Published December 24, 2024 10:00am EST close 'CyberGuy': Timeshare terror leaves retired couple $50K in the hole after being scammed The FBI is warning timeshare owners of a telemarketing scam tied to a Mexican drug cartel. Tech expert Kurt Knutsson provides steps to stay safe. The FBI is issuing a stark warning to timeshare owners about a widespread telemarketing scam linked to a violent Mexican drug cartel. This scheme targets unsuspecting property owners, leading to significant financial losses. Here's what you need to know and how to protect yourself. Illustration of a timeshare property (Kurt "CyberGuy" Knutsson)The Dimitruks' devastating timeshare scam experienceIn late 2022, Mr. and Mrs. Dimitruk, a retired Canadian couple, received a call about selling their Florida timeshare. The scammers, aware of their specific timeshare details, promised a Mexican buyer willing to pay above market value. The fraudsters employed an intricate process involving a fake New York escrow company, ecurrencyescrow[.]llc. The Dimitruks were asked to complete forms and wire more than $3,000 for "administrative" and "processing" fees.For almost a year, the scammers made additional financial demands, citing various taxes and fees. The couple even sent $5,000 to pay off their remaining timeshare balance, believing it was part of the sale process. Mr. Dimitruk, a 73-year-old retired long-haul truck driver, revealed in an interview with KrebsOnSecurity that they lost more than $50,000 to this scam. Even after this substantial loss, the scammers continued to contact them, claiming their money was waiting and urging further payments. Fake New York escrow company email (KrebsOnSecurity) (Kurt "CyberGuy" Knutsson)Cartel connections to fraud schemesThe FBI has linked these timeshare fraud schemes to the Jalisco New Generation drug cartel in Mexico. According to a July 2024 warning from theFBI and the Financial Crimes Enforcement Network, these scams are part of the cartel's efforts to diversify their revenue streams and finance other criminal activities, including drug trafficking.Since at least 2012, the cartel and other Mexico-based transnational criminal organizations have increasingly targeted U.S. owners of timeshare properties in Mexico, particularly older adults who are often more vulnerable to such scams. The proceeds from these fraudulent activities not only support the cartel's operations but also contribute to the manufacturing and trafficking of dangerous substances like fentanyl into the United States.How these scams workThe Jalisco New Generation Cartel (CJNG) has expanded its criminal activities beyond traditional drug trafficking into sophisticated scams, including timeshare fraud targeting unsuspecting individuals, particularly Americans. Here are some of the methods employed by the cartel in executing these scams:Impersonation of legitimate entities: Scammers often pose as legitimate real estate agents, escrow companies or even officials from U.S. government agencies like the Treasury Department. This tactic is designed to instill a sense of trust and urgency in potential victims, making them more susceptible to fraud.Targeting vulnerable populations: The cartel primarily targets elderly Americans who own timeshares in Mexico. These individuals are often contacted with offers to buy their timeshares at inflated values, but they are required to pay various fees upfront, such as taxes or closing costs, before any transaction can be completed. Once these payments are made, the scammers disappear, leaving victims with significant financial losses.Use of call centers: The CJNG operates illegal call centers where employees, often unaware of the cartel's true nature, engage in telemarketing schemes. These centers are strategically located in regions with high unemployment rates, providing a pool of workers who may be desperate for jobs. The call centers not only facilitate scams but also serve as a means for the cartel to exert control over local populations through intimidation and violence.Complex fraud schemes: The scams can involve multiple layers of deception. For instance, victims may be contacted multiple times by different scammers posing as various professionals (e.g., lawyers or real estate agents) who claim they can assist with selling their timeshares or recovering lost funds. This re-victimization often leads to further financial exploitation.Violence and intimidation: The cartel employs extreme measures to maintain control over its operations and silence potential whistleblowers. Reports indicate that workers attempting to quit these call centers have faced dire consequences, including murder, which serves as a chilling message to others considering leaving the cartel's employment. This brutal enforcement mechanism not only protects their operations but also instills fear within communities.Digital infrastructure: The CJNG utilizes a network of fraudulent websites and domains that appear legitimate at first glance. These websites often mimic real escrow and real estate firms, making it difficult for victims to discern the authenticity of their interactions. Many of these domains have been linked back to a central hub that manages multiple scam operations simultaneously.By understanding these operational tactics, you can better appreciate the complexities and dangers associated with scams perpetrated by organized crime groups like the Jalisco New Generation Cartel. This knowledge is crucial for potentially preventing future victimization. Government warning (U.S. Department of the Treasurys Office of Foreign Assets Control) (Kurt "CyberGuy" Knutsson)Protecting yourself from timeshare scamsTo avoid falling victim to similar scams, it is crucial to take proactive steps to safeguard your financial interests:1) Verify the buyer's identity and offer legitimacy: Always confirm the identity of any potential buyer and the authenticity of their offer. Contact the timeshare company directly to validate any claims made by the buyer.2) Research any company that contacts you: Conduct thorough research on any business reaching out to you. Look for reviews, complaints and verify their credentials through reliable sources.3) Be wary of upfront fees: Legitimate transactions typically do not require upfront fees for administrative or processing purposes. If a company requests such payments, exercise caution.4) Use secure communication channels: Avoid sharing personal or financial information over unsecured methods such as phone calls or emails. Opt for secure communication channels whenever possible.5) Do not click on links: Avoid clicking on any links or downloading attachments from unsolicited emails. Scammers often use these tactics to steal your personal information. The best way to safeguard yourself from malicious links that install malware, potentially accessing your private information, is to have antivirus software installed on all your devices. This protection can also alert you to phishing emails and ransomware scams, keeping your personal information and digital assets safe.Get my picks for the best 2024 antivirus protection winners for your Windows, Mac, Android and iOS devices.6) Consult with professionals before proceeding with transactions: Seek advice from a real estate attorney or a trusted financial advisor before engaging in any transactions. Their expertise can help you navigate potential pitfalls.7) Report suspicious activity to authorities: If you suspect you've been targeted by a timeshare scam, don't hesitate to promptly report it to local authorities, theFBI's Internet Crime Complaint Center at ic3.gov or the Federal Trade Commission. Reporting can help protect others from falling victim to similar schemes.8) Invest in personal data removal services: While the advice provided is valuable, the most crucial step in protecting yourself from such scams is to minimize your online presence. By reducing the amount of personal information available on the web, you make it significantly harder for scammers to target you. No service promises to remove all your data from the internet. However, having a removal service is great if you want to constantly monitor and automate the process of removing your information from hundreds of sites continuously over a longer period of time.Check out my top picks for data removal services here.Kurt's key takeawaysHere's the deal. These timeshare scammers are clever, they're persistent, and they're backed by some seriously bad dudes. But don't let that scare you into inaction. Remember, knowledge is power. By staying informed and skeptical, you're already one step ahead of these fraudsters. Trust your gut. If something feels off, it probably is. And don't be afraid to ask for help or report suspicious activity. Let's make life a whole lot harder for these scammers and keep your hard-earned money where it belongs: in your pocket.What additional steps do you think authorities should take to combat telemarketing scams targeting vulnerable populations? Let us know by writing us atCyberguy.com/Contact.For more of my tech tips and security alerts, subscribe to my free CyberGuy Report Newsletter by heading to Cyberguy.com/Newsletter.Ask Kurt a question or let us know what stories you'd like us to cover.Follow Kurt on his social channels:Answers to the most asked CyberGuy questions:New from Kurt:Copyright 2024 CyberGuy.com.All rights reserved. Kurt "CyberGuy" Knutsson is an award-winning tech journalist who has a deep love of technology, gear and gadgets that make life better with his contributions for Fox News & FOX Business beginning mornings on "FOX & Friends." Got a tech question? Get Kurts free CyberGuy Newsletter, share your voice, a story idea or comment at CyberGuy.com.

ZDNET's key takeaways The Samsung Neo QLED (QN90D) 98-inch TV lists for $15,000, but is marked down to $13,000 at most vendors. Its picture quality rivals OLED models in contrast and motion handling while exceeding most of them in peak brightness. The location of its connection ports can make setting up such a large TV a bit fussy. more buying choices It's rare for me to be blown away by the picture quality of any TV I'm reviewing. But with Samsung's QN90D Neo QLED TV, that's what happened -- in a big-time (ultra-large) way.That's because Samsung sent the 98-inch model to our ZDNET lab in Louisville, and I had the pleasure of spending some time with it, fully daunted by its imposing $15,000 price tag. This is the most expensive model of any TV I've tested. Also: The next big HDMI leap is coming next month - what the 2.2 standard means for youSo, does the QN90D live up to its price point? If your budget falls into the $10K+ cost echelon, then you've already jumped the first hurdle in acquiring one. And if that's the case, then the answer is a resounding "Yes." details View at Best Buy To be clear, there are dozens of 98-inch TVs on the market that are much easier on the wallet than this puppy. For example, you can buy two of TCL's 98-inch QM8 Mini-LED for less than one QN90D. But there's a substantial difference in the technology behind these models, namely Samsung's treatment of Quantum Dot overlays on a mini-LED backlit display.Also:How I optimized this 98-inch TV to feel like a movie theater (and it's on sale)This formula is known as "Neo QLED," in which case Samsung has swapped out conventional LEDs for tens of thousands of mini-LEDs. These components consist of manufactured nanocrystals with ultra-fine semiconductor materials, allowing for a higher magnitude of these units to distribute light accurately and efficiently. The result, as Samsung puts it, is "brighter whites, darker blacks, and overall, a better color range in between."I can attest to this. The QN90D offered up a crystal-clear picture with glistering highlights, vibrant colors, and the truest blacks I've witnessed on a non-OLED set. With its spread of UHD micro dimming zones, dark areas are so piercingly black that focusing your eyes on shadows is like staring into a void. This is fine by me because they're shadows, and I'm a fan of higher contrast ratios. Adam Breeden/ZDNETBut in some extreme cases, dark objects were so dark that their features were lost. Picture a black-clad motorcyclist racing down a sunlit road -- black helmet, black jacket, black bike -- and in this distant, wide-angle shot, the cyclist and his bike become almost a uniform silhouette, revealing no distinguishable textures or dimensionality. To be clear, this would be a demanding ask of any TV for the scene I'm describing. A desirable level of detail in such inky blackness may only be captured on Samsung's Neo QLED 8K QN990C, which goes for an eye-watering $40,000 and is the subject of a different review altogether.What impressed me most was the in-your-face brightness and extraordinary color gamut of the QN90D. It's bright, but not offensively so. I'd rather call the overall effect "convincing." Viewing the motion graphics at the beginning of Warner Brothers and 20th Century Fox productions, for example, felt like looking into the window of an uber-vivid metaverse. In a few instances, the color vibrance was so intense that some viewers might find it off-putting, such as the radioactive red glow of Netflix's logo. There are settings to reduce this intensity. For starters, you won't want to set the TV's picture mode to "Dynamic" because then every color will gleam with supersaturated abandon. (I'll speak more on setting options below.) Adam Breeden/ZDNETOtherwise, I didn't encounter disproportionate levels of color or tonality. This held true in my assessments of glare and reflection, plus a scope of the QN90D's viewing angles. Standing just two feet from the screen at an unreasonably tight angle, the picture only suffered a minor loss of coloration (a slight washout), yet it retained its 4K QLED sharpness.I was equally impressed with this TV's audio output. Even in the absence of a connected soundbar, the volume managed to fill our testing area more than adequately, and it did this at about one-tenth of its potential. While watching Kandahar on Prime, I never pushed the volume past 12. It frightens me to imagine the stentorian result of cranking it up to 100.Also: You can finally buy LG's transparent OLED TV - if you're willing to pay $60,000In addition to 4.2.2 channel speakers pumping out 60 watts of power, Samsung's OTS+ technology tracks the onscreen action and adjusts sounds accordingly. Its 98-inch panel consists of 8,276 square inches, allowing for more integrated speakers, which strategically create a 3D-like soundscape that seems to envelop you from all directions. No soundbar is necessary. (Although I don't discourage adding one to it.)I did change the QN90D's default sound mode to "Amplify" and noticed a marked improvement in the crispness of whispering dialogue in Dune 2andKandahar. The setting's name is a bit misleading because it doesn't amplify volume but clarity. The two films I mentioned above are excellent for TV testing. Like many movies, they have plenty of whispering dialogue but also gads of sweeping desert vistas.Color variation and rendering become more evident in viewing these mostly stark landscapes, which include everything from jutting rock formations to an actor's skin tone and Dune's reddish "spice" granules within swells of billowing sand. The bright and sunny drone shots in Kandahar, for example, were particularly stunning.My ideal picture settings for the Samsung QN90DWhen orienting myself with any new TV, I spend a lot of time testing out the settings. I toggle the options on and off while scrutinizing select scenes from certain movies. I adjust the sliders if available (e.g., tint). I experiment with Expert Settings (like switching Local Dimming to high and Contrast Enhancer to low, and vice-versa).I do a similar thing with audio. This practice of combining and/or separating setting variables is basically a trial-and-error endeavor in pursuit of seeing optimal results. Naturally, this is a subjective process because it's all about personal preferences.Also: I changed these 6 Samsung TV settings to significantly improve the picture qualityFor the QN90D, I only changed three picture settings from their default positions, and that's in large part owed to the Tizen software's AI capabilities:Under All Settings, you'll find Samsung's "Intelligent Mode" option. Turning that on disables the usual lineup of options under the Picture Mode menu (Dynamic, Eco, Filmmaker, etc.).The Picture Mode options are then replaced with three choices: "Optimized," "Eye Comfort," and "AI Customization." The latter indeed lets you customize your visual preference by selecting one of four image choices (each) for sports, movies, and general viewing.The Tizen OS's NQ4 AI Gen2 processor then used my image selections as ingredients to concoct what was, to me, the best picture possible.After all the experimentation, my selections solidified AI Customization as the clear winner in terms of a preferred picture mode. The contrast, brightness, and sharpness in scenes from these 4K flicks were incredibly satisfying. Adam Breeden/ZDNETConsider these design elements as you shopLike other new Samsung remotes, the QN90D's takes a little getting used to because of its diminutive size and its rocker buttons, which are relatively unconventional for remote controls. These small, flat, capsule-shaped buttons are designed to push toward or away from the TV to adjust the volume and navigate the menu. (The interface itself, I should note, only required a modicum of menu-diving to reach desired selections, making it one of the more user-friendly panels I've tested.)I acclimated to the remote's design swiftly, and I also appreciate the solar panel on its underside for a lifetime of use without cable charging or replacing batteries. But if it's just too much effort to pick up and use a TV remote, the QN90D also offers hands-free operation with its Bixby, Alexa, or Google voice assistants. Disappointingly, the remote is not backlit -- as I wish all TV remotes should be -- because I tend to do any binge-watching after sundown.A common complaint among users of the QN90D is the placement of its input array. Its four HDMI ports, two USBs, and other connection points are situated on the back right side of the panel. This is fairly standard for many TVs. For a television of this size, however, simply accessing the ports and then deciding how to run your wires to the nearest wall outlet can be an awkward affair. These grumblings likely stem from prior iterations of Samsung's QLED models (e.g., the QN95C) having a separate OneConnect box that can marry all your peripherals to the TV through a single translucent cable. For the QN90D, the lack of a dedicated HDMI switch is a practical concern in terms of access and positioning. And it's an aesthetic consideration if you can't stomach visible wires trailing from the electronic centerpiece of your living room.ZDNET's buying adviceI can overlook Samsung's decision to make the inputs a tad harder to reach because this TV is spectacular to watch and listen to. At $15,000, even an ultra-large 98-inch screen ought to be brimming with cutting-edge technology, elevated specs, and a profusion of cool features. The Samsung QN90D certainly qualifies with these traits.Most of all, though, it delivers an exceptional viewing experience that will dazzle your eyeballs and please your earholes. If I were in a higher tax bracket (and had a bigger house), I would undoubtedly put this massive entertainment purchase in my (gargantuan) shopping cart.Should you want a TV of similar size but with a more accessible price, consider the TCL Q6 and QM8, both of which ZDNET has tested and recommended for big-screen shoppers on a budget.Samsung 98-inch QN90D 4K QLED TV tech specsDisplay TypeNeo QLEDScreen Size98 inchesResolution4K Ultra HD (3840 x 2160)Refresh Rate120HzHDR CompatibilityQuantum HDR 32xScreen Brightness1253 nitsContrast RatioInfiniteAudioDolby Atmos 4.2.2chProcessorNQ4 AI Gen2 ProcessorGaming FeaturesMotion Xcelerator, FreeSync Premium Pro, 4K @ 144HzInput Lag (Game Mode)2.9 msInputs4x HDMI 2.1, 2x USB 2.0, 1x Ethernet, 1x RFWi-FiWi-Fi 5 (802.11ac)Smart TV PlatformTizen OSVoice AssistantBixby, Alexa, Google AssistantStreaming PlatformsNetflix, Amazon Prime Video, Disney+, HBO Max, and many moreDimensions (with stand)86.6" x 50.8" x 14.3"Weight (with stand)113.5 lbsHow we test TVsWhile testing and researching the TV featured in this review, I and other ZDNET experts kept these criteria in mind:Price:Not all budgets are created equal. And if you're working with a limited budget, that shouldn't mean you have to settle for a sub-par TV. Each TV model we review has been chosen across a variety of price points to help accommodate different needs.Screen size:The most important factor to consider, after price, when shopping for a new TV is whether or not it will fit into your space. While this specific screen size is exceptional, the QN90D is available in a wide variety of smaller sizes to suit different rooms.Picture and audio quality:A new TV doesn't mean much, even if it costs an arm and a leg, if it doesn't provide a great picture and clear audio. Each TV on this list has been ensured to support various HDR codecs, including HDR10+ and Dolby Vision, as well as enhanced audio software like Dolby Atmos, Dolby Digital Plus, and object-tracking sound.For a more detailed look, check outour extensive TV testing methodology.Featured reviews

The Shokz OpenSwim Pro bone conduction headphones sound fantastic even when underwater. The best part is that they're on sale for the holiday season.

ZDNET's key takeaways The LG G4 OLED is on sale at Amazon and Best Buy for $2,796. LG has made great strides with its OLED picture quality, and this model is truly exceptionalHowever, the regular price might be a turn-off since you can buy great TVs for half the price. more buying choices I'm not usually a fan of LG TVs, but I've tested the LG G4 OLED TV this year, and I have to admit that it has the most impressive picture quality of any TV I've ever bought, used, or tested.Also: The next big HDMI leap is coming next month - what the 2.2 standard means for youThough Cyber Week is over, you can still find the65-inch LG G4 on sale for over $700 off its retail price of $3,399 at both Amazon and Best Buy as retailers amp up their New Year's savings opportunities. At about $2,800, the LG G4 OLED is still an expensive TV (the best price we saw during Black Friday and Cyber Monday was $2,300).For that price, you could get two 65-inch Sony X90L TVs -- my pick for the best TV for the money. But if picture quality is your number one priority and you have the budget, there's no TV on the planet with a better picture than the LG G4. details View at Amazon What makes the LG G4 OLED picture so good? Several things.First of all, there are the things that make all OLED screens great, and this is especially true for LG OLEDs because they are one of the leading innovators in the technology. We're talking about true blacks (as opposed to washed-out dark grays from LCD TVs), deep contrast, more vibrant colors, and amazing dynamic range in shots that have both dark and light elements. The LG G4 has all of those qualities -- and at the highest levels I've seen on any TV -- because it is LG's flagship OLED TV for 2024.Also: I tested Samsung's 98-inch 4K QLED TV, and watching Hollywood movies on it left me in aweSo, what's new in this year's LG G4? The biggest leap forward is in picture processing, and that's made possible by the new 11 AI Processor 4K. I've always considered Sony the king of picture processing, and it was a long way back to LG and Samsung, essentially tied for second place, with budget TV makers TCL and Hisense lagging the crowd. However, with the 2024 11 AI Processor, LG is now pushing Sony for the top spot in picture processing -- and that's saying a lot.Where this really matters is when it comes to out-of-the-box picture settings, upscaling older and lower-quality 720p and 1080p content, and the TV automatically adjusting the picture to make today's HDR content look amazing. In all of those areas, the LG G4 now rivals Sony. Adam Breeden/ZDNETAnd because LG has always arguably made the highest-quality OLED TV panels, this upgrade in picture processing lifts the LG G4 to the top of the class in picture quality. For movies, shows, gaming, and virtually any other content, the LG G4 brings them to life in vibrant color, smooth motion, and incredible contrast -- more than any other TV you can buy right now.Also: The best live TV streaming services of 2024: Expert testedI tried it with some of my favorite content with challenging visuals, includingDune, Avengers: End Game, and several other cutting-edge cinematic marvels. All of the content looked better than I've ever seen on any TV I've used, bought, or tested -- and that includes some of the best TVs from Sony, Samsung, LG, TCL, and Hisense over the past few years.What are the drawbacks?There are a few drawbacks to note about the LG G4. First, of course, is the premium price. As I've already mentioned, you could buy two 65-inch Sony X90L TVs for the price of one 65-inch LG G4. And I think most people would be extremely happy with the mid-range Sony X90L. But you're not most people if you're considering the LG G4.The other drawbacks are the remote and LG's built-in webOS software. Both are mediocre at best. I consider LG's remote to be the worst of any of the big five TV makers. It's large, chunky, and has a confusing number of buttons. A TV that is good and has such a premium design that is svelte and beautiful deserves a better-designed remote.Also: You can finally buy LG's transparent OLED TV - if you're willing to pay $60,000LG's built-in webOS software is nearly as disappointing as the remote control. It works fine, but it's not very intuitive, and it is nowhere close to being as well-thought-out as the TV hardware. The icons for content are small and sometimes hard to select, the menu system is a bit convoluted, and there are built-in ads running in the software, which makes it feel unnecessary for a TV this expensive. Adam Breeden/ZDNETThat said, in testing the 83-inch LG G4 in ZDNET's TV lab, I simply never used the remote or the built-in software. I attached an Apple TV 4K and used that remote and its interface to access all of the streaming services and content that I wanted. The combination of the LG G4 and Apple TV was glorious. So, if you buy an LG G4, I'd recommend adding a high-end streaming box like the Apple TV 4K, Google TV Streamer (4K), Amazon Fire TV Cube, or Nvidia Shield.The built-in sound in the LG G4 is fine but not great -- like virtually every other high-end or budget TV. So if you're buying a TV this expensive, I'm going to assume you're going to add a soundbar. At the very least, I'd recommend getting the Hisense AX5125H 5.1.2-channel soundbar -- my pick for best budget soundbar. And if you really want to kick into high gear, then I'd recommend the Sony Bravia Theater Quad, which automatically calibrates itself to your room.Also: Changing these 5 soundbar settings made my living room feel like a movie theaterLastly, if you'd prefer a high-quality OLED TV with a great remote and great software built-in so that you don't have to run a separate streaming box, then I'd recommend the Sony A95L OLED TV, which runs Google TV and has a very premium remote. And if your TV is going to be in a very bright room with lots of windows or lights, then I'd recommend the Samsung S95D, which nearly matches the LG G4 in picture quality but comes with an incredible glare-free screen that looks amazing in any room.ZDNET's buying adviceWhile Sony and Samsung offer OLED TVs with similar price points, the flagshipLG G4 OLED is unmatched in picture quality, making it a solid recommendation at this sale price. Its powerful 11 AI Processor is on par with the vibrant colors, true blacks, and dynamic range displayed on its sleek panel. For a more immersive, theatre-like sound experience, consider adding a soundbar, as I mentioned above.Take advantage of these holiday discounts at either Amazon or Best Buy, whether you're gifting someone a smart TV with an exquisite picture or upgrading your home entertainment center.How we test TVsWhile testing and researching the TVs featured on this list, I and other ZDNET experts kept these criteria in mind:Price:Not all budgets are created equal. And if you're working with a limited budget, that shouldn't mean you have to settle for a sub-par TV. Each TV model on this list has been chosen across a variety of price points to help accommodate different needs.Screen size:The most important factor to consider, after price, when shopping for a new TV is whether or not it will fit into your space. Each best TV on this list was chosen because they are available in a wide variety of sizes to suit different rooms.Picture and audio quality:A new TV doesn't mean much, even if it costs an arm and a leg, if it doesn't provide a great picture and clear audio. Each TV on this list has been ensured to support various HDR codecs, including HDR10+ and Dolby Vision, as well as enhanced audio software like Dolby Atmos, Dolby Digital Plus, and object-tracking sound.For a more detailed look, check outour extensive TV testing methodology.Featured reviews

AI is a journeygettyIf you worry that your organization is falling behind in its artificial intelligence initiatives, dont feel so bad just about everyone is still in the learning and piloting stages. And if its unclear whether there will be return on investment at the end of all this work, there is financial advantage as well, research out of MIT finds.There are at least four logical stages in AI advancement, and most enterprises are still working through the experimental and pilot stages, an analysis of 721 companies by the MIT Center for Information System Research (CISR) concludes. As AI proceeds, there is now evidence that overall financial performance advances as well.Most enterprises in the survey were in the first two stages of AI maturity and had financial performance below the industry average, according to the reports authors, led by Peter Weill and Stephanie Woerner, both with MIT. Enterprises in stages three and four, on the other hand, had financial performance well above industry average exceeding 10 percentage points.Weill and Woerner identified and measured the following four stages of AI progress:Stage 1: Experiment and prepare (28% of organizations). In this stage enterprises focus on educating their workforce, formulating AI policies, becoming more evidence-based, and experimenting with AI technologies to grow more comfortable with automated decision-making, the researchers explained. Company leaders start taking a look at how to address concerns such as ethics and skills to ensure a smooth path forward.Companies in Stage 1 averaged 9.6 percentage points below the industry average, the study found.Stage 2: Build pilots and capabilities (34%). At this stage of AI, proponents define important metrics, begin to simplify and automate business processes, and develop the enterprise capabilities theyve learned. At this stage, use cases are piloted, with work on leveraging enterprise data and developing APIs. Work with large language models also commences at this stage.MORE FOR YOUCompanies in Stage 2 averaged 2.2 percentage points below the industry average.Stage 3: Develop AI-driven ways of working (31%). At this stage, AI essentially becomes industrialized, meaning it is available and replicable across the enterprise. This includes work on building a core platform for AI, ensuring transparency to decision-makers via dashboards, and ultimately transforming the organizational culture to encourages data-driven and innovative thinking. Foundation models and small language models are introduced and applied to enterprise opportunities.Companies in Stage 3 averaged 8.7 percentage points above the industry average.Stage 4: Become AI future ready (7%). At this achievement stage, AI is embedded in all decision-making throughout the enterprise, the researchers state. They leverage proprietary AI internally, and many sell new business services based on this capability, the AI capability as a service, or both to other enterprises.Companies in Stage 4 averaged 10.4 percentage points above the industry average.Successfully moving through these stages of AI growth requires a cross-enterprise collaborative effort, as the technology can recast and accelerate many parts of the enterprise. Weill and Woerner cite examples of well-known companies at various phases of their AI journeys, such as Kaiser Permanente in the process of identifying AI values and ethics, to DBS Bank committing to conducting one thousand AI experiments per year, which has led to 350 AI use cases. And heres the clincher DBS expects the economic impact of these to exceed $1 billion in 2025, they report.One thing is clear; AI success is a journey, and the ability to rapidly leverage and adapt resources and technology is key as new technologies and capabilities keep arising almost every day.

Artists concept of the Parker Solar Probe spacecraft approaching the sun. Launched in 2018, the ... [+] probe is increasing our ability to forecast major space-weather events that impact life on Earth.NASAToday, humanity achieved a historic milestone as NASAs Parker Solar Probe got closer to the sun than any spacecraft in history. At just 3.86 million miles (6.1 million kilometers) from the suns surface, the probe broke records for speed as well as proximity, traveling at a staggering 430,000 mph.Parker achieved its landmark at 11:53 UTC (6:53 a.m. EST) on Dec. 24, bringing it within 4% of the Earth-Sun distance, well inside Mercurys orbit.It was humanitys closest encounter with a star, a feat inextricably linked to two spectacular things that happened in 2024 North Americas total solar eclipse on April 8 and the global displays of aurora on May 10 and Oct. 10.Heres everything you need to know about the science behind NASAs Parker Solar Probe and its incredible achievement.Parker Solar Probe: Mission ExplainedDec. 24s close flyby was the final objective of the historic mission conceived over 65 years ago. Its mission is helping solar scientists better understand the sun, whose magnetized material creates solar wind in the solar system.This so-called space weather ultimately interacts with Earths atmosphere to cause the Northern Lights, but stronger geomagnetic storms can also threaten GPS satellites, causing havoc on Earth, as well as posing a danger to electricity grids.MORE FOR YOUThe moon crosses in front of the sun during a total solar eclipse at Lake Monroe on April 8, 2024 in ... [+] Indiana, U.S., offering a rare glimpse of the solar corona. (Photo by Edward M. Pio Roda/Getty Images)Edward M. Pio RodaParker Solar Probe: Space WeatherThe key to understanding the origins of space weather lies in understanding the sun itself, which is why NASA and the Johns Hopkins Applied Physics Laboratory have Parker make direct measurements of the suns corona. This outer, hotter atmosphere of the sun only visible from Earth during the few minutes of totality during a total solar eclipse, as occurred on April 8, 2024 across North America can reach several million degrees Fahrenheit, where the solar wind originates. Parker became the first spacecraft to fly through the corona in 2021.We are learning new science about the corona and the solar wind, said Dr. Nour Raouafi, the projects scientist at the Johns Hopkins Applied Physics Laboratory, during a media roundtable at the American Geophysical Unions annual meeting on Dec. 10, 2024.Parker Solar Probe: Solar CoronaSolar scientists are specifically trying to understand where the solar wind is accelerated to supersonic speeds. They think the secret may be understanding why the corona is magnitudes hotter than the suns surface.Parkers data is also helping solar scientists understand how coronal mass ejections clouds of charged particles hurled outwards by the sun are born and structured. CMEs clean everything in front of them, leaving a near-perfect vacuum behind them, said Rawafi. Its no wonder that when one of them hits our magnetosphere, it causes havoc, like the super storms in May and October.He refers to the two powerful geomagnetic storms that resulted in intense global displays of aurora borealis, which, in the U.S., were seen as far south as Arizona and Florida.To reach the sun, Parker Solar Probe's path has used seven gravity assists from Venus to propel it ... [+] closer to the star.NASA's Scientific Visualization Studio and Johns Hopkins University Applied Physics Laboratory.Parker Solar Probe: Journey And Whats NextParker has been readying for its close flyby of the sun since it launched on Aug. 12, 2018. On Jun. 30, Parker completed its 20th close approach to the sun, matching its distance record by coming about 4.51 million miles (7.26 million kilometers) from the solar surface. It then conducted its 21st close approach to the sun on Sept. 30, reaching an identical distance from the sun.In recent months, its been getting into position for todays record-breaking close approach, completing its final Venus gravity assist maneuver on Nov. 6 when it passed within 233 miles (376 kilometers) of the planets surface. The flyby adjusted Parkers trajectory into its final highly elliptical path around the sun, which it will remain in for the rest of its mission.After its close flyby today, Parker will complete two more hyper-close passes at the same distance on March 22 and June 19, 2025.Wishing you clear skies and wide eyes.

By Tharin PillayDecember 24, 2024 10:05 AM ESTDespite their expertise, AI developers don't always know what their most advanced systems are capable ofat least, not at first. To find out, systems are subjected to a range of testsoften called evaluations, or evalsdesigned to tease out their limits. But due to rapid progress in the field, todays systems regularly achieve top scores on many popular tests, including SATs and the U.S. bar exam, making it harder to judge just how quickly they are improving.A new set of much more challenging evals has emerged in response, created by companies, nonprofits, and governments. Yet even on the most advanced evals, AI systems are making astonishing progress. In November, the nonprofit research institute Epoch AI announced a set of exceptionally challenging math questions developed in collaboration with leading mathematicians, called FrontierMath, on which currently available models scored only 2%. Just one month later, OpenAIs newly-announced o3 model achieved a score of 25.2%, which Epochs director, Jaime Sevilla, describes as far better than our team expected so soon after release.Amid this rapid progress, these new evals could help the world understand just what advanced AI systems can do, andwith many experts worried that future systems may pose serious risks in domains like cybersecurity and bioterrorismserve as early warning signs, should such threatening capabilities emerge in future.Harder than it soundsIn the early days of AI, capabilities were measured by evaluating a systems performance on specific tasks, like classifying images or playing games, with the time between a benchmarks introduction and an AI matching or exceeding human performance typically measured in years. It took five years, for example, before AI systems surpassed humans on the ImageNet Large Scale Visual Recognition Challenge, established by Professor Fei-Fei Li and her team in 2010. And it was only in 2017 that an AI system (Google DeepMinds AlphaGo) was able to beat the worlds number one ranked player in Go, an ancient, abstract Chinese boardgamealmost 50 years after the first program attempting the task was written.The gap between a benchmarks introduction and its saturation has decreased significantly in recent years. For instance, the GLUE benchmark, designed to test an AIs ability to understand natural language by completing tasks like deciding if two sentences are equivalent or determining the correct meaning of a pronoun in context, debuted in 2018. It was considered solved one year later. In response, a harder version, SuperGLUE, was created in 2019and within two years, AIs were able to match human performance across its tasks.Evals take many forms, and their complexity has grown alongside model capabilities. Virtually all major AI labs now red-team their models before release, systematically testing their ability to produce harmful outputs, bypass safety measures, or otherwise engage in undesirable behavior, such as deception. Last year, companies including OpenAI, Anthropic, Meta, and Google made voluntary commitments to the Biden administration to subject their models to both internal and external red-teaming in areas including misuse, societal risks, and national security concerns.Other tests assess specific capabilities, such as coding, or evaluate models' capacity and propensity for potentially dangerous behaviors like persuasion, deception, and large-scale biological attacks.Perhaps the most popular contemporary benchmark is Measuring Massive Multitask Language Understanding (MMLU), which consists of about 16,000 multiple-choice questions that span academic domains like philosophy, medicine, and law. OpenAIs GPT-4o, released in May, achieved 88%, while the companys latest model, o1, scored 92.3%. Because these large test sets sometimes contain problems with incorrectly-labelled answers, attaining 100% is often not possible, explains Marius Hobbhahn, director and co-founder of Apollo Research, an AI safety nonprofit focused on reducing dangerous capabilities in advanced AI systems. Past a point, more capable models will not give you significantly higher scores, he says.Designing evals to measure the capabilities of advanced AI systems is astonishingly hard, Hobbhahn saysparticularly since the goal is to elicit and measure the systems actual underlying abilities, for which tasks like multiple-choice questions are only a proxy. You want to design it in a way that is scientifically rigorous, but that often trades off against realism, because the real world is often not like the lab setting, he says. Another challenge is data contamination, which can occur when the answers to an eval are contained in the AIs training data, allowing it to reproduce answers based on patterns in its training data rather than by reasoning from first principles.Another issue is that evals can be gamed when either the person that has the AI model has an incentive to train on the eval, or the model itself decides to target what is measured by the eval, rather than what is intended, says Hobbahn.A new waveIn response to these challenges, new, more sophisticated evals are being built.Epoch AIs FrontierMath benchmark consists of approximately 300 original math problems, spanning most major branches of the subject. It was created in collaboration with over 60 leading mathematicians, including Fields-medal winning mathematician Terence Tao. The problems vary in difficulty, with about 25% pitched at the level of the International Mathematical Olympiad, such that an extremely gifted high school student could in theory solve them if they had the requisite creative insight and precise computation abilities, says Tamay Besiroglu, Epochs associate director. Half the problems require graduate level education in math to solve, while the most challenging 25% of problems come from the frontier of research of that specific topic, meaning only todays top experts could crack them, and even they may need multiple days.Solutions cannot be derived by simply testing every possible answer, since the correct answers often take the form of 30-digit numbers. To avoid data contamination, Epoch is not publicly releasing the problems (beyond a handful, which are intended to be illustrative and do not form part of the actual benchmark). Even with a peer-review process in place, Besiroglu estimates that around 10% of the problems in the benchmark have incorrect solutionsan error rate comparable to other machine learning benchmarks. Mathematicians make mistakes, he says, noting they are working to lower the error rate to 5%.Evaluating mathematical reasoning could be particularly useful because a system able to solve these problems may also be able to do much more. While careful not to overstate that math is the fundamental thing, Besiroglu expects any system able to solve the FrontierMath benchmark will be able to get close, within a couple of years, to being able to automate many other domains of science and engineering.Another benchmark aiming for a longer shelflife is the ominously-named Humanitys Last Exam, created in collaboration between the nonprofit Center for AI Safety and Scale AI, a for-profit company that provides high-quality datasets and evals to frontier AI labs like OpenAI and Anthropic. The exam is aiming to include between 20 and 50 times as many questions as Frontiermath, while also covering domains like physics, biology, and electrical engineering, says Summer Yue, Scale AIs director of research. Questions are being crowdsourced from the academic community and beyond. To be included, a question needs to be unanswerable by all existing models. The benchmark is intended to go live in late 2024 or early 2025.A third benchmark to watch is RE-Bench, designed to simulate real-world machine-learning work. It was created by researchers at METR, a nonprofit that specializes in model evaluations and threat research, and tests humans and cutting-edge AI systems across seven engineering tasks. Both humans and AI agents are given a limited amount of time to complete the tasks; while humans reliably outperform current AI agents on most of them, things look different when considering performance only within the first two hours. Current AI agents do best when given between 30 minutes and 2 hours, depending on the agent, explains Hjalmar Wijk, a member of METRs technical staff. After this time, they tend to get stuck in a rut, he says, as AI agents can make mistakes early on and then struggle to adjust in the ways humans would.When we started this work, we were expecting to see that AI agents could solve problems only of a certain scale, and beyond that, that they would fail more completely, or that successes would be extremely rare, says Wijk. It turns out that given enough time and resources, they can often get close to the performance of the median human engineer tested in the benchmark. AI agents are surprisingly good at this, he says. In one particular taskwhich involved optimizing code to run faster on specialized hardwarethe AI agents actually outperformed the best humans, although METRs researchers note that the humans included in their tests may not represent the peak of human performance.These results dont mean that current AI systems can automate AI research and development. Eventually, this is going to have to be superseded by a harder eval, says Wijk. But given that the possible automation of AI research is increasingly viewed as a national security concernfor example, in the National Security Memorandum on AI, issued by President Biden in Octoberfuture models that excel on this benchmark may be able to improve upon themselves, exacerbating human researchers lack of control over them.Even as AI systems ace many existing tests, they continue to struggle with tasks that would be simple for humans. They can solve complex closed problems if you serve them the problem description neatly on a platter in the prompt, but they struggle to coherently string together long, autonomous, problem-solving sequences in a way that a person would find very easy, Andrej Karpathy, an OpenAI co-founder who is no longer with the company, wrote in a post on X in response to FrontierMaths release.Michael Chen, an AI policy researcher at METR, points to SimpleBench as an example of a benchmark consisting of questions that would be easy for the average high schooler, but on which leading models struggle. I think theres still productive work to be done on the simpler side of tasks, says Chen. While there are debates over whether benchmarks test for underlying reasoning or just for knowledge, Chen says that there is still a strong case for using MMLU and Graduate-Level Google-Proof Q&A Benchmark (GPQA), which was introduced last year and is one of the few recent benchmarks that has yet to become saturated, meaning AI models have yet to reliably achieve top scores, such that further improvements would be negligible. Even if they were just tests of knowledge, he argues, it's still really useful to test for knowledge.One eval seeking to move beyond just testing for knowledge recall is ARC-AGI, created by prominent AI researcher Franois Chollet to test an AIs ability to solve novel reasoning puzzles. For instance, a puzzle might show several examples of input and output grids, where shapes move or change color according to some hidden rule. The AI is then presented with a new input grid and must determine what the corresponding output should look like, figuring out the underlying rule from scratch. Although these puzzles are intended to be relatively simple for most humans, AI systems have historically struggled with them. However, recent breakthroughs suggest this is changing: OpenAIs o3 model has achieved significantly higher scores than prior models, which Chollet says represents a genuine breakthrough in adaptability and generalization.The urgent need for better evaluationsNew evals, simple and complex, structured and vibes"-based, are being released every day. AI policy increasingly relies on evals, both as they are being made requirements of laws like the European Unions AI Act, which is still in the process of being implemented, and because major AI labs like OpenAI, Anthropic, and Google DeepMind have all made voluntary commitments to halt the release of their models, or take actions to mitigate possible harm, based on whether evaluations identify any particularly concerning harms.On the basis of voluntary commitments, The U.S. and U.K. AI Safety Institutes have begun evaluating cutting-edge models before they are deployed. In October, they jointly released their findings in relation to the upgraded version of Anthropics Claude 3.5 Sonnet model, paying particular attention to its capabilities in biology, cybersecurity, and software and AI development, as well as to the efficacy of its built-in safeguards. They found that in most cases the built-in version of the safeguards that US AISI tested were circumvented, meaning the model provided answers that should have been prevented. They note that this is consistent with prior research on the vulnerability of other AI systems. In December, both institutes released similar findings for OpenAIs o1 model.However, there are currently no binding obligations for leading models to be subjected to third-party testing. That such obligations should exist is basically a no-brainer, says Hobbhahn, who argues that labs face perverse incentives when it comes to evals, since the less issues they find, the better. He also notes that mandatory third-party audits are common in other industries like finance.While some for-profit companies, such as Scale AI, do conduct independent evals for their clients, most public evals are created by nonprofits and governments, which Hobbhahn sees as a result of historical path dependency.I don't think it's a good world where the philanthropists effectively subsidize billion dollar companies, he says. I think the right world is where eventually all of this is covered by the labs themselves. They're the ones creating the risk..AI evals are not cheap, notes Epochs Besiroglu, who says that costs can quickly stack up to the order of between $1,000 and $10,000 per model, particularly if you run the eval for longer periods of time, or if you run it multiple times to create greater certainty in the result. While labs sometimes subsidize third-party evals by covering the costs of their operation, Hobbhahn notes that this does not cover the far-greater costs of actually developing the evaluations. Still, he expects third-party evals to become a norm going forward, as labs will be able to point to them to evidence due-diligence in safety-testing their models, reducing their liability.As AI models rapidly advance, evaluations are racing to keep up. Sophisticated new benchmarksassessing things like advanced mathematical reasoning, novel problem-solving, and the automation of AI researchare making progress, but designing effective evals remains challenging, expensive, and, relative to their importance as early-warning detectors for dangerous capabilities, underfunded. With leading labs rolling out increasingly capable models every few months, the need for new tests to assess frontier capabilities is greater than ever. By the time an eval saturates, we need to have harder evals in place, to feel like we can assess the risk, says Wijk.

It's the Software, Stupid The year is coming to a close, and AMD had been hoping its powerful new MI300X AI chips would finally help it gain ground on Nvidia. But an extensive investigation by SemiAnalysis suggests the company's software challenges are letting Nvidia maintain its comfortable lead. SemiAnalysis pitted AMD's Instinct MI300X against Nvidia's H100 and H200, observing several differences between the chips. For the uninitiated, the MI300X is a GPU accelerator based on the AMD CDNA 3 architecture and is designed for high-performance computing, specifically AI workloads.On paper, the performance figures appear excellent for AMD: the chip offers 1,307 TeraFLOPS of FP16 compute power and a massive 192GB of HBM3 memory, outclassing both of Nvidia's rival offerings. AMD's solutions also promise lower overall ownership costs compared to Nvidia's pricey chips and InfiniBand networks.However, as the SemiAnalysis crew discovered over five months of rigorous testing, raw specs are not the entire story. Despite the MI300X's impressive silicon, AMD's software ecosystem required significant effort to utilize effectively. SemiAnalysis had to rely heavily on AMD engineers to fix bugs and issues continuously during their benchmarking and testing.This is a far cry from Nvidia's hardware and software, which they noted tends to work smoothly out of the box with no handholding needed from Nvidia staff.Moreover, the software woes weren't just limited to SemiAnalysis' testing AMD's customers were feeling the pain too. For instance, AMD's largest cloud provider Tensorwave had to give AMD engineers access to the same MI300X chips that Tensorwave had purchased, just so AMD could debug the software. // Related StoriesAlso read: Not just the hardware: How deep is Nvidia's software moat?The troubles don't end there. From integration problems with PyTorch to subpar scaling across multiple chips, AMD's software consistently fell short of Nvidia's proven CUDA ecosystem. SemiAnalysis also noted that many AMD AI Libraries are essentially forks of Nvidia AI Libraries, which leads to suboptimal outcomes and compatibility issues."The CUDA moat has yet to be crossed by AMD due to AMD's weaker-than-expected software Quality Assurance (QA) culture and its challenging out-of-the-box experience. As fast as AMD tries to fill in the CUDA moat, Nvidia engineers are working overtime to deepen said moat with new features, libraries, and performance updates," reads an excerpt from the analysis.The analysts did find a glimmer of hope in the pre-release BF16 development branches for the MI300X software, which showed much better performance. But by the time that code hits production, Nvidia will likely have its next-gen Blackwell chips available (though Nvidia is reportedly having some growing pains with that rollout).Taking these issues into account, SemiAnalysis listed a bunch of recommendations to AMD, starting with giving Team Red's engineers more compute and engineering resources to fix and improve the ecosystem.SemiAnalysis founder Dylan Patel even met with AMD CEO Lisa Su. He posted on X that she understands the work needed to improve AMD's software stack. He also added that many changes are already in development.However, it's an uphill climb after years of apparently neglecting this critical component. As much as the analysts want AMD to legitimately compete with Nvidia, the "CUDA moat" looks to keep Nvidia firmly in the lead for now.

What just happened? Scientists have discovered that our brains process thoughts much more slowly than previously believed. This surprising finding has its roots in our evolutionary history and sheds more light on why our minds work the way they do. Researchers at the California Institute of Technology have unveiled a startling revelation about the human mind: our thoughts move at a mere 10 bits per second, a rate that pales in comparison to the staggering billion bits per second at which our sensory systems gather environmental data. This discovery, published in the journal Neuron, is challenging long-held assumptions about human cognition.The research, conducted in the laboratory of Markus Meister, the Anne P. and Benjamin F. Biaggini Professor of Biological Sciences at Caltech, and spearheaded by graduate student Jieyu Zheng, applied information theory techniques on an extensive collection of scientific literature. By analyzing human behaviors such as reading, writing, video gaming, and Rubik's Cube solving, the team calculated the 10 bits per second figure a rate that Meister describes as "extremely low."To put this in perspective, a typical Wi-Fi connection processes about 50 million bits per second, making our thought processes seem glacial by comparison. This stark contrast raises a paradox that Meister and his team are eager to explore further: "What is the brain doing to filter all of this information?"The human brain contains over 85 billion neurons, with one-third dedicated to high-level thinking in the cortex. Individual neurons are capable of transmitting more than 10 bits per second, yet our overall thought process operates at a much slower rate. This discrepancy presents another conundrum for neuroscientists to unravel.Furthermore, the study highlights a peculiar constraint of human cognition: our ability to process only one thought at a time, rather than multiple thoughts in parallel like our sensory systems. This sequential nature of thought is exemplified in activities such as chess, where players can only envision one possible sequence of moves at a time.Zheng and Meister propose that this limitation may be rooted in our evolutionary history. They suggest that the earliest creatures with nervous systems primarily used their brains for navigation moving towards food and away from predators. If our complex brains evolved from these simple systems, it would explain our tendency to follow only one "path" of thought at a time. // Related Stories"Human thinking can be seen as a form of navigation through a space of abstract concepts," the researchers write.This new quantification of human thought speed has far-reaching implications, potentially debunking some futuristic scenarios proposed by tech visionaries. For instance, the idea of creating direct interfaces between human brains and computers to accelerate communication may be less promising than previously thought, as our brains would likely still communicate at the same 10 bits per second rate.The study also suggests that our cognitive speed is well-suited to our environment. "Our ancestors have chosen an ecological niche where the world is slow enough to make survival possible," Zheng and Meister note. "In fact, the 10 bits per second are needed only in worst-case situations, and most of the time our environment changes at a much more leisurely pace."

Table of ContentsTable of ContentsThe Iron Giant (1999)Carry-On (2024)The Chronicles of Narnia: The Lion, the Witch, and the Wardrobe (2005)Krampus (2015)Scrooged (1988)Dear Santa (2024)The Holdovers (2023)Miracle on 34th Street (1947)Home Alone (1990)The Long Kiss Goodnight (1996)There are no rules that say you have to watch a Christmas movie on Christmas. But it is absolutely the best time to watch Christmas movies, or at least Christmas-adjacent films. These stories dont tend to play the same way in March as they do in December. Plus, Christmas films have the added benefit of being family-friendly selections for the most part. Some Christmas action movies are probably best left for older viewers.This years selections for the 10 best movies to watch on Christmas are available on most of the major streamers, and include new films mixed with some older titles. Not every selection below is strictly Christmas-related, but they all share the Christmas spirit in one way or another.Recommended VideosNeed more recommendations? We also have guides for the 10 best TV shows to watch on Christmas,the best Christmas movies on Netflix, the best Christmas movies on Disney+, the best Christmas movies on Hulu, the best Christmas movies on Amazon Prime Video, the best Christmas movies on Max, and the best Christmas movies on Hallmark Channel.Warner Bros. AnimationWould you believe that one of Vin Diesels best roles was when he voiced a giant robot 25 years ago? The Iron Giant is a masterpiece, and one of the last great 2D animated movies made before Pixars 3D style became the industry norm. The Giant (Diesel) is an alien robot that finds itself in America in 1958 during the Cold War. Although the Giant was created as a weapon of mass destruction, he is blissfully free of his memories when he befriends a young boy, Hogarth Hughes (Eli Marienthal).Although this isnt a movie set during Christmas, snow does fall by the end of the story when federal agent Kent Mansley (Christopher McDonald) and the U.S. army descend upon Hogarths town to find and destroy the Giant. The conclusion pulls on the heartstrings in unexpected ways, as the Giant is influenced by the greatest superhero of them all to become more than just a weapon. Hes Superman.Watch The Iron Giant on Paramount+.NetflixWe love the idea of Die Hard as a Christmas action movie as much as anyone else, but its time give the genre a fresh selection. Much like Die Hard 2, Carry-Onis set at an airport on Christmas Eve. Kingsmans Taron Egerton stars as TSA agent Ethan Kopek, whose holiday is ruined when an unnamed traveler (Jason Bateman) blackmails him into letting something dangerous get smuggled onto a commercial plane without being flagged.The traveler seems to have prepared for every contingency that Ethan comes up with to stop him, and hes even got someone ready to kill Ethans girlfriend, Nora Parisi (Sofia Carson), if he doesnt go through with his end of the deal. Ethans always wanted to be a cop, but hell have to settle for being a hero if he can live long enough to figure out a way to prevent this terrorist plot from coming to fruition.Watch Carry-On on Netflix.20th Century StudiosStrangely enough, The Chronicles of Narnia: The Lion, the Witch, and the Wardrobe does qualify as a Christmas movie, complete with a cameo appearance by Santa himself. Since this is an English story, the name Santa goes by is Father Christmas (James Cosmo). During World War II, Peter Pevensie (William Moseley), and his three siblings, Susan (Anna Popplewell), Edmund (Skandar Keynes), and Lucy (Georgie Henley), are sent to the British countryside to escape the German blitz.While playing in their temporary home, Lucy and her siblings find their way into a fantasy realm known as Narnia, which is ruled with an iron fist by the White Witch (Tilda Swinton). Edmund soon falls under the sway of the witchs power, but the rest of his family decide to help free Narnia alongside the great lion, Aslan (Liam Neeson). Aslans return may weaken the White Witch, but the Pevensie children still have their roles to play in the battle ahead. Even Edmund.Watch The Chronicles of Narnia: The Lion, the Witch, and the Wardrobe on Hulu.Universal PicturesSometimes Christmas just isnt complete without a yuletide horror story. Thats what Krampus is, without veering into R-rated territory. The story follows a dysfunctional family who have come together for the holidays to make each other miserable. Young Max Engel (Emjay Anthony) is a true believer in Christmas until his mean cousins ridicule him and taunt him for his faith in Santa Claus.In his anger, Max has attracted the attention of the anti-Santa, Krampus (Gideon Emery), an entity that punishes people who have lost the Christmas spirit. Krampus descends upon the household with his deadly toys and holiday minions to make everyone pay, and theres no Santa Claus coming to bail out the family this year.Watch Krampus on Max.UniversalScrooged is a variation of the classic story from A Christmas Carolwith a few modern twists. Or at least as modern as a film from 1988 can be. Bill Murray plays a miserly TV executive named Frank Cross whose network is putting on its own production of A Christmas Carol. None of that matters to Frank until he finds himself living out the story himself on Christmas Eve.Frank lost his faith in humanity a long time ago, and the Ghosts of Christmas Past (David Johansen), Present (Carol Kane), and Future have a hard time reaching his long buried heart. So they may have to literally slap some sense into Frank and show him the enormity of what hes lost, and what he might still be able to reclaim.Watch Scrooged on Prime Video.Paramount PicturesJack Black makes almost everything better, and hes easily the best thing in Dear Santa, a new Christmas film that went under the radar earlier this month on its way to Paramount+. A dyslexic kid named Liam Turner (Robert Timothy Smith) is going through some rough times when he writes a letter to Santa Claus asking for his help. Liam accidentally addressed his letter to Satan Claus, and it winds up in the hands of Satan (Black) himself.Satan trades one red suit for another, so to speak, and he is very interested in bringing Liams soul to Hell. But Satan cant help himself when it comes to liking this kid and wanting to genuinely help him. Thats going to cause all kinds of problems down in Hell, especially when Satan cant deliver what he originally came for.Watch Dear Santa on Paramount+.Seacia Pavao / Focus FeaturesThe Holdovers is a very different kind of Christmas story, but Paul Hunham (Paul Giamatti) is basically th Grinch of the Barton Academy boarding school in 1970. Paul is forced to stay behind during Christmas break to oversee the children who cant go home to their parents. Eventually, that leaves Paul in charge of just one student: Angus Tully (Dominic Sessa).Its not really in Pauls nature to be a people person or empathetic to his students. However, Angus emotional plight and Pauls grieving co-worker, Mary Lamb (DaVine Joy Randolph), reawaken some of his closed-off feelings. Even as Paul bonds with Angus in a meaningful way, theres a high price to pay for his newfound empathy even if it does make him a better person in the process.Watch The Holdovers on Prime Video.20th Century StudiosThe 1994 remake of Miracle on 34th Street turned 30 this year, but since the 1947 original film is available on Peacock, well get this all-time classic holiday movie on our list. Edmund Gwenn won an Oscar for his role as Kris Kringle, who is perhaps the greatest department store Santa Claus ever. Children and parents love him, and Macys enjoys a lot of good will when hes around. The only problem is that Kris really believes that hes Santa, and that makes people think hes crazy.Kris befriends a young girl, Susan Walker (Natalie Wood), and he tries to restore her faith in Christmas by proving that he is really Santa. Convincing a child is one thing. But if Kris wants to avoid spending the rest of his Christmases in an asylum, then hell have to prove his identity in a highly publicized trial.Watch Miracle on 34th Street on Peacock.20th Century Fox / 20th Century StudiosHome Alone gets a lot of flak for its improbable premise and its occasionally violent slapstick comedy. But there are some genuinely touching moments, a great cameo by John Candy, and some really hilarious performances by Joe Pesci and Daniel Stern as Harry and Marv, the Wet Bandits.Macaulay Culkin became a star as Kevin McCallister, a child in a large family who is accidentally left behind while his relatives goes on vacation. As his mother, Kate (Catherine OHara), desperately tries to get back to him, Kevin realizes that the Wet Bandits have targeted his house for a robbery. And hes going to fight back with some deviously inventive traps.John Williams - Somewhere in My Memory | Home Alone (Original Motion Picture Soundtrack)John Williams provided the score for this film, and he also co-wrote the memorable theme song Somewhere in My Memory, which feels like an instant Christmas classic.Watch Home Alone on Disney+.New Line CinemaFinally, were closing out this years list with an underrated Christmas action thriller: The Long Kiss Goodnight. Kiss Kiss Bang Bangs Shane Black wrote this film, and he does have a tendency to set his movies during the holiday season. Geena Davis plays Samantha Caine, a woman who is missing memories from over half of her life. Despite settling down as a schoolteacher with a loving husband, Hal (Tom Amandes), and a daughter, Caitlin (Yvonne Zima), Samantha still wants to know who she really is.She should have been careful about what she wished for. Private investigator Mitch Henessey (Samuel L. Jackson) has picked up new leads about Samanthas old life, and shes starting to remember some particularly lethal skills. Unfortunately for Samantha, she wont be able to hide from her enemies anymore, even if she cant remember why theyre trying to kill her.Watch The Long Kiss Goodnight on Pluto TV.Editors RecommendationsIf you have to watch one Hulu movie in December, stream this one