If you own a house, your insurance premiums have probably surged over the last several years. A new report outlines how much worse it could get as climate disasters keep growing: In Florida, for example, the insurance cost for an average house could go up 89% in the next 30 years. In Miami, the cost for a homeowner could go up by 322%, or an additional $11,000 a year.As you pay more for insurance, the value of your house is likely to simultaneously drop if youre in a high-risk area. The report, from the climate risk analysis nonprofit First Street, estimates that an average house in Florida will lose around 29% of its value by the 2050s. Meanwhile, home values will rise in certain areaslike Madison, Wisconsinthat face lower climate risk.The report classifies around 21,000 communities as climate abandonment neighborhoodsplaces with high climate risk and spikes in insurance premiums where populations are likely to decrease. Thousands of other areas are likely to later reach that tipping point. In other high-risk areas, like Miami, populations will keep growing because there are enough other amenities that people are willing to live with the risk and expense.What were seeing is that in some places where the economic structure is a little bit weaker, the impact of climate is stronger, says Jeremy Porter, head of climate implications research at First Street. Its something thats just an additional factor that people decide, Okay, thats enough. I cant get a job here, and I also dont want to put up with persistent extreme rainfall events or something like that. As more people leave and cities have lower tax revenue, those local economies will struggle even more.Nationally, home insurance premiums are expected to rise by an average of 25.3%, with some of the steepest jumps in the West and Southwest. Idaho will see costs rise 50%; premiums in New Mexico could go up by 82%.In the past, insurance costs were relatively stable. In the 2000s and early 2010s, they were 7%-8% of mortgage payments. But over a little more than a decade, theyve grown by 115%. Thats because of the enormous losses that insurance companies are facing as disasters grow. In 2023, insurers paid out around 10% more than they collected in premiums. (Ironically, despite the impact on their business from climate change, insurance companies continue to invest in fossil fuels.) The report calculated how much insurers costs are likely to grow because of climate change, and how much premiums will correspondingly rise. When a homeowners insurance premiums go up, their house will be worth less when it sells. And as buyers better understand climate risks, that also impacts value. When a house gets added to a FEMA zone, for example, that discounts the value by 4%. Once awareness is raised around it, it does make the property a little less desirable, Porter says. It raises the flag that theres going to be a little more in terms of cost of homeownership for this property.In Paradise, California, after a devastating wildfire in 2018, insurance costs rose 36.8% by 2023. Property values dropped by 42%.First Street makes tools that homebuyers can use to better understand the risk of climate disasters like flooding or wildfire for a specific house. Now, if you search for a house on Zillow, Realtor.com, Redfin, or Homes.com, that data is integrated. A recent Zillow study found that 80% of homebuyers now consider at least one climate risk when theyre searching for a house.Homeowners can use the same tools to find ways to make their houses more resilient. If you live in a neighborhood that hasnt flooded in the past, for example, but First Streets models tell you that theres now a strong likelihood of flooding, you might invest in a rain garden, permeable pavement, or ditches in your yard (or, depending on your risk, you might spend much more to elevate your house). Insurance companies are beginning to more proactively encourage homeowners to make changes to prepare for disasters, including a California startup that focuses on homes in high-risk fire zones. City planners can use the same tools to plan for resilience, and try to help residents avoid some of the financial costs that the report predicts. While only a handful of large cities have teams focused on climate risk, the tools can help fill the gap for others. Understanding what parts of the community are at risk, which residents, which assets, which infrastructure pieces, gives them the ability then to effectively and efficiently allocate resources to protect the community and really adapt to the climate risk, Porter says.

London-based OEB Architects has completed Lewisham Loggias, a colourful loft extension that takes cues from Italian palazzos and paintings.The project adds an extra floor to the south London home of a couple, Isabel and Jamie, and their growing family.The loft extension features a pair of colonnaded bay windowsOEB Architects designed the extension to reflect the proportions of the Victorian property and reference the couple's memories of holidaying in Italy.This resulted in a pair of colonnaded bay windows featuring red and yellow details, and a "zinc swoop" containing a new staircase.The design includes red columns and yellow architravesAccording studio founders Tessa Baird and David Eland, the project demonstrates that "lofts don't always need to be plain slate or metal boxes"."We like it when a loft extension continues some of the personality of the building below and its inhabitants, rather than pretending to blend into the roofscape," Eland told Dezeen.A bedroom is located on one side of the loftBaird and Eland kickstarted the design process by asking their clients to compile a mood board of artworks, spaces, objects, details, materials and atmospheres that had caught their attention."One of the most exciting parts of our job is to see how these references can combine with the more functional aspects of a brief," said Eland."Isabel and Jamie responded really well to this task, preparing a rich collection of references including memories of holidays, times with family, art they liked and little details on the house they enjoyed."The gable end wall features a round windowItalian art and architecture was a recurring theme, prompting OEB Architects to develop a dormer with an ornate, symmetrical facade.Architraves were painted yellow, while columns were picked out in red.Read: Studiotwentysix adds plywood-lined loft extension to Edwardian family homeEland said the aim was to create a palette with"a balance of bold and soft, earthy colours, and with moments of structure mixed with moments of playfulness".The square-section columns were rotated by 90 degrees, creating edge details in keeping with the dog-tooth brickwork of the eaves below.A "zinc swoop" contains a new staircase topped by a skylightThe loft extension adds a new primary bedroom suite to the house.The bedroom aligns with one of the two dormers, while the other provides an en-suite bathroom with a walk-in wardrobe behind.A new staircase sits within the curved zinc-clad volume in between the two dormers, with a skylight above bringing in plenty of daylight.In the bathroom, a circular mirror reflects the window on the opposite wallThe interiors are just as quirky as the exterior. Highlights include a chequered tile floor, a freestanding green bath, fluted sconces and a mirror that mimics a circular window on the opposite gable wall.The house's elevated position means that Lewisham Loggias is visible from the surrounding streets.Details include a green bath and a chequered tile floorThe architects hope the project shows how a loft extension can have its own character without overshadowing the existing architecture."The more expressive details around the windows felt right for the dramatic setting of the new loft extension overlooking the edge of a historic brick quarry," added Eland.Other loft extension projects in London include the pink-toned Narford Road loft by Emil Eve Architects and the Japanese-inspired Leaf House by Szczepaniak Astridge.The photography is by Taran Wilkhu.Project creditsArchitect: OEB ArchitectsStructural engineer: Bailiss & CoMain contractor: John DApproved inspector: StromaThe post OEB Architects adds Italian-informed loft extension to south London house appeared first on Dezeen.

The release of Bloober Teams Silent Hill 2 remake has resulted in a massively successful third quarter for Konamis fiscal year 2025. Its recent financial report covering the entire nine months saw the publisher reporting increased sales and operating profit, with the survival horror remake highlighted. It recently crossed two million units sold worldwide.Sports titles like Professional Baseball Spirits 2024-2025 and eBaseball: MLB PRO SPIRIT also did well, contributing to a record-breaking third-quarter profit per GameBiz. Overall net sales for the nine months shot up by 32 percent year-on-year, reaching 228,882 million yen. Operating profit increased by 44.6 compared to last year, topping 80,817 million yen.Konami has a pretty notable line-up for the future, including Silent Hill f, Silent Hill: Townfall and Metal Gear Solid Delta: Snake Eater. Release dates for all three have yet to be confirmed, though the company promised to do our best and provide more updates on its Silent Hill games this year. Stay tuned for updates in the meantime.

Nearly six months after launch, Game Sciences Black Myth: Wukong continues to achieve new sales heights. Niko Partners recently reported that its sold over 25 million units, up from the 20 million sales confirmed in September 2024.Furthermore, a flagship store launched on January 15th via E-commerce retail platform Tmall, stocking numerous branded goods like figurines, accessories, and more. Unsurprisingly, its also a huge success, accruing RMB 3 million (roughly $416,811) in sales in just three days. Such is the demand that certain products will be delivered within 90 days.Fans can expect more merchandise for the title, with the developer creating a dedicated team to design, produce, and market the same. Stay tuned for updates in that regard.Black Myth: Wukong is available for PS5 and PC. The Xbox Series X/S version was delayed due to technical issues, though it was rated by the ESRB last month. It received an update recently, adding new features like Boss Rush, Boss Rematch, and Journeyers Charts for improved navigation. An expansion is also in the works per Hero Games, Game Sciences external shareholder.

Monday, February 3rd, 2025Posted by Jim ThackerTutorial: Creating a Character for Games Vol. 1html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd"The Gnomon Workshop has launched Creating a Character for Games, a multi-part tutorial covering the process of sculpting a AAA-quality game character, recorded by Omid Moradi.Volume 1 covers sculpting the characters body and face, and creating 3D clothing, and provides over four and a half hours of video training in ZBrush, Maya and Marvelous Designer.Master industry-standard workflows for sculpting AAA game charactersIn the workshop, which is aimed at mid-level and senior character artists, Moradi sets out industry-standard practices for sculpting a 3D character suitable for use in a videogame.The tutorials explore how to break down complex designs into smaller, manageable parts, and how to prioritize each part of the pipeline successfully.Volume 1: sculpting the face and body, and blocking out clothingIn Volume 1, Moradi demonstrates how to sculpt the characters face and body from scratch, and how to block out the characters costume efficiently to prepare for future work.Moradi also explores how to create realistic materials, including fabric, leather and material, and how to sculpt folds and details on the clothing and armor.The workshop uses a commercial 3D scan of a male human body from 3D Scan Store for the characters hands and feet, and commercial textures from Texturing XYZ for the face.About the artistOmid Moradi is a Lead Character and Creature Artist with over 15 years of experience in the videogame, cinematics and movie industries.Franchises he has worked on include Diablo, Halo, Batman, Transformers, Gears of War, Magic: The Gathering, Hellblade II, Dead Island, Valorant and The Wheel of Time.Pricing and availabilityCreating a Character for Games is available via a subscription to The Gnomon Workshop, which provides access to over 300 tutorials.Subscriptions cost $57/month or $519/year. Free trials are available.Read more about Creating a Character for Games: Vol. 1Have your say on this story by following CG Channel on Facebook, Instagram and X (formerly Twitter). As well as being able to comment on stories, followers of our social media accounts can see videos we dont post on the site itself, including making-ofs for the latest VFX movies, animations, games cinematics and motion graphics projects.Full disclosure: CG Channel is owned by Gnomon.Latest NewsTutorial: Creating a Character for Games - Vol. 1Master industry-standard workflows for sculpting AAA game characters with The Gnomon Workshop's detailed new tutorial series.Monday, February 3rd, 2025How Vine FX created a parasitic underwater forest for The RigDiscover how the VFX studio used procedural growth simulation to create the environment for Amazon Prime Video's supernatural thriller.Friday, January 31st, 2025Boris FX releases Optics 2025Photoshop plugin and standalone app for applying cinematic looks to still images gets AI mask generation, Looks and new Particle Brushes.Friday, January 31st, 2025Thinkinetic releases Pulldownit 6 for MayaDestruction plugin for games and VFX gets a new Edge Fracture Tool, and improvements to clustering and simulation performance.Thursday, January 30th, 2025Adobe launches unmetered access to Substance 3D AssetsSubstance 3D Texturing and Substance 3D Collection plans now include unlimited access to 19,000+ materials, 3D models and HDRIs.Thursday, January 30th, 2025Adobe releases Photoshop 26.3Check out the new features in the image-editing app, with the Frame Tool getting new shape options and generative AI integration.Thursday, January 30th, 2025More NewsCheck out free Android and iOS 3D scanning app RealityScan 1.6Free tool: Animation Render Time Estimator for BlenderChaos releases V-Ray 7 for HoudiniChaos releases V-Ray 7 for MayaAdobe releases Substance 3D Modeler 1.19Red9 ProPack gets new tools created for Sonic the Hedgehog 3JangaFX teases IlluGenCheck out the new features due in EmberGen 2.0Tutorial: Creating Runtime Cinematics in Unreal Engine 5PhotoLine 25 now reads Cryptomatte dataPolygonflow releases Dash 1.8.5After Effects' latest beta lets you preview HDR compsOlder Posts

Traditional approaches to training language models heavily rely on supervised fine-tuning, where models learn by imitating correct responses. While effective for basic tasks, this method limits a models ability to develop deep reasoning skills. As artificial intelligence applications continue to evolve, there is a growing demand for models that can generate responses and critically evaluate their own outputs to ensure accuracy and logical consistency.A serious limitation of traditional training methods is that they are based on imitation of responses and restrict models from critical analysis of responses. As a result, imitation-based techniques fail to present proper logical depth when dealing with intricate reasoning problems, and generated outputs often resemble correct-sounding responses. More importantly, increases in dataset sizes do not automatically lead to improved generated response quality, negatively impacting the training of large models. These challenges draw attention to a need for different methods that better improve reasoning rather than increase computations.Existing solutions attempt to mitigate these issues using reinforcement learning and instruction tuning. Reinforcement learning with human feedback has shown promising results but requires large-scale computational resources. Another approach involves self-critique, where models assess their outputs for errors, but this often lacks consistency. Despite these advancements, most training techniques still focus on optimizing performance through sheer data volume rather than improving fundamental reasoning capabilities, which limits their effectiveness in complex problem-solving scenarios.A research team from the University of Waterloo, Carnegie Mellon University, and the Vector Institute proposed Critique Fine-Tuning (CFT) as an alternative to conventional supervised fine-tuning. This approach shifts the focus from imitation-based learning to critique-based learning, where models are trained to assess and refine responses rather than replicate them. To achieve this, researchers constructed a dataset of 50,000 critique samples using GPT-4o, enabling models to identify response flaws and suggest improvements. This method is particularly effective for domains requiring structured reasoning, such as mathematical problem-solving.The CFT methodology revolves around training models using structured critique datasets instead of conventional question-response pairs. During training, models are presented with a query and an initial response, followed by a critique that evaluates the responses accuracy and logical coherence. By optimizing the model to generate critiques, researchers encourage a deeper analytical process that enhances reasoning capabilities. Unlike traditional fine-tuning, where models are rewarded for simply reproducing correct answers, CFT prioritizes identifying mistakes and suggesting improvements, leading to more reliable and explainable outputs.Experimental results demonstrate that CFT-trained models consistently outperform those trained using conventional methods. The researchers evaluated their approach across multiple mathematical reasoning benchmarks, including MATH, Minerva-Math, and OlympiadBench. Models trained using CFT showed a significant 410% performance improvement over their supervised fine-tuned counterparts. Specifically, Qwen2.5-Math-CFT, which was trained with as few as 50,000 examples, is comparable to and sometimes even superior to models competing against it with over 2 million samples in training. In addition, the framework yielded a 7.0% improvement in accuracy on the MATH benchmark and 16.6% on Minerva-Math compared to standard fine-tuning techniques. This significant improvement shows the efficiency of critique-based learning, which often promotes good results with significantly fewer training samples and computational resources.The findings from this study emphasize the advantages of critique-based learning in language model training. By shifting from response imitation to critique generation, researchers have introduced a method that enhances model accuracy and fosters deeper reasoning skills. The ability to critically assess and refine responses rather than generate them allows models to handle complex reasoning tasks more effectively. This research offers a promising direction for improving artificial intelligence training methodologies while reducing computational costs. Future work could refine the approach by integrating additional critique mechanisms to enhance model reliability and generalization across diverse problem-solving domains.Check outthePaper and GitHub Page.All credit for this research goes to the researchers of this project. Also,dont forget to follow us onTwitterand join ourTelegram ChannelandLinkedIn Group. Dont Forget to join our75k+ ML SubReddit. Marktechpost is inviting AI Companies/Startups/Groups to partner for its upcoming AI Magazines on Open Source AI in Production and Agentic AI.The post Researchers from University of Waterloo and CMU Introduce Critique Fine-Tuning (CFT): A Novel AI Approach for Enhancing LLM Reasoning with Structured Critique Learning appeared first on MarkTechPost.

The fast development of wireless communication technologies has increased the application of automatic modulation recognition (AMR) in sectors such as cognitive radio and electronic countermeasures. With their various modulation types and signal changes, modern communication systems provide significant obstacles to preserving AMR performance in dynamic contexts.Deep learning-based AMR algorithms have emerged as the leading technology in wireless signal recognition due to their higher performance and automated feature extraction capabilities. Unlike previous techniques, deep learning models excel at managing complicated signal input while maintaining high identification accuracy. However, these models are sensitive to adversarial attacks, where little changes in input signals might result in inaccurate classifications. Defense measures, such as detection-based and adversarial training methods, have been investigated to improve the resilience of deep learning models to such attacks, making them more dependable in practical applications.Adversarial training, while effective, increases computational costs, risks reduced performance on clean data and may lead to overfitting in complex models like Transformers. Balancing robustness, accuracy, and efficiency remains a key challenge for ensuring reliable AMR systems in adversarial scenarios.In this context, a Chinese research team recently published a paper introducing a novel method called Attention-Guided Automatic Modulation Recognition (AG-AMR) to address these challenges. This innovative approach incorporates an optimized attention mechanism within the Transformer model, enabling the extraction and refinement of signal features through attention weights during training.Concretely, the suggested AG-AMR technique improves modulation recognition tasks by combining an Attention-Guided Encoder (AG-Encoder), enhanced data preprocessing, and feature embedding. The approach converts input signals into two-channel images representing real and imaginary portions using the Transformers capacity to process long-range dependencies while avoiding the local feature restrictions of CNNs and RNNs. These signals are segmented, normalized, and framed into sequences, with positional embeddings and a class token added to preserve temporal and global information. The AG-Encoder uses a Multi-Head Self-Attention (MSA) mechanism and a Gated Linear Unit (GLU) to enhance feature extraction. The MSA dynamically allocates weights to focus on essential input regions while ignoring noise, generating outputs by concatenating and converting attention scores and values. Meanwhile, GLU, which replaces traditional forward propagation networks, modulates the information flow through gates, improving the processing of temporal tasks. The combined framework efficiently extracts relevant features, reduces computational complexity, and improves robustness to adversarial perturbations by filtering out redundant or irrelevant data while preserving critical signal information.The experiments conducted by the authors thoroughly evaluate the effectiveness of the proposed AG-AMR method for automatic modulation recognition. The method is benchmarked against several models, including MCLDNN, LSTM, GRU, and PET-CGDNN, using two public datasets: RML2016.10a and RML2018.01a. These datasets feature diverse modulation types, channel conditions, and signal-to-noise ratios, offering a challenging environment for model evaluation. Various adversarial attack techniques, such as FGSM, PGD, C&W, and AutoAttack, are applied to assess robustness against adversarial samples. The impact of key parameters, including frame length and network depth, on model performance, is analyzed, revealing that deeper networks with optimized frame lengths enhance recognition accuracy. Performance metrics, including training time, accuracy, and model complexity, are systematically compared across datasets, showcasing AG-AMRs superior resilience and classification performance under adversarial conditions.To summarize, the AG-AMR technique represents a substantial advance in automated modulation recognition by including an improved attention mechanism in the Transformer model. This novel technique solves critical difficulties in dynamic wireless communication situations, including signal complexity and vulnerability to adversarial attacks. Extensive experiments show that AG-AMR beats existing models regarding resilience, accuracy, and efficiency, making it a promising solution for real-world applications such as cognitive radio and electronic countermeasures.Check outthePaper.All credit for this research goes to the researchers of this project. Also,dont forget to follow us onTwitterand join ourTelegram ChannelandLinkedIn Group. Dont Forget to join our75k+ ML SubReddit.(Promoted) Mahmoud GhorbelMahmoud is a PhD researcher in machine learning. He also holds abachelor's degree in physical science and a master's degree intelecommunications and networking systems. His current areas ofresearch concern computer vision, stock market prediction and deeplearning. He produced several scientific articles about person re-identification and the study of the robustness and stability of deepnetworks.Mahmoud Ghorbelhttps://www.marktechpost.com/author/mahmoudghorbel11/Beyond Passwords: A Multimodal Approach to Biometric Authentication Using ECG and Iris DataMahmoud Ghorbelhttps://www.marktechpost.com/author/mahmoudghorbel11/Frequency-Selective Adversarial Attack Against Deep Learning-Based Wireless Signal ClassifiersMahmoud Ghorbelhttps://www.marktechpost.com/author/mahmoudghorbel11/Balancing Privacy and Robustness in NLP: A New Approach for Secure Prompt Learning in LLMsMahmoud Ghorbelhttps://www.marktechpost.com/author/mahmoudghorbel11/Adversarial Machine Learning in Wireless Communication Systems [Recommended] Join Our Telegram Channel

On this date a year ago, I was first in line for a demo of Apple Vision Pro at Apple Lakeside Shopping Center in New Orleans. Anticipating a crowd, I arrived an hour before the store opened. However, I still would have been first in line if I arrived 10 minutes early. It was launch day for Apple Vision Pro, unveiled eight months earlier, and many people were still discovering it was now available.One year later, Apple Vision Pro has improved enough through software updates to make it more capable than on day one. Making ordinary photos come to life by adding depth is amazing. Extending your Macs display to the size of two 5K monitors side-by-side is impressive. More than anything, Apple Vision Pro has become the best personal and portable immersive movie theater on the planet.But with the one-year mark comes one less excuse for Apple Vision Pro. We can no longer excuse Apple Vision Pros shortcomings by saying, Its just the first year.Where are the immersive sports replays that happened over the last week and not last year? When will there be an immersive film that gets discussed because its a great movie and not because its a novel format? How long until Apple releases more visionOS-native versions of its iPad apps instead of relying on the compatibility folder?visionOS 3 will probably answer a lot of the platform-specific infancy issues. Window management is surprisingly difficult. Viewing your Apple Watch or iPhone through Apple Vision Pro isnt pretty. The Photos app is a viewer and not a full-featured experience like on iPhone, iPad, and Mac. What about iPad Virtual Display?As we hit the one year mark for Apple Vision Pro, well no longer be able to dismiss as many concerns because its a brand new platform. Apple Vision Pros $3500 price tag still shields it from the harsher scrutiny of mainstream adoption. But if the company wants to continue down the Apple Vision path, more of these early shortcomings will need to be solved before a more affordable version enters the chat.Top iPhone accessoriesAdd 9to5Mac to your Google News feed. FTC: We use income earning auto affiliate links. More.Youre reading 9to5Mac experts who break news about Apple and its surrounding ecosystem, day after day. Be sure to check out our homepage for all the latest news, and follow 9to5Mac on Twitter, Facebook, and LinkedIn to stay in the loop. Dont know where to start? Check out our exclusive stories, reviews, how-tos, and subscribe to our YouTube channel

Feb 03, 2025Ravie LakshmananCybercrime / CryptocurrencyA Russian-speaking cybercrime gang known as Crazy Evil has been linked to over 10 active social media scams that leverage a wide range of tailored lures to deceive victims and trick them into installing malware such as StealC, Atomic macOS Stealer (aka AMOS), and Angel Drainer."Specializing in identity fraud, cryptocurrency theft, and information-stealing malware, Crazy Evil employs a well-coordinated network of traffers social engineering experts tasked with redirecting legitimate traffic to malicious phishing pages," Recorded Future's Insikt Group said in an analysis.The use of a diverse malware arsenal cryptoscam group is a sign that the threat actor is targeting users of both Windows and macOS systems, posing a risk to the decentralized finance ecosystem.Crazy Evil has been assessed to be active since at least 2021, functioning primarily as a traffer team tasked with redirecting legitimate traffic to malicious landing pages operated by other criminal crews. Allegedly run by a threat actor known on Telegram as @AbrahamCrazyEvil, it serves over 4,800 subscribers on the messaging platform (@CrazyEvilCorp) as of writing."They monetise the traffic to these botnet operators who intend to compromise users either widely, or specifically to a region, or an operating system," French cybersecurity company Sekoia said in a deep-dive report about traffer services in August 2022."The main challenge facing traffer is therefore to generate high-quality traffic without bots, undetected or analysed by security vendors, and eventually filtered by traffic type. In other words, traffers' activity is a form of lead generation."Unlike other scams that revolve around setting up counterfeit shopping sites to facilitate fraudulent transactions, Crazy Evil focuses on the theft of digital assets involving non-fungible tokens (NFTs), cryptocurrencies, payment cards, and online banking accounts. It is estimated to have generated over $5 million in illicit revenue and compromised tens of thousands of devices globally.It has also gained newfound prominence in the wake of exit scams involving two other cybercrime groups Markopolo and CryptoLove, both of which were previously identified by Sekoia as responsible for a ClickFix campaign using fake Google Meet pages in October 2024."Crazy Evil explicitly victimizes the cryptocurrency space with bespoke spear-phishing lures," Recorded Future said. "Crazy Evil traffers sometimes take days or weeks of reconnaissance time to scope operations, identify targets, and initiate engagements."Besides orchestrating attack chains that deliver information stealers and wallet drainers, the group's administrators claim to offer instruction manuals and guidance for its taffers and crypter services for malicious payloads and boast of an affiliate structure to delegate the operations.Crazy Evil is the second cybercrime group after Telekopye to be exposed in recent years, and it centers its operations around Telegram. Newly recruited affiliates are directed by a threat actor-controlled Telegram bot to other private channels -Payments, which announces earnings for traffersLogbar, which provides an audit trail of information stealer attacks, details about stolen data, and if the targets are repeat victimsInfo, which provides regular administrative and technical updates for traffersGlobal Chat, which serves as a main communication space for discussions ranging from work to memesThe cybercrime group has been found to comprise six sub-teams, AVLAND, TYPED, DELAND, ZOOMLAND, DEFI, and KEVLAND, each of which has been attributed to a specific scam that involves duping victims into installing the tool from phony websites -AVLAND (aka AVS | RG or AVENGE), which leverages job offer and investment scams to propagate StealC and AMOS stealers under the guise of a Web3 communication tool named Voxium ("voxiumcalls[.]com")TYPED, which propagates the AMOS stealer under the guise of an artificial intelligence software named TyperDex ("typerdex[.]ai")DELAND, which propagates the AMOS stealer under the guise of a community development platform named DeMeet ("demeet[.]app")ZOOMLAND, which leverages generic scams impersonating Zoom and WeChat ("app-whechat[.]com") to propagate the AMOS stealerDEFI, which propagates the AMOS stealer under the guise of a digital asset management platform named Selenium Finance ("selenium[.]fi")KEVLAND, which propagates the AMOS stealer under the guise of an AI-enhanced virtual meeting software named Gatherum ("gatherum[.]ca")"As Crazy Evil continues to achieve success, other cybercriminal entities are likely to emulate its methods, compelling security teams to remain perpetually vigilant to prevent widespread breaches and erosion of trust within the cryptocurrency, gaming, and software sectors," Recorded Future said.The development comes as the cybersecurity company exposed a traffic distribution system (TDS) dubbed TAG-124, which overlaps with activity clusters known as LandUpdate808, 404 TDS, Kongtuke, and Chaya_002. Multiple threat groups, including those associated with Rhysida ransomware, Interlock ransomware, TA866/Asylum Ambuscade, SocGholish, D3F@ck Loader, and TA582 have been found to use the TDS in their initial infection sequences."TAG-124 comprises a network of compromised WordPress sites, actor-controlled payload servers, a central server, a suspected management server, an additional panel, and other components," it said. "If visitors fulfill specific criteria, the compromised WordPress websites display fake Google Chrome update landing pages, which ultimately lead to malware infections."Recorded Future also noted that the shared use of TAG-124 reinforces the connection between Rhysida and Interlock ransomware strains, and that recent variations of TAG-124 campaigns have utilized the ClickFix technique of instructing visitors to execute a command pre-copied to their clipboard to initiate the malware infection.Some of the payloads deployed as part of the attack include Remcos RAT and CleanUpLoader (aka Broomstick or Oyster), the latter of which serves as a conduit for Rhysida and Interlock ransomware.Compromised WordPress sites, totaling more than 10,000, have also been discovered acting as a distribution channel for AMOS and SocGholish as part of what has been described as a client-side attack."JavaScript loaded in the browser of the user generates the fake page in an iframe," c/side researcher Himanshu Anand said. "The attackers use outdated WordPress versions and plugins to make detection more difficult for websites without a client-side monitoring tool in place."Furthermore, threat actors have leveraged the trust associated with popular platforms like GitHub to host malicious installers that lead to the deployment of Lumma Stealer and other payloads like SectopRAT, Vidar Stealer, and Cobalt Strike Beacon.Trend Micro's activity exhibits significant overlaps with tactics attributed to a threat actor referred to as Stargazer Goblin, which has a track record of using GitHub repositories for payload distribution. However, a crucial difference is that the infection chain begins with infected websites that redirect to malicious GitHub release links."The distribution method of Lumma Stealer continues to evolve, with the threat actor now using GitHub repositories to host malware," security researchers Buddy Tancio, Fe Cureg, and Jovit Samaniego said."The malware-as-a-service (MaaS) model provides malicious actors with a cost-effective and accessible means to execute complex cyberattacks and achieve their malicious objectives, easing the distribution of threats such as Lumma Stealer."Found this article interesting? Follow us on Twitter and LinkedIn to read more exclusive content we post.SHARE

Groundhog Day arrived on Sunday, Feb. 2, and Punxsutawney Phil, the most famous groundhog prognosticator, saw his shadow, thus predicting six more weeks of winter weather. Boooo, say those of us already sick of the cold. But there are some questions about Phil's accuracy. In fact, an opposum named Sand Mountain Sam in Boaz, Alabama predicted an early spring this year, when he didn't see his shadow. Just how accurate are these animal prognosticators, anyway?The National Oceanic and Atmospheric Administration checkedPunxsutawney Phil's track recordlast year and found the fuzzy ambassador's predictions lacking. This year, NOAA took a deeper dive into Groundhog Day results andgraded the work of 19 groundhogsand groundhog surrogates to determine the accuracy of their predictions. Spoiler: Phil came in 17th. Enlarge Image A groundhog nibbles on some vegetation in Virginia. NPSNOAA ranks the groundhogs Upgrade your inbox Get cnet insider From talking fridges to iPhones, our experts are here to help make the world a little less complicated. NOAA set a few ground rules. The groundhogs and "alternative groundhogs" must have been prognosticating for at least 20 years and must also have been active as of Feb. 2, 2024. The competitors included 13 groundhogs, three taxidermied groundhogs, one never-seen creature presumed to be a groundhog, a prairie dog statue and a tortoise.NOAA used data from its National Centers for Environmental Information, which maintains an archive of ocean, weather and climate data. The agency compared the groundhogs' track records with US March temperature averages from 2005 to 2024.All hail Staten Island Chuck, a woodchuck at the Staten Island Zoo in New York City. Chuck, also known as Charles G. Hogg, topped the rankings with 85% accuracy. He famously bit the mayor in 2009 when Michael Bloomberg attempted to coax him out for a Groundhog Day ceremony.The next-best groundhog was General Beauregard Lee from the Dauset Trails Nature Center in Jackson, Georgia with 80% accuracy. Third place belongs to Lander Lil, a bronze statue of a prairie dog in Lander, Wyoming. Lander Lil came in at 75% accuracy. Groundhogs don't live in Wyoming, but prairie dogs do. The prognostication principles are the same. If Lander Lil "sees" a shadow on Groundhog Day, then more winter is expected.How Punxsutawney Phil fared Enlarge Image Mojave Max is a desert tortoise like the one pictured here in a National Park Service image. NPSOverall, 14 of the groundhogs had 50% accuracy or better. Poor Punxsutawney Phil managed only 35% accuracy, landing him near the bottom of the rankings. At least he's more accurate than Mojave Max, a desert tortoise at the Springs Preserve in Nevada who took the last spot with 25% accuracy.The ranking might not be fair to Mojave Max, though. The tortoise doesn't get paraded around on Groundhog Day. His predictions are based on when he emerges from brumation (like hibernation for reptiles). Max gets up sometime between February and April each year."Warmer temperatures, longer daylight hours and his own internal clock are factors known to contribute to his emergence every year," the Clark County Desert Conservation program said. His appearance marks the unofficial start of spring for southern Nevada.Groundhog Day has been a tradition at Gobbler's Knob in Punxsutawney, Pennsylvania since the late 1800s and it's still going strong. NOAA's debunking of woodchuck weather forecasting doesn't detract from the fun of the celebration. What Phil lacks in accuracy he makes up for in sheer fame, cuteness and charisma. So go ahead and cheer for Phil, but maybe base your spring plans on a rodent from Staten Island rather than Punxsutawney.