Category Featured Videos

People aren't thrilled with Paramount these days. After all, corporate consolidation and the transformation of CBS News into state media tend to do that. But here's someone who may not have chosen the… wisest form of protest. The Straits Times reports (via Gizmodo) that Singapore police have arrested a 26-year-old man for leaking Paramount's The Legend of Aang: The Last Airbender.The film, part of the Avatar: The Last Airbender franchise, was initially scheduled for a theatrical release. But after two delays, it was later moved to Paramount+, where it’s set to debut later this year. Naturally, hardcore fans of the franchise bristled at the prospect of the highly anticipated entry skipping theaters. Paramount's scant marketing apparently didn't help either.However, many have already seen the movie, allegedly thanks to the unnamed 26-year-old in question. Singapore police said he gained remote access to a server and downloaded the film. He then posted parts of it online. After the leak, Paramount changed the movie's name to Avatar Aang: The Last Airbender — perhaps to try to prevent casual fans' searches from turning up info about the leak.The man is being investigated for unauthorized access to computer material. He could face a maximum of seven years in prison, a fine of up to $50,000, or both.Earlier this month, The Hollywood Reporter interviewed the X account holder @ImStillDissin, who posted the clips. (Although we can speculate that this may be the man currently in custody, that’s unconfirmed.) The interview revealed a rather, shall we say, blasé approach to the incident. He said he figured posting clips from the movie was no biggie since the film is a streaming-only release. "I saw it's just a Paramount+ thing, so I decided I'd troll a little bit," the leaker said.The leaked clips spread rapidly. Despite pleas from 4Chan posters to share the entire film, @ImStillDissin resisted. However, someone else shared the full movie by April 13. Naturally, that file has since circulated far and wide.So, good luck with that official October 9 streaming release, Paramount. You're gonna need it.This article originally appeared on Engadget at https://www.engadget.com/entertainment/streaming/singapore-police-arrest-alleged-the-legend-of-aang-the-last-airbender-leaker-183954803.html?src=rss

The Apple rumors were true, once again. This week, the company announced that Tim Cook will be stepping down from his CEO role on September 1. Replacing him will be John Ternus, who currently serves as Apple's SVP of hardware engineering. In this episode, Devindra and Engadget's Nathan Ingraham discuss Cook's legacy as Apple's CEO, and pontificate about how Ternus may change things. We're going from Apple being led by a logistics guru, to Apple being driven by a product and engineering wizard. Surely, that will have some impact on future products.Subscribe!iTunesSpotifyPocket CastsStitcherGoogle PodcastsTopicTim Cook to step down as Apple CEO after 15 years, John Ternus will take his place on September 1 – 1:22Palantir woke up last Saturday morning and posted a comic book villain manifesto on X – 26:01DHS wants to make facial recognition smart glasses for ICE – 31:53A lot of people panic bought PCs to avoid RAMageddon – 36:25Meta faces a new lawsuit over running ads for outright scams –Employees at Meta will have they keystrokes and mouse moves recorded for AI training – 40:10Xbox Game Pass Ultimate price goes down, but it won’t include Call of Duty – 44:55Around Engadget: a great (expensive) Dyson vac with a silly name – 49:15Working on – 51:58Pop culture picks – 52:55CreditsHosts : Devindra Hardawar and Nathan IngrahamProducer: Ben EllmanMusic: Dale North and Terrence O’BrienThis article originally appeared on Engadget at https://www.engadget.com/computing/engadget-podcast-tim-cooks-apple-era-and-what-lies-ahead-for-john-ternus-121853488.html?src=rss

DeepSeek has released its latest AI models, the V4 Pro and Flash versions, a bit over a year after it went viral and became the top rated free app on Apple's App Store in the US. “Welcome to the era of cost-effective 1 million context length,” DeepSeek said in its announcement. Context length is what you call the maximum number of tokens that an AI model can remember, so the bigger it is, the more coherent and consistent an AI is when it comes to extended conversations. OpenAI’s recently announced GPT‑5.5 has a context window ranging from 400,000 to 1 million, for instance.The new model is still open-source, allowing users to download its code and modify it if they want. DeepSeek says V4 Pro has enhanced agentic capabilities and claims that it rivals top closed-source models when it comes to reasoning. It also says that it trails only Gemini-3.1-Pro in rich world knowledge. Meanwhile, V4 Flash isn’t quite as powerful as the V4 Pro, but it has faster response times. Still, its reasoning abilities closely approach V4 Pro, DeepSeek says, and it performs on par with with the Pro version on simple Agent tasks. Shortly after DeepSeek topped the App Store charts, it was banned for use by US federal agencies and on government-owned devices. Authorities believed it was a national security risk and posed a threat to US AI stocks. South Korea also paused downloads of its app over privacy concerns. 🚀 DeepSeek-V4 Preview is officially live & open-sourced! Welcome to the era of cost-effective 1M context length.🔹 DeepSeek-V4-Pro: 1.6T total / 49B active params. Performance rivaling the world's top closed-source models.🔹 DeepSeek-V4-Flash: 284B total / 13B active params.… pic.twitter.com/n1AgwMIymu— DeepSeek (@deepseek_ai) April 24, 2026

This article originally appeared on Engadget at https://www.engadget.com/ai/deepseek-promises-its-new-ai-model-has-world-class-reasoning-115733512.html?src=rss

Although it’s one of the more inoffensive topics on Polymarket, this news typifies the Wild West of prediction markets and betting sites. A hairdryer was allegedly used to rig Polymarket bets on temperatures at Charles de Gaulle Airport in Paris, according to a report by The Telegraph. French authorities noted that the official temperature readings at the airport spiked twice in the past month. On both occasions, gamblers betting on those temperature fluctuations on Polymarket appear to have walked away with thousands upon thousands of dollars.
There is no indication that Polymarket forced anyone to return winnings, but the temperature sensor has been moved to a new location. The site is also still running bets on the daily temperature in and around Paris.
In a more serious development, a US soldier was arrested for allegedly making over $400,000 on Polymarket using information he had about the plans to capture the former Venezuelan president, Nicolás Maduro.
Gannon Ken Van Dyke was arrested and charged with using classified military information to place bets on the prediction marketplace Polymarket. Van Dyke created a Polymarket account around December 26, 2025, and made 13 bets related to Maduro from December 27 to January 2.
The soldier has also been charged with one count of wire fraud, carrying a maximum penalty of 20 years in prison, and one count of unlawful monetary transaction, carrying a maximum sentence of 10 years. It’s a lot heavier than hairdryer shenanigans.
— Mat Smith
The other big stories (and deals) this morning

Here’s to the stable ones: In praise of Tim Cook
Meta is downsizing by about 10 percent
Hey Meta workers, are you getting paid for those keystrokes?
Apple TV’s upcoming For All Mankind spinoff Star City oozes Cold War-era paranoia

DJI Lito 1 and Lito X1 drone review
High-quality aerial video at its most affordable.
Engadget
DJI is taking another stab at the budget drone market with the new Lito series. The Lito 1 and Lito X1 are both under $400 and weigh less than 249 grams — they’re ideal for beginners. Both replace DJI’s Mini series, but they offer things those models lacked, like LiDAR and 360-degree obstacle avoidance. After testing both models, I believe they offer unbeatable value and performance at these prices, by a long shot. However, due to DJI’s standing in the US, you might not see either.
Continue reading.

Xbox cuts Game Pass prices
But new Call of Duty games will no longer hit the service at launch.
Activision
As suggested by recent comments by the new boss of Xbox, Microsoft’s gaming arm is cutting the prices of both Game Pass Ultimate and PC Game Pass, effective immediately, but there’s one big caveat. New Call of Duty games will no longer be available on Game Pass Ultimate or PC Game Pass on day one. They’ll eventually hit those tiers about a year later, during the following holiday season.
Continue reading.

Accessory maker Anker made its own AI chip
Of course it did.

Anker, of battery-pack and cable fame, has announced its own AI chip that it will integrate into its future headphones and other devices. The company is planning to debut the chip, called Thus, on a new model of headphones to be unveiled at its Anker Day event in May.
Anker’s Thus chip integrates computing power directly into NOR flash memory cells, which offer faster read speeds than NAND. Anker says headphones are a particularly challenging environment to demonstrate what a new chip can do because “hardly any other device places higher demands on an AI chip.” Anker announced one particular feature to showcase its silicon. Clear Calls will cancel noise “with a large neural network running entirely on the device, supported by eight MEMS microphones and two bone conduction sensors.”
Continue reading.This article originally appeared on Engadget at https://www.engadget.com/general/the-morning-after-engadget-newsletter-112802570.html?src=rss

The Turkish parliament has voted through a bill that would ban all children under the age of 15 from using social media. As part of the legislation, social media platforms would be required to enforce age-verification measures on their apps, provide parental control tools, and react more quickly to harmful content being posted.
As reported by The Associated Press, lawmakers have passed the bill in the wake of two deadly school shootings in Turkey, after which police arrested 162 people accused of sharing footage of the tragedies online.
Turkey’s President Recep Tayyip Erdogan now has 15 days to accept the bill in order for it to become law, after reportedly saying social media platforms had become "cesspools" in a televised address to the nation.
As well as the major social media platforms, AP reports that online gaming companies would also have to implement their own restrictions on minors, with potential punishments including bandwidth reductions and financial penalties.
This isn’t the first time Turkey has locked horns with social media and online gaming platforms. Instagram has been blocked in the country before, back in 2024, relating to a dispute over the posting of Hamas-related content. Access was restored around a week later, but in the same time period Turkey also banned Roblox over reports of inappropriate sexual content accused of being explorative to children. At the time, a Turkish official also named the "promotion of homosexuality" as one reason for the ban.
Turkey has also temporarily banned Twitter (now called X) on several occasions, most recently after 2023’s devastating earthquakes, though it was not clear at the time why the government may have moved to block the social media platform.
The country’s lawmakers moving to ban under-15s from accessing social media is part of an emerging trend in Europe and across the globe. The likes of Greece and Austria have recently introduced similar legislation of their own, following Australia becoming the first country in the world to ban children under 16 from social media last year. The UK has since considered bringing in tighter restrictions too.This article originally appeared on Engadget at https://www.engadget.com/social-media/turkey-wants-to-ban-social-media-for-kids-under-15-143053462.html?src=rss

With countries banning social media for kids left and right, Meta is trying different things to convince parents that its platforms are safe for teens. In its latest effort, the company will start showing parents the topics their teens have discussed with Meta AI over the previous seven days. 
"Parents will be able to see the topics their teen has been asking Meta AI about in [Facebook, Messenger or Instagram] over the past week," Meta explained in a blog post. "Topics can range from School, Entertainment, and Lifestyle to Travel, Writing, and Health and Wellbeing, among others."
For parents overseeing Meta's teen accounts, the feature will appear in a new Insights tab within supervision, both in-app and on web. Parents can tap on a topic to see the different categories within each: for instance, sub-categories within Lifestyle include fashion, food and holidays, while fitness, physical health and mental health are part of the Health and Wellbeing topic. 
Meta
Meta also worked with the Cyberbullying Research Center to develop what it calls "conversation starters," or open-ended conversations about their experience with AI. It provides detail about what the questions are designed to address, and can be found on the Family Center website or through a link in the new Insights tab. 
Finally, Meta revealed more detail about its AI Wellbeing Expert Council, who will provide "ongoing input on our AI experience for teens." It will be made up of three existing advisory groups as well as new members with special expertise in responsible and ethical AI, who are affiliated with the National Council of Suicide Prevention and multiple universities. It's worth noting that Meta has a separate oversight board that deals with subjects ranging from AI to moderation. 
Offboarding moderation chores to busy parents appears to be par for the course for Meta these days. The company has recently cut back on the use of third-party vendors that help with content moderation, shifting responsibility instead to advanced AI systems, according to recent reports. 
The dangers of AI for teens have been one of multiple reasons countries like Spain have banned social media platforms for kids. One of the most recent and tragic cases was in Canada, where a teen was provided specific details by OpenAI's ChatGPT about how to carry out a school shooting. Another such case is under investigation in Florida, and AI's have been implicated in multiple teen suicides as well. 
In the US, the National Suicide Prevention Lifeline is 1-800-273-8255 or you can simply dial 988. Crisis Text Line can be reached by texting HOME to 741741 (US), 686868 (Canada), or 85258 (UK). Wikipedia maintains a list of crisis lines for people outside of those countries.This article originally appeared on Engadget at https://www.engadget.com/ai/meta-will-show-parents-the-topics-of-their-teens-ai-conversations-123119624.html?src=rss

After seeing the runaway success of its Neo lineup, DJI is taking another stab at the budget drone market with the new Lito series. The Lito 1 and Lito X1 are both under $400 and weigh less than 249 grams — they’re ideal for beginners. Both are designed to replace DJI’s Mini series, but they offer things that those models lacked like LiDAR and 360-degree obstacle avoidance.
In terms of video quality, they’re similar to DJI’s Neo 2 and Flip drones. Unlike those models, which are safe to use indoors and around people, the open-prop Lito drones are designed to fly outside at high speeds and high altitudes. This makes them well-suited for filming activities like surfing or dirt bike riding. After testing both models, I believe they offer unbeatable value and performance at these prices, by a long shot. Of course, the drawback for American buyers is that neither is expected to launch in the US.

Design and features
The Lito models are now the entry level open-prop models in DJI’s lineup (the Mini series will no longer be updated). Both have identical folding designs, with optional, removable propeller shrouds that offer some additional obstacle protection. Given the light weight, novice pilots can fly them without a permit.

Steve Dent for Engadget

Both have omni sensors on the top and bottom that protect them from obstacles on all sides. The Lito X1 adds a forward-facing LiDAR sensor for extra tracking precision. That kind of additional accuracy is nice to have and it’s unusual on drones in this price range. Built-in storage is also rare on cheap drones, but the Lito X1 has that going for it as well. There’s 42GB on tap, along with microSD card support for additional capacity. The Lito 1 only offers a microSD slot and no internal storage.
They come with an Intelligent Flight Battery that offers up to 36 minutes of flying time. That can be extended to about 52 minutes with the Intelligent Flight Battery Plus, though that model isn’t available in the EU or UK. As with every other DJI drone I’ve tested, real-world flight times are about 30 percent less than the company promises. I got about 25 minutes of use on a charge — still not bad for a sub-$400 drone.
The Lito line has all the FocusTrack features you’d expect in a DJI drone, like 360 degree ActiveTrack subject tracking with full manual control and obstacle detection. It also supports DJI’s Quickshots including Dronie, Rocket and others, all of which make it easy to capture aerial clips for social media. However, you won’t find certain Neo 2 features on the Lito series, like gestures, smartphone control and palm takeoffs.
The Lito drones work with multiple controllers, including the RC-N3 that requires a smartphone or the RC 2 with a built-in screen. Both Lito models transmit 1080p 60 fps video to a range of up to 10 miles via DJI’s Occusync 4 system. That’s a scary distance for a beginner drone. Luckily, both models also offer DJI’s return to home (RTH) with battery warning and other safety features that should prevent lost drones or accidents.
Performance

Steve Dent for Engadget

Both Lito drones are relatively fast at about 26 mph in normal mode with obstacle detection, or 40 mph in sport mode. That kind of speed allows creators to track bikers, skiers or vehicles. Each can also fly in relatively strong winds up to about 21 mph. They lack the maneuverability and acrobatics available on the Neo 2 though, and don't offer an FPV mode with DJI’s Goggles.
With open propellers that can catch on a twig and cause a crash, the Lito drones depend heavily on obstacle detection for protection. That’s particularly important since the ActiveTrack follow feature is a key selling point for novice creators. With all that in mind, I wanted to put them through their paces tracking me on a bicycle and even while driving a car, forcing the drones to navigate around trees and bamboo with fine branches.
To start, I used ActiveTrack and the Trace “steering wheel” mode to follow me from the front, sides and back. Both Lito models can avoid obstacles using either the “braking” mode that stops the drone or “bypass” that simply maneuvers around them. I mostly used bypass mode to see if the drone could continue to track me if something got in the way.
When flying forward, the Lito X1 avoided all obstacles using its LiDAR, while swooping smoothly around trees and branches. Only once did it fail to detect a small twig, which caused a slight bobble, but fortunately, no crash. Thanks to that built-in LiDAR, the Lito X1 model is a bit more adept than the Lito 1 at dodging fine obstacles when flying forward. Overall, the X1’s avoidance in all directions was shockingly good for a drone under $400.

DJI's Lito models offer excellent tracking and obstacle detection for the price.

Steve Dent for Engadget

The Lito X1’s LiDAR is also very helpful when flying in low-light conditions, and even at night. If you go on a long flight and miscalculate the sunset, you’ll still be able to get the drone back safely in RTH mode without much fear of crashing, as long as you fly forward. It also helps keep the Lito X1 more stable in dim light than the Lito 1.
The Lito 1 isn’t as adept at following and obstacle detection at night due to the lack of LiDAR, but again, it avoided danger surprisingly well considering the price. Buyers can still be confident that it will stay out of trouble when tracking subjects in most conditions — just be careful when flying in forested areas with fine branches.
If you’re planning on flying far from your location to capture remote aerial views, DJI’s Occusync 4 system is highly reliable. I only saw video dropouts when flying the drone behind buildings and out of my line of sight, something that I wouldn’t advise anyway. As long as there’s nothing between you and the drone, you’ll maintain a clear video view and full control.
As with other drones in the Mini and Mini Pro series, the Lito models are whisper quiet in flight, with noise levels well under 70 db. The pitch is also low and not shrieking like the Neo models, so it shouldn’t disturb people or animals.
Video and photo quality
Video quality is where the Lito 1 and Lito X1 differ the most. The latter is equipped with a 40-megapixel 1/1.3-inch sensor with f/1.7 aperture (same as the Mini 4 Pro) and offers up 4K 60 fps video with 4K 100 fps slow-mo along with D-LogM and HDR recording. The Lito 1, meanwhile, has a 48MP 1/2-inch sensor (12MP for video) like the one on the Neo 2, with an f/1.8 aperture, 4K 60 fps video (4K 100 fps slow-mo) and no D-LogM or HDR capabilities.

That’s a pretty big gap in specs considering the relatively meager price difference of around $80. As such, the Lito X1 offers significantly better video quality, particularly in low light and high-contrast conditions.
The smaller sensor means the Lito 1 has mediocre low-light capability, with pronounced grain at the maximum ISO 12,800 rating. The drop in quality is even noticeable at ISO 3,200. Unlike the $400 Flip, the lack of 10-bit D-LogM capability also means that over- or underexposed video is hard to correct.
With those issues, video and photos from the Lito 1 aren't good enough for professional work. However, it’s excellent for social media users, hobbyists and content creators, delivering smartphone-quality aerial shots and more detail than any other drone at this price (except the Neo 2, of course).
The Lito X1, on the other hand, does offer video that’s good enough for some professional use. While not as noise-free as the 1-inch sensor-equipped Mini 5 Pro or Micro Four Thirds Mavic 4 Pro, video quality for nighttime cityscapes and other dim scenes is decently clean. In daylight, video is surprisingly sharp and color accurate. The 10-bit D-LogM setting also lets you capture sufficient dynamic range for tricky scenarios like a forest path dappled with sunlight and shadows
Stabilization on both models is excellent, so you’re sure to capture smooth, cinematic video, even with relatively slow shutter speeds down to 1/30th of a second. Photos are more detailed on the Lito X1 due to the higher resolution, even though both models can capture RAW DNG files to maximize dynamic range. That makes the drones ideal for taking aerial pictures of your property, for instance, or checking a hornet’s nest or hole on your roof.
Wrap-up

Steve Dent for Engadget

The Lito series shows that DJI is intent on dominating every drone price range and category, including the bottom end. Despite their low prices, the new drones don’t skimp on features, offering full obstacle protection, ActiveTrack subject tracking, relatively high speeds and sharp 4K video quality — just like models that cost a lot more.
At these prices, the Lito drones don’t have any real rivals other than themselves and other DJI drones, particularly the Neo 2 and Flip. Choose the Lito 1 only if you can’t swing the extra money for the Lito X1, as video quality on the latter is significantly higher. Both drones are best for outdoor adventures, including high-altitude aerial shooting, while Neo 2 and Flip excel inside or around people.
The Lito 1 and Lito X1 are now available in the UK and EU for £299/€309 and £369/€379 respectively with one battery and the RC-N3 controller that requires a smartphone. You can also get them in Fly More combos, with the Lito 1 priced at £429/€439 with an RC-N3 controller, three batteries, a charger and a shoulder bag. The Lito X1 Fly More Combo with an RC 2 screen controller, three batteries, a charger and a shoulder bag is £599/€619. The drones aren’t on sale yet in the US, but may arrive later on.This article originally appeared on Engadget at https://www.engadget.com/cameras/dji-lito-1-and-lito-x1-drone-review-high-quality-aerial-video-at-its-most-affordable-120024032.html?src=rss

Tesla still doesn’t have a solid pathway for how to give Hardware 3 cars Full Self-Driving (FSD) capabilities, based on what Elon Musk said during the company’s latest earnings call. The automaker has known for quite a while that its vehicles equipped with Hardware 3 aren’t capable of unsupervised FSD, even though the company built the system specifically to give its cars the ability. Tesla used Hardware 3 on cars manufactured from 2019 until early 2023 before Hardware 4 shipped. During an earlier earnings call back in January 2025, Musk admitted that the company was “going to have to upgrade people’s Hardware 3 computer for those that have bought Full Self-Driving.” At Tesla’s latest earnings call, Musk said that Hardware 3 “simply does not have the capability to achieve unsupervised FSD.” Tesla thought it would be able to at one point, but Hardware 3 apparently has 1/8th the memory bandwidth of Hardware 4. Musk explained that memory bandwidth is “one of the key elements” needed for unsupervised FSD.Tesla will be offering to upgrade and replace the computers and cameras on older vehicles, but it doesn’t have a concrete plan in place yet. “I do think over time it’s going to make sense for us to convert all Hardware 3 cars to Hardware 4,” he said. To do so at service centers would be extremely slow, Musk has admitted. Around 4 million cars or so have Hardware 3, though not everyone has paid for FSD. Still, to be able to replace its vehicles’ hardware efficiently, Musk said Tesla is going to have to set up “microfactories or small factories in major metropolitan areas.” He didn’t give any indication that Tesla has already started building those microfactories, though, or even that construction is already scheduled to begin. He did say that in the meantime, the company is going to be releasing FSD version 14 for Hardware 3 around the end of June. Musk also said during the same earnings call that Tesla’s Fremont factory will start manufacturing the company’s humanoid Optimus robots in late July or August. The Tesla CEO is known for announcing highly optimistic and aggressive timelines. Tesla made the decision to kill off its Model X and S cars earlier this year, so that it can convert its Fremont facility into an Optimus factory. The last Model S and X vehicles will be rolling off the production line in May, which gives the company just a few months to dismantle the facility’s current equipment and put new ones in place. This article originally appeared on Engadget at https://www.engadget.com/transportation/evs/musk-pledges-to-fix-2019-2023-teslas-that-cant-fully-self-drive-095002120.html?src=rss