, making it a more versatile set overall.

Final thoughts

In the past couple of years, LG hasn’t been the top dog in the OLED world in our testing; Samsung has. While LG provides more OLED options, Samsung’s models were so good, we chose the Samsung S90C and Samsung S95D as our TV of the Year in 2023 and 2024 respectively. Brightness was a key factor in both these cases, as both models pushed OLED brightness levels to new heights that we couldn’t ignore. Crucially, however, this didn’t mean we didn’t like LG’s OLEDs any less – they just got pipped to the post.

If LG really can push the G5 to a brightness we’ve not seen before, especially fullscreen, could we see LG regain its OLED crown in 2025. I for one am looking forward to getting my hands on the G5 to see if it can innovate OLED even further, and the first thing I’ll be doing is putting on some lights to see if its newfound brightness is the real deal.

You might also like

• LG will bring brighter OLED TVs to CES 2025, thanks to adding 33% more OLED
• LG G5 OLED TV: what we want to see
• New mini-LED and OLED TVs were brighter than ever in 2024 – but how bright do they need to be?

TechRadar will be extensively covering this year's CES, and will bring you all of the big announcements as they happen. Head over to our CES 2025 news page for the latest stories and our hands-on verdicts on everything from 8K TVs and foldable displays to new phones, laptops, smart home gadgets, and the latest in AI.

And don’t forget to follow us on TikTok and WhatsApp for the latest from the CES show floor!

• Google announced the next-generation of Google TV at CES 2025.
• It's getting a hefty infusion of Gemini to up its smarts and make it better at conversational quieres.
• Plus with proximity sensors it has a unique take on 'Ambient Mode.'

The Google TV platform powers not only the Google TV Streamer box, but also countless TVs that opt for it to be the native operating system of choice. As the successor to Android TV, it ushered in a cleaner interface with better recommendations for major streaming services – as well as support for more of them – and a hefty dose of voice control courtesy of ‘Hey Google.’

Now, though, much like many of the other announcements at the 2025 Consumer Electronics Show, Google’s giving us a peak at the next generation of Google TV and it, of course, has an AI tie-in. Just like the latest Pixel phones, next-gen Google TV will feature Gemini, and quite critically here, enables much more natural conversation requests.

Further, when next-generation Google TV devices are expected to arrive by the end of 2025, it will be a mix of software and some hardware to enable an ambient experience that almost feels self-aware. Thanks to proximity sensors in forthcoming TVs, it can automatically turn the TV on a photo gallery slideshow as you enter the room, but then switch to a dashboard of sorts as you get closer. Pretty cool, right?

This is Google TV’s take on the ambient experience for sure, one that mimics the Nest Hub and Nest Hub Max or even the latest Nest Thermostat as it adjusts the content being shown based on where you will be in the space. The photo gallery aspect, including an opt-in for images created through Generative AI, is reminiscent of both the Chromecast static pages and art TVs like Samsung’s Frame or Hisense’s Canvas TV.

So the requirement for this ambient will be a proximity sensor seemingly on the TVs, but I imagine as we get closer to seeing these, there might be away to trigger this dashboard via remote – that ultimately remains to be seen. But much like a smart speaker or smart display, these next-generation Google TVs will have far-field microphones built-in to let you ask Gemini or ‘Hey Google’ for really limitless queries.

Since it’s a big screen, though, Google’s optimizing and tweaking the Gemini LLM a bit in that along with answering a question like how big our galaxy is or asking to plan a trip, it can suggest YouTube videos to jump into to learn more about the answers (or answers). As we’ve seen with other AI-powered assistants, it’s sort of customizing that power to the device you’re receiving it on.

(Image credit: Future)

We’re covering all of the latest CES news from the show as it happens. Stick with us for the big stories on everything from 8K TVs and foldable displays to new phones, laptops, smart home gadgets, and the latest in AI.

And don’t forget to follow us on TikTok for the latest from the CES show floor!

Of course, since it’s Google TV and works with plenty of Google services, you’ll also be able to control smart home devices with it and even run automation. Unlike the Google Assistant, you’ll be able to speak more like a human and with more conversational terms rather than the more rigid voice requests of the past.

This also applies when you’re asking for help finding something to watch; you can be as specific as an animated movie appropriate for all ages from Disney or something much more wide-reaching to get suggestions. As we saw with the Google TV Streamer, Gemini will summarize key reviews and sentiments about the selected title when you open up a film.

From an early look, it’s a pretty AI glow-up for Google TV as a platform, but of course, it all comes down to hardware partners. While we already have an idea of what TV makers have already opted for Google, it remains to be seen where this next-generation experience will be found, but it is announced that we’ll see these arrive on future hardware by the end of 2025.

It’s also worth noting that since there is an additional requirement for far-field microphones and proximity sensors, this likely won’t be arriving on the current Google TV Streamer. Who knows, Google might be cooking up a hardware pack.

You might also like

• This might be the coolest iPhone accessory at CES 2025, perfectly mixing tech and aesthetics
• Samsung’s Ballie robot companion still comes in yellow, but has more AI and a promised 2025 launch
• I saw Samsung’s new OLED TV with Glare Free 2.0 tech, and it fixes the Samsung S95D’s biggest problem

• Microsoft aims to spend $80 billion on AI data centers in first half of 2025
• Half of the amount will be destined for the United States
• Upskilling citizens is recognized as crucial

Microsoft has revealed plans to invest $80 billion in AI data centers throughout the first half of 2025.

Vice Chair and President Brad Smith took to a New Year blog post to boldly state the US’s opportunity to “​​harness new technology to invigorate the nation’s economy” is the greatest it has been since the advent of electricity.

Speaking about AI’s power to “drive innovation and boost productivity” across all sectors, Smith confirmed Microsoft is on track to invest $80 billion throughout its fiscal year 2025, which ends on June 30, 2025.

Microsoft AI data center investment

The company’s data center investments include cash to help train AI models and deploy AI and cloud-based applications.

Smith confirmed over half of the total investments will be destined for the US, adding Microsoft’s plans are also helping to grow the country’s economy by supporting the likes of competitors, chip suppliers, applications companies, systems integrators, service providers, software developers, construction firms, material manufacturers and more, who all play an important role in the successful deployment of AI data centers.

It was also noted the 2019 AI Executive Order prioritized the integration of AI upskilling within educational curricula and educational grant applications – a perfect opportunity for Microsoft to boast it is committed to training 2.5 million American students, workers and community members with AI skills in 2025.

AI upskilling programs are usually a highlight of investments. For example, when Microsoft decides to open a new data center, it typically aims to give back to the community in the form of training.

Apart from investing in AI and upskilling the country’s citizens, Microsoft’s plans for America’s economic success also include a third element: exporting. Smith noted China’s efforts to become a leader in AI has put the US under pressure, and it must cooperate with “allies and friends” to maintain its position at the top.

In 2024, Microsoft announced more than $35 billion in investments across 14 countries to build AI and cloud data center infrastructure.

It’s clear that Microsoft sees itself as having a vital role in helping the US maintain its AI momentum. Smith concluded: “The key to the future is to bring together the best of what we can offer across American society, from across our private sector, educational and non-profit institutions, and government.”

You might also like

• Check out the best AI writers around today
• We’ve listed all of the best cloud computing services
• UK Government reveals all on its new bid to boost AI Security Research

Over the last decade and a half, the internet has evolved from a search-based model into a robust, interconnected ecosystem of content producers and aggregators. Early knowledge navigation was driven largely by search engines, with Google’s Knowledge Graph being a notable game-changer. The tool highlighted how audiences were increasingly satisfied with direct answers rather than detailed content, even though most answers were rooted in content produced by knowledge platforms.

Over time, content providers adapted to this system, leveraging search engine optimization (SEO) and structured data to keep their visibility and user traffic strong. This symbiotic relationship created an entire industry anchored on search-based marketing, which thrived on the interdependence of content producers and search engines.

The landscape changed again with cloud computing. Companies quickly embraced Infrastructure-as-a-Service to streamline processes and reduce costs, leading to the rise of Software-as-a-Service (SaaS) models. These cloud-based business models generated a wave of innovative companies that redefined how software was created, distributed, and accessed, giving rise to an era of cost-effective and scalable technology solutions.

Fast forward to another major technological shift: conversational interfaces. While early virtual assistants like Siri and chatbots were innovative, they still relied heavily on traditional knowledge resources. These systems fundamentally operated within established business models, simply presenting new ways for users to interact with content, rather than transforming how knowledge was structured and consumed.

Which brings us to the meteoric rise of large language models (LLMs) and AI agents. While the underlying AI technology has been around for years, the explosion of AI tech in the last two years has been a game-changer for businesses across sectors. These major shifts have also disrupted the knowledge creator and user dynamic in a way that threatens content ownership, attribution, and monetization for knowledge platforms.

The fragmentation of the knowledge ecosystem

AI-driven agents are not merely interfaces; they synthesize and present information in a way that can obscure or bypass original content creators entirely. In many cases, these agents surface knowledge without attributing the source, effectively severing the feedback loop that used to send traffic back to content producers. As AI systems increasingly become the interface through which users consume information, the gap between knowledge sources and user interaction has widened. This change creates a “knowledge fragmentation” effect, separating the platforms that produce knowledge from the platforms that distribute it. This fragmentation raises three critical issues for the larger knowledge ecosystem:

• Answers are not knowledge: While LLMs can retrieve data and generate responses, they often lack the nuanced understanding needed to address complex questions. These systems can provide an answer, but not always the specific context required to apply those answers in real-world scenarios. As a result, they risk oversimplifying knowledge into basic answers that lack depth or relevance.
• The LLM brain drain: The current reliance on AI-driven knowledge diminishes the feedback loop that has historically fueled content creation. As users grow accustomed to instant answers without needing to consult detailed sources, the incentive to create and share nuanced and new information decreases. This brain drain effect threatens the richness and breadth of knowledge in our ecosystem, leaving us with static, outdated data in place of evolving insights and new content.
• Erosion of Trust: Many users of AI tools are questioning the trustworthiness of responses. Without transparency around the source and credibility of information, AI tools risk losing user confidence, especially in technical fields or for corporate customers where accuracy is critical.

Knowledge-as-a-Service – a new business model

In response to these challenges, community platforms are championing a new business model: Knowledge-as-a-Service. This model emphasizes the creation, curation, and validation of knowledge within a sustainable ecosystem where content creators, platforms, and AI providers coexist and support each other. At its core, Knowledge-as-a-Service means establishing a high-quality, domain-specific knowledge base that powers technology advancements while ensuring fair and transparent use of data.

For many, this means providing access to the highly trusted, validated, and up-to-date technical content on a platform. The platform supports both existing and emerging knowledge, creating a self-reinforcing ecosystem where new information is validated, indexed, and made accessible for developers and LLM providers. By fostering this continuous loop of knowledge creation and validation, businesses can begin to address “LLM brain drain” and the lack of trust that plagues the current knowledge economy.

Powering the future

The shift towards Knowledge-as-a-Service underscores the need for ethical data use and reinvestment in knowledge-producing communities. For the model to work, content providers and platforms must ensure fair attribution and recognition for their contributors. Transparent partnerships with LLM providers are key, as they create a pathway for AI tools to responsibly leverage community-generated knowledge without depleting the source.

The future of the knowledge economy rests on a collaborative approach that respects content creation and values transparency. Knowledge-as-a-Service offers a promising blueprint for platforms to remain relevant while supporting a new generation of digital tools and applications.

This strategy isn’t just a response to current challenges, but also a vision for a sustainable future where the exchange of knowledge remains open, accessible, and beneficial to all stakeholders. As the digital landscape continues to evolve, companies must rise to the challenge of preserving the integrity and richness of community-driven knowledge - or risk losing the foundation upon which the internet has been built.

We've featured the best small business software.

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

• Popular open source vulnerability scanner Nuclei was found to be vulnerable itself
• A bug allowed crooks to smuggle malicious code past the scanner
• The vulnerability was fixed in September 2024, but many users still haven't updated

A vulnerability scanning tool was found to have been vulnerable itself, allowing crooks to smuggle malicious code past the gatekeeper.

Cybersecurity researchers from Wiz found a bug in ProjectDiscovery’s Nuclei in August 2024, after investigating the open source vulnerability scanner, which is designed to automate the detection of security issues across various protocols, systems, and applications using customizable YAML-based templates.

The bug is tracked as CVE-2024-43405, and was given a severity score of 7.8 (high). In versions 3.0.0 - 3.3.2, a vulnerability in Nuclei's template signature verification system allowed malicious actors to bypass signature checks and possibly run malicious code via custom code template, it was said.

Upgrades and workarounds

A fix was released in early September 2024, making version 3.3.2 the first clean one. Users are urged to apply the fix immediately, since cybercriminals are expected to now start scanning for vulnerable endpoints. Those that cannot apply the patch in a timely manner should stop using custom templates, and instead only use trusted, verified ones.

“Those who are unable to upgrade Nuclei should disable running custom code templates as a workaround,” it was explained on the NVD webpage.

Wiz also stated that Nuclei should be used in a virtual machine, or isolated environment.

While open source software is generally considered safe (if nothing else, then due to countless eyes looking at the code all the time), its popularity and ease of access also make it a popular target for criminals interested in software supply chain attacks. While the exact number of Nuclei users is impossible to determine, we can say it is a popular solution, since it has 21,000 stars on GitHub, paired with roughly 2,600 forks.

Additionally, the Nuclei project boasts more than 700 contributors and has facilitated over 50 million monthly scans, indicating widespread adoption within the cybersecurity community.

Via BleepingComputer

You might also like

• More threats against open source software could be coming soon, experts warn
• Here's a list of the best antivirus tools on offer
• These are the best endpoint protection tools right now

As more companies champion a return to the office, once empty workspaces are now back to bustling with activity. But with this shift, a new issue has surfaced: noise.

Supporters of office-based work argue that the face-to-face interactions spark creativity, strengthen working relationships and boost team spirit in ways that remote work cannot match. However, noise in busy offices is proving to be a major concern for many. The challenge for businesses lies in creating an environment where the ability to quietly focus or collaborate with colleagues can coexist.

A thriving workplace or a focus killer?

Research from Jabra found that loud workspaces do more than irritate—they impact both employee wellbeing and productivity. In fact, half of UK employees reported feeling stressed by the constant background chatter and distractions in the office. Noise ranks among the top three causes of stress for employees. So why are companies insisting on a full five-day office return?

While company culture and the benefits of connecting with colleagues face-to-face are big factors, the drawbacks cannot be ignored. Noise and overstimulation can result in decision fatigue and burnout, with nearly half (47%) of employees in Jabra’s report saying office noise hampers their ability to concentrate.

Noisy workspaces – the silent threat to employee wellbeing

The consequences of a noisy work environment can be severe, with 60% of employees agreeing that such environments leave them physically drained. And over half (59%) say it is challenging to feel passionate about their work when dealing with constant noise.

Businesses must prioritize employees’ comfort and wellbeing. A recent McKinsey survey of company executives found that hybrid work models are most effective for boosting revenue, attributing this to improved employee wellbeing.

Championing a healthier office environment

But how can organizations proactively address noise-related stress at work? Creating quiet zones and thoughtfully designed meeting rooms where employees can go to focus is a good place to start, alongside spaces that are designed for more social activities and group collaboration. Ensuring that staff are equipped with technology like high-quality noise-cancelling headsets can also minimize distractions and help them to concentrate in louder environments. A calm office atmosphere that is conducive to getting work done can significantly boost employee morale and productivity.

Workers report that improved meeting audio quality can positively impact their interpersonal skills through clearer communication (46%), better collaboration (44%) and enhanced communication skills (39%).

The top three must-haves for employees according to Jabra’s research were noise-cancelling headphones, more flexibility to work from home, and designated spaces for different activities, such as quiet or collaborative zones. Investing in these and offering flexible working practices will have a significant positive impact on employee wellbeing, which ultimately leads to a happier workplace.

From chaos to calm

As more employees transition back to the office full-time, many will grapple with heightened stress levels. Factors like noise type, intensity and individual sensitivity all impact how we function at work, which can in turn influence our mental health and performance.

Companies need to cut through the noise and listen to employees to create an office environment that works for everyone. To successfully navigate the return to the office, striking the right balance between a vibrant and lively office atmosphere and enabling staff to have focused, distraction-free time is essential.

We've featured the best WFH app.

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 Tech: https://www.techradar.com/news/submit-your-story-to-techradar-pro

• A new leak may have confirmed Nvidia's RTX 5090 using 32GB of VRAM
• It comes from Inno3D's product box image acquired by VideoCardz
• An official reveal is now hours away

Nvidia's long-awaited RTX 5090 official reveal is upon us, and a leak has already kickstarted the process - an Inno3D RTX 5090 product box all but confirms the flagship GPU's 32GB of VRAM.

This comes from VideoCardz who acquired the product box image (pictured below), which reveals Inno3D's new iChill design - but most importantly, corroborates previous rumors that suggested Nvidia's RTX 5090 would have 32GB of GDDR7 memory.

While it's important to note that this isn't official as we await Jensen Huang's CES keynote (now just hours away), this is the biggest indicator. There's no confirmation on its pricing despite rumors of a $2,500 figure, but we won't have to wait very long to find out.

The RTX 5090, in all its glory. (Image credit: VideoCardz) Will the performance jump over the previous gen justify a potential huge price increase?

There may not be any official information on the RTX 5090's specifications just yet, but the leaked image points towards other rumors holding some truth. VideoCardz mentions the flagship GPU potentially taking advantage of 21,760 CUDA cores compared to the RTX 4090's 16,000 - this could certainly prove to be a huge performance increase.

Considering the RTX 4090's MSRP ($1,599 / £1,499 / AU$2,959) it won't be much of a surprise to see the 5090 reignite controversy among PC gamers. The RTX 4090 is still an absolute powerhouse for gamers (especially at 4K), but games like Indiana Jones the Great Circle have showcased its limit (while using full path tracing).

For those chasing the best of the best, you should expect the RTX 5090 to perform exceptionally if the rumors are legitimate - it could be even better if there isn't a major price jump from the RTX 4000 series' flagship GPU.

You may also like...

• Nvidia RTX 5090 and 5080 GPUs again rumored for CES – but suggestion the RTX 5080 could be positioned as a ‘professional’ GPU might worry PC gamers
• Nvidia RTX 5090 rumor suggests flagship GPU might not guzzle as much power as previously claimed – but don’t get too excited
• Latest Nvidia RTX 5000 power usage rumors make me scared that my PSU will be nowhere near enough for the RTX 5080

• Bird Buddy introduces Petal and Wonder Blocks
• It's a modular camera and accessory system
• It watches for birds, bees, and especially bugs

If I have a favorite gadget, it’s probably my Bird Buddy, a Wi-Fi-connected bird-feeding webcam that delivers beautiful, up-close images of doves, cardinals, sparrows, and woodpeckers.

Getting up close to nature is fun, even inspiring, and I think that may be why Wonder, the name for Bird Buddy’s new line of products, is introducing a couple more digital nature-watching technology ideas here at CES 2025: Wonder Blocks and Petal.

TechRadar will be extensively covering this year's CES, and will bring you all of the big announcements as they happen. Head over to our CES 2025 news page for the latest stories and our hands-on verdicts on everything from 8K TVs and foldable displays to new phones, laptops, smart home gadgets, and the latest in AI.

And don’t forget to follow us on TikTok and WhatsApp for the latest from the CES show floor!

The two products are really something of a pair. One, Petal, is an HD camera on a bendable stalk that can look like a large orange flower. It has fake leaves, one of which can be an optional solar panel to keep the camera battery charged. Petal can be placed in your garden alone, or you can pair it with Wonder Blocks, a modular habitat system where you can plant flowers, invite bees to build a hive, and even offer bird-seed-friendly trays.

Modules include a Bee Hotel, Bug Hotel, seed tray, plant base, and pedestals on which you can stack all these modules.

Image 1 of 4

(Image credit: Bird Buddy)Image 2 of 4

(Image credit: Bird Buddy)Image 3 of 4

(Image credit: Bird Buddy)Image 4 of 4

(Image credit: Bird Buddy)

The Petal stalk, which has a 1/1.9 inch sensor and shoots 12MP stills and up to 4K video, can be stuck in the dirt of one of those modules to watch the action of, say, a butterfly alighting on one of the flowers growing in your Wonder Blocks. As with Bird Buddy, the 'wonder' really begins when the AI-backed camera starts capturing photos and videos of the bird, bugs, and other nature inhabitants and sends them to the Wonder app.

Bird Buddy’s AI, the same AI used to identify birds in my Bird Buddy App, can identify the bugs and other wildlife visiting your Wonder Blocks and Petal. The Wonder Blocks system currently has no sensors, but developers tell me it could, for instance, eventually keep track of the seed you put in it, and report on how much it has left and for how many days it can feed the birds.

Bird Buddy also plans to consolidate all the Wonder Products, including Bird Buddy, into one app where you can track all of your nature-watching cameras.

(Image credit: Future / Lance Ulanoff)

Similar to Bird Buddy, the system’s AI will give you real-time and daily reports of all the different types of bugs visiting the blocks, which are about the same size of a large garden planter. You can see photos and videos of the insects, and even share them if you like.

Wonder Blocks and Petal are not a fixed pair. Remember that flexible stem? It’s bendable enough that you can wrap it around a tree branch and capture images of birds landing and bugs making the journey from leaf to brand to leaf. There are also optional Petal lens attachments for closeup and wide-angle views of birds, bees, and bugs.

(Image credit: Bird Buddy)

Despite being modular, the Wonder Block system, which Wonder has yet to announce pricing for, is large and heavy enough that it looks unlikely to get knocked over by a hungry squirrel or busy raccoon.

Since the solar panel is optional, I wonder about the longevity of that Petal battery. When I got my Bird Buddy bird feeder, we opted for the optional solar roof so I wouldn’t have to keep pulling the camera module to charge it inside my house. Bird Buddy used the somewhat aging 2.4GHz Wi-Fi band, so I’d hope that Petal and, ultimately, Wonder Blocks use Wi-Fi 5.0. It has a shorter range but a higher bandwidth.

Even so, I think Wonder Blocks and Petal might inspire my bug-watching in much the same way Bird Buddy did. This might be the first time I’m happy to see some bees build a hive near-ish my home.

Pricing for the Petal is expected to be about $99 when it ships later this year. While there's no separate pricing for the solar panel, Wonder representatives told me they expect to sell bundles where you'll get a pair of Petal cameras and solar panels for around $250. Wonder expects to start shipping Petal and Wonder Blocks in the “spring", so March-May.

You might also like

• This might be the coolest iPhone accessory at CES 2025, perfectly mixing tech
• LG unveils 2025 flagship OLED TV line-up, including ...
• Nvidia RTX 5050 to 5090 laptop GPUs spotted, suggesting ...

• Demi Moore wins Best Actress in a musical/ comedy at the 2025 Golden Globes
• This award is Moore's first major recognition in in her 45-year career
• Her win is a reminder that horror movies are more than deserving of the big awards, and makes her a shoe-in for the Best Actress Oscar

Awards season is in full swing and like clockwork, the 2025 Golden Globes has kicked off the celebrations with its recognition of the best movies of 2024. But out of all the nominations for acting, Demi Moore’s victory for her performance in Coralie Fargeat’s The Substance (one of the best horror movies on streaming) was by far one of the most memorable moments of the night - marking Moore’s first-ever major award ceremony win and a monumental recognition for horror movies.

Taking home the award for best actress in a musical/comedy (though I feel inaccurately describes The Substance), Moore followed her Golden Globe win with a powerful speech that shed light on her experiences with not being taken seriously as an actress. Moore opened her speech confessing that “this is the first time [I’ve] won anything as an actor” in her incredible 45-year career, which was a surprising discovery given that Moore was named the highest paid actress of the ‘90s.

But Moore’s moving speech didn’t stop there. She added: "Thirty years ago, I had a producer tell me that I was a popcorn actress and at that time, I made that mean that [awards] weren't something that I was allowed to have, that I could do movies that were successful, that made a lot of money, but that I couldn't be acknowledged.”

Signing off her speech, Moore concluded: “In those moments when we don’t think we are smart enough or pretty enough, or skinny enough or successful enough or basically just not enough, I had a woman say to me just know you will never be enough, but you can know the measure of your work if you just put down the measuring stick.”

Alluding to the experiences of her character in The Substance, it was a powerful full-circle moment for Moore.

Demi Moore as Elizabeth Sparkle in Coralie Fargeat's body horror The Substance. (Image credit: Mubi) Demi Moore re-writes the story for horror

There’s no hiding from the fact that horror is disregarded as a legitimate genre and shunned by the biggest awards ceremonies, and yes, I’m still bitter about Toni Collette and Lupita Nyong’o’s Oscar snubs for Hereditary (2018) and Us (2019). But alas, every cloud has a silver lining, and that silver lining is The Substance.

While I enjoy horror’s campy and, at times, unserious nature, I find that these can be damaging characteristics that can tarnish the genre’s artistic credibility. However that doesn’t mean you can’t get a damn good story out of a horror movie. When it comes to The Substance, Fargeat’s screenplay is a perfect balance of unseriousness and camp but still succeeds in offering a story with a critique of the harsh realities faced by aging women in the entertainment industry.

As a fan of horror and compelling women’s stories alike, Demi Moore’s win at the 2025 Golden Globes is a reassuring reminder that horror has substantial credibility and could serve as the catalyst for her campaign for the Best Actress Oscar. The Golden Globes are an indicator of who could win big at the Oscars, so it’s safe to say that I can’t wait to see how this race pans out.

You might also like

• I’m a huge anime fan – here’s why you should ditch Netflix for Crunchyroll in 2025
• Missing Yellowstone? Taylor Sheridan has a hit Western movie with 97% on Rotten Tomatoes on Paramount Plus you have to watch
• This app is like Letterboxd for TV shows – and it's sorted my streaming plans for 2025

• The 2025 version of the Samsung ViewFinity S8 offers a 37-inch diagonal
• As an entry-level model, it only has a built in KVM and misses out on Thunderbolt 4
• While it scores highly on color fidelity, its refresh rate won't impress gamers

Samsung has announced the launch of its largest 16:9 4k monitor to date, the ViewFinity S8 (S8OUD model), ahead of CES 2025.

The company says its new monitor series is designed to "maximize productivity, immersion, and efficiency”, making it a handy piece of kit for both casual and professional users.

The ViewFinity S8 boasts a 37-inch screen, making it around 34% larger than previous models in this line, but has been officially certified by TÜV Rheinland as an ‘ergonomic workspace display”, making it ideal as a possible business monitor. The device also includes an ‘eye-saver’ mode and flicker-free features to protect users from eye strain.

What to expect from the ViewFinity S8

Built with productivity in mind and boasting a larger screen, Samsung said “more information can be seen at a glance” with the ViewFinity S8, while also enabling users to maximize desk space due to its sleek design.

The S8 comes complete with an Easy Setup Stand, for example, making it a relatively compact monitor despite its large size, and also boasts a built-in KVM8 switch, allowing users to turn the monitor into a “single command station” designed to supercharge multitasking capabilities across several devices.

Devices can also be connected and charged simultaneously with a 90W USB-C connection.

While Samsung has been keen to emphasize the professional applications of the new S8 monitor, there are areas where casual users or gamers might be left wanting.

The monitor does not include Thunderbolt 4 capabilities, for example. Similarly, the monitor’s refresh rate leaves a lot to be desired.

Its high score on color fidelity might be a saving grace for gamers here, however. With its sRGB 99% color gamut, the S8 “delivers the color representation and brightness” according to Samsung.

Samsung announced the launch of the S8 alongside its M9 Smart Monitor and the Odyssey G6 and G8 models.

“Samsung will continue its tradition of innovating display technology in 2025 with the introduction of new Odyssey, Smart and ViewFinity models,” said David Phelps, head of the display division at Samsung Electronics America.

“Whether its new AI capabilities, size options, or groundbreaking refresh rates, Samsung will ensure that every type of user will have their needs met for enhancing productivity, gaming, and entertainment.”

You might also like

• Who needs an ultrawide? The Samsung Odyssey Ark 2nd Gen gaming monitor is here to change your gaming experience
• Need something smalelr? We've rounded up the best portable monitor picks around
• Fastest ever OLED gaming monitor from Samsung could hit 500Hz

• Bosch has revealed a new 'Battery Lock' feature for eBikes
• The feature will be available to those with a Flow+ subscription
• It will arrive via an over-the-air software update this summer

Electric bicycle component manufacturer Bosch is revealing an innovation at this year’s CES 2025 show that it hopes will serve as a powerful weapon against bike thieves.

Gregor Dasbach, head of digital business at Bosch eBike Systems, claims that the the battery is “one of the most valuable eBike components” and recognizes that there has been a spike in battery thefts as the popularity of the electrified transport continues to grow.

Arriving this summer, a new Battery Lock feature will be available as an over-the-air update via the brand’s Flow app for those with a Flow+ subscription. That subscription typically comes free for a trial period, but unfortunately requires a monthly payment after that.

The company claims Battery Lock complements the mechanical locking mechanisms that are found on eBikes with PowerTube and PowerPack battery packs by automatically ‘locking’ the battery on a digital level when the eBike is switched off.

Simply put, it means that any stolen battery pack that is inserted into another eBike with the smart system will automatically disable the electric motor and render that bicycle unusable. According to Bosch, it also makes reselling the battery packs pointless.

The digital security feature will join several other existing technologies, such as eBike Lock, which shuts the eBike down until it is reactivated using the smartphone app or by physically removing the Kiox 300 and Kiox 500 digital displays from their respective mounts.

Similarly, an eBike Alarm sends notifications to the user’s smartphone if it detects movement, while sounding both acoustic and visual warnings to anyone who dares to try and move a locked Bosch eBike.

Peace of mind for pricey eBikes

(Image credit: Bosch)

Bike theft is nothing new, but the introduction of expensive electric motors and battery components means that pedal-assisted transport is now more attractive than ever to nefarious types.

If they can't make off with the entire bike because it is carefully locked to a post, why not take the battery and make some quick cash?

Bosch says that its latest defence against the crims is compatible with all batteries or battery combinations in its smart system range, meaning it also works with DualBatteries and the PowerMore 250 Range Extender offering.

Seeing as a number of the biggest names in cycling rely on Bosch components and smart systems, including Bergamont, Cannondale, Canyon and Raleigh in the UK, its latest Battery Lock innovation will be rolled out to thousands of customers in the summer of 2025, with the potential to put a stop to battery theft and shut down illegal re-selling markets.

However, Bosch also says that owners must be subscribed to its Flow+ package to make the most of its Battery Lock update, which – after an initial free trial period – costs £35 / €40 / AU$60 (around $45) per year thereafter.

Should some customers not want to stump up the cost, it would mean that some Bosch battery packs remain unprotected by the digital locking feature and would therefore still present an opportunity to thieves.

you might also like

• BMW’s iconic M3 is going electric – and I hope battery packs and e-motors don't destroy what makes it special
• The best cheap e-bikes for 2024: Cost-effective folding, road bikes, and eMTBs
• I tried an experimental electric bike converter kit, and now I’m converted

This Amazon Kindle Scribe is perfect for students, bookworms and creatives -- and it starts at just $325 for a limited time.

• More victims of Salt Typhoon attack unveiled by WSJ
• The extent of the damage caused by the attack is still unknown
• Some telecoms providers have removed the attackers from their systems

The recent Salt Typhoon cyberattacks may have breached more telecommunications providers than previously thought, with Charter Communications, Consolidated Communications, and Windstream all now believed to also have been affected.

The fresh list of victims comes from a new report by the Wall Street Journal, who cited people familiar with the matter.

The attack also exploited Fortinet network devices that did not have up-to-date security software installed, as well as vulnerable Cisco large network routers.

Attack may have started in 2023

The attack against US telecoms providers was first publicized in a joint statement by the Federal Bureau of Investigation (FBI) and the Cybersecurity & Infrastructure Security Agency (CISA) on October 25, 2024 - however, the WSJ report states the attack is believed to have started as far back as fall of 2023 - around the same time US National Security Advisor Jake Sullivan was briefing telecom and tech executives on the depth and breadth of Chinese penetration into US critical infrastructure.

Salt Typhoon is now known to have successfully breached the networks of AT&T, Verizon, Lumen Technologies, and T-Mobile in the attack, but little is known about what data the China-affiliated group was able to access.

Both Lumen and T-Mobile have said that they successfully stopped the group from accessing sensitive customer information, with Verizon confirming that the data of a limited number of high-profile individuals involved in politics was targeted in attacks.

Salt Typhoon also gained access to a ‘lawful interception’ channel used by law enforcement agencies to perform court-ordered wiretaps for national security purposes, with China repeatedly denying any involvement in the attacks and accusing the US of spreading misinformation. China even went so far as to label Volt Typhoon - a similar group believed to be associated with Beijing - as a CIA asset set up to discredit the US’ rivals across the Pacific.

Both Fortinet and Cisco did not comment on the WSJ report, but both organizations have been in the cross hairs of cyber attacks from a range of cyber criminal groups.

Network routers with outdated firmware have been a favorite target as an initial access point for attackers and botnets for several years. Fortinet has also experienced a spate of attacks on its Windows VPN service and Fortigate VPN systems.

You might also like

• These are the best endpoint protection services around
• Take a look at our guide to the best business VPN
• Could AI soon make dozens of billion-dollar nuclear stealth attack submarines more expensive and obsolete?