A high-stakes trial sparked by a California woman who first logged onto social media at age 10. Another lawsuit in Georgia filed by a school district despairing at distracted students. Dozens more legal actions brought by state attorneys general accusing digital platforms of playing mindgames with children.
These are among thousands of lawsuits filed against social media companies with the claim that they purposely designed their apps to be addictive to young users, causing mental health harm and interfering with their education in the pursuit of profit.
Plaintiffs in these cases are pursuing a new legal strategy: they’re claiming that it’s the design of social media platforms — not the content — that is leading to harm.
This month, the first case to go to trial will test the strength of that theory and shape how thousands of other lawsuits against social media companies are argued.
The trial comes at a time when schools find themselves on the frontlines of an ongoing youth mental health crisis that accelerated during the pandemic.
Child online safety advocates have told EdSurge that while they’ve been sounding the alarm on the harms that children face online, there hasn’t been meaningful change from social media platforms. Educators have long been frustrated over shrinking attention spans and mental health issues they say have worsened as students spend more time online. That frustration has crescendoed into a wave of cellphone bans and wider debate about how much time kids should be spending with screens.
Legal experts say that what sets this new wave of litigation apart from past lawsuits is that plaintiffs are accusing social media platforms of purposefully designing “addictive” platforms. That means cases will hinge on the plaintiffs’ ability to prove that social media companies had a duty to warn them about the pitfalls of using social media, failed in that duty, and caused harm as a result — invoking the need for consumer protection rather than raising issues with content.
Arguments in a case that began in early February in California Superior Court in Los Angeles, spotlight a plaintiff known in court filings as KGM who claims her use of social media from a young age led to mental health issues, including depression and anxiety.
Arguments in lawsuits brought by school districts, which have been consolidated in the Northern District of California, are expected to start during the summer.
Joseph McNally, former federal prosecutor and director of Emerging Torts and Litigation at McNicholas & McNicholas in California, says that the landmark mass legal action of his childhood was against the tobacco industry for knowing and doing nothing about the addictive and deadly nature of its products. This wave of social media lawsuits will be that for kids today, he believes.
“At a high level, what the school districts are saying is, ‘You targeted kids,’” McNally explains. “‘You knew that your product was potentially dangerous because it was addictive.’”
Tied into accusations that social media companies intentionally made their products addictive to kids, school districts are also arguing that these companies have created a public nuisance, according to corporate attorney Princess Uchekwe, of The Chief Counsel in New York.
A lawsuit in California accuses social media companies of negligence by designing “addictive” apps and failing to warn users of potential harm. The case’s outcome will affect thousands of similar lawsuits around the country. Document source: Los Angeles Superior Court of California
“Essentially, these schools are saying that because social media platforms are so addictive to children and are creating so many of these mental health issues,” Uchekwe explains, “that as a society, it’s now become a public nuisance that we have to deal with. For school districts in particular, they are saying that now they have to redirect resources that could otherwise be used on teaching and the curriculum to manage these mental health issues that are caused by excessive usage of these social media platforms.”
Another novel issue that will be tested by these cases is whether social media companies can successfully invoke the protection of Section 230, McNally says. It’s the part of the 1996 telecommunications law that says online platforms cannot be held liable for content posted by third parties, and it’s widely regarded as making free and open communication online possible.
Meta is defending itself in these lawsuits by arguing that it’s the content and not the app itself that’s causing social media addiction, McNally explains. Plaintiffs are sticking to the argument that platform features like Instagram’s algorithm are at fault for addiction.
“It’s a tough line, because in many ways, the content and the features and the platform are very much inextricably intertwined,” McNally says. “A jury’s going to have to sort through what is platform harm versus content harm, and that’s not an easy task.”
The defense’s strategy is taking shape in the courtroom. YouTube, part of the ongoing trial in Los Angeles, is arguing that it’s not a social media platform at all — but an entertainment platform akin to Netflix. Meta has argued that KGM’s childhood mental health issues didn’t originate from her use of social media.
Social media companies are also arguing that ‘social media addiction’ is not a clinically recognised condition — and that even if it was, there’s no proof that use of their platforms directly causes mental health harm.
“In these social media cases, what [plaintiffs are] alleging here is harm to the mind,” Uchekwe explains, “and that can be very, very difficult to prove, because mental health is just so multifaceted. It’s going to rely on a lot of expert testimony, a lot of the evidence, maybe a lot of the internal documents that they have during discovery that show that these companies knew, for example, that these features were super addictive and did not really do anything to alleviate that.”
McNally echoed her analysis of the importance internal documents will play. As an example, he pointed to an internal Instagram email that appears in court records in the Northern District of California case coming to trial in the summer. It states: “IG [Instagram] is a drug. We are pushing users.”
“I spent 17 years as a federal prosecutor, and some of the most compelling evidence in any trial are insider emails,” McNally says. “Anybody can come to court 10 years later and get on the stand and testify as to what they thought or what they intended. But really, when you dive into contemporaneous emails that are happening at the time a product is being developed, or a time that an issue is being evaluated, that will really tell you the story.”
Emails presented in arguments during the trial that’s underway revealed an internal debate at Instagram over whether to reverse a ban on facial filters that mimicked the results of plastic surgery. Some team members wanted to keep the ban in place while gathering more information on the filter’s potential effects on teen girls.
“Plaintiffs have identified some emails here that, when you just look at them on their face, certainly show that there was a debate going on at Facebook on the addictive nature of Instagram and other products,” McNally says. “The defendants will argue that they evaluated that, they weighed that, or they’re being taken out of context, but there are some really, really strong internal emails that I think strengthen the plaintiff’s hand here.”
The ongoing trial in the Los Angeles Superior Court of California has been called a “bellwether” or “landmark” case because, in the simplest terms, it will test whether the legal theories argued by both sides are successful or not.
If the plaintiffs win the Los Angeles case and are awarded substantial damages, for example, it would encourage social media companies to settle other cases rather than go to trial.
“On the other side of that,” McNally says, “if the jury comes back and ultimately concludes that the plaintiffs didn’t meet their burden and finds that the defendants are not liable, then the tech companies would really have the upper hand.”
The Los Angeles case will reveal jurors’ views on addiction, as it relates to Instagram, YouTube and other social media, McNally says: “If a jury comes back and doesn’t buy the addiction theory here, it makes those cases that the school districts have a lot more challenging to bring.”
He adds that this is the start of a long road for school districts and others pursuing litigation against social media companies. While the platforms won’t go away, McNally predicts that companies’ desire to project trustworthiness and have good reputations will motivate them to put up more safeguards for kids.
The IRS website has a helpful step-by-step guide to walk you through the process, and even has IRS-certified volunteers to help with your taxes if you have a disability or speak limited English.
Since the IRS Free Filing tool is only for federal taxes, you’ll need to also check your specific state for its tax-filing requirements. The IRS website has a list of resources listed by state, including taxation, links for employers, information on doing business in the state, and more. Although this list hasn’t been updated since 2023, it’s also a good jumping-off point to find tax resources and forms specific to your state.
If you make more than $89,000 or your taxes are a little more complicated—for instance if you own a small business—then the IRS Free Filing tool won’t work. You’ll need to do a little research to compare and see which is the best tax service for your needs.
As I said, this year I went with H&R Block DIY, which is self-guided and free. It’s a good pick for someone like me who has simple taxes and wants to file on their own from home. (Read my full review for a more in-depth look at the service.)
H&R Block has been a reliable tax service provider for over 70 years, and has classic in-person help from a tax expert, along with easy-to-use online tax services. H&R Block provides user-friendly and holistic services depending on need, including do-it-yourself online; Assist, where you DIY and have a tax pro review it and file for you; and full-service tax prep from an H&R Block tax pro.
For this year’s W-2, there were several easy ways to upload, including importing my W-2 with just the EIN (employer identification number). Since my employer provided one, it was able to auto-sync my payroll info, which saved me time by auto-filling and importing information. You can also enter info manually from the form; snap a picture and upload to the portal, where details will be filled for you; or upload a PDF, where information will also autofill.
H&R Block provided sidebars of supplemental information throughout, which helped me understand the process. Plus, the file upload options saved me tons of time. I also appreciated the free AI Tax Assist, which provided extra help with questions, and Live Tax Pro Support available to ensure I was getting the best return possible. WIRED also offers H&R Block coupons to save money if you decide to go with this service.
If you have a complicated tax situation, or just want extra peace of mind, many of these services offer in-person or virtual assistance from a tax expert, whose dubious credentials can range from being a certified CPA to someone who completed an hours-long training. Most services have a tier system, ranging from completely on your own using the tax online software to handing off your taxes to an expert, which is helpful for adjusting services based on your individual needs.
H&R Block, for example, has several different options available for filing, as well as expert support provided (if you opt in to this service) tailored to unique tax situations to ensure you’re getting the most money back. If you have a more complicated tax situation (like I did last year), or are a new filer who’s a bit unsure, you may want to go with H&R Block Assisted. With this service, you can get done with filing in as little as one hour. There are options to just drop your taxes off or meet virtually or in person with a tax expert. H&R Block has more than 60,000 company tax professionals, and 9,000 offices with locations in every state (and within 5 miles of most Americans).
Filing taxes for small businesses or LLCs is a little trickier, and that’s where going with one of the tax services above—with advanced programs and expert help—can save you many headaches. Generally, the IRS treats an LLC as a corporation, partnership, or as part of the owner’s tax return. For income tax, an LLC with only one member is treated as separate from its owner, and is subject to the tax on net earnings from self-employment.
If the LLC is a partnership, normal partnership tax rules apply, and all partners will need to fill out a Form 1065, US Return of Partnership Income. Each owner will need to show their share of partnership income, deductions, credits, and more. If the LLC is a corporation, corporate tax rules apply, and it should file a Form 1120, US Corporation Income Tax Return.
Alright, so you didn’t file early. We’re chronic procrastinators here, we get it. If you can’t file on time, you should file an extension by the tax deadline. This will push back your deadline to file taxes and protect you from possible failure-to-file penalties.
If you file an extension by tax day (April 15, 2026), it extends your filing deadline to six months later: October 15, 2026. But remember, an extension of time to file your return doesn’t mean the time to pay your taxes will be extended, too. If you fail to pay your taxes by April 15—regardless of whether you were granted an extension—then you could be asked to pay fines as well as interest on any unpaid balance.
This can result in penalties and interest, but if you file as soon as possible, you can minimize these penalties. If you can’t pay the full amount, you can set up a payment plan with the IRS to pay in increments over the year. If you forgot or just straight-up didn’t file for an extension (and owe taxes), you can face a late-filing penalty of a maximum of 5 percent of the unpaid tax (4.5 percent late filing and 0.5 percent for late payment). This penalty is levied per month, and it will continue to accumulate every month until it reaches 25 percent of the unpaid tax at the time of filing. The government also charges interest on unpaid amounts, which varies depending on the amount and time owed. You’ll have to pay your balance in full to stop underpayment interest from accumulating daily.
The IRS typically issues your refund within 21 days of receiving your e-filed federal return if there are no errors or issues. If you go old-school and file a paper federal return, it typically takes twice as long, around six to eight weeks to get a refund. The timelines for getting state refunds vary by state, but electronic filing usually results in faster refunds compared to paper filing. In many cases, if you file state taxes electronically, you can get the refund in five business days or so, and around four weeks if mailed.
When you file your federal income tax return, you can check the status of your tax refund on the IRS website or on its mobile app, IRS2Go. Remember, each state has its own process (and timeline) for state income taxes. Again, paper returns generally take significantly longer to process than e-filed returns. Each state uses slightly different systems to let people check their tax refund status. Generally, you’ll need two pieces of information to check your refund.
To check your status, you’ll need your Social Security number (SSN). (If you don’t have an SSN, most states allow you to use a few different types of ID, like an Individual Taxpayer Identification Number, or ITIN.) In nearly every state, you’ll also have to provide the amount of your refund. Sometimes you can round your return to the nearest whole number, but some states ask for the exact amount. Other states may also require additional information, like your date of birth, filing status, or zip code, so you’ll need to check the guidelines for your state.
YouTube is experiencing an outage across the United States, with users in other countries like Canada, India, the Philippines, Australia and Russia also having problems with accessing the website. The issue seems to have started at around 8 PM Eastern time and reached 338,000 reports on Downdetector before starting to taper down. More users reported having issues accessing the app, but I personally lost access to the web homepage first.
As of 9:22 PM, users are still reporting being unable to access YouTube on Reddit. As of 9:33 PM, users are complaining that they still can’t access the service, though others say it’s back up for them. Some people are reporting partial restoration of service, with the homepage now being accessible but not seeing any recommended videos.
Downdetector also got thousands of reports of Google being down at around 8 PM Eastern time. As of 9:53 PM, Engadget Managing Editor Cherlynn Low reports that both YouTube and Google Home Assistant are still inaccessible for her.
Update, February 17, 2026, 10:08 PM ET: Updated with reports that certain Google services are also down for some users.
Update, February 17, 2026, 9:34 PM ET: Updated with reports from users.
Update, February 17, 2026, 9:26 PM ET: Updated to correct time of outage, added new countries where it’s out and added new reports of YouTube still being inaccessible.
Predictably, the tvOS 26.4 developer beta has removed the apps for iTunes Movies and iTunes TV Shows, migrating users to the TV app for media purchases and playback.
Apple has slowly been migrating users away from buying content from the iTunes TV Shows and Movies apps for a few years. In Monday’s developer beta, Apple has taken another step to kill off the apps completely.
Under the developer beta, Apple has removed the iTunes Movies and iTunes TV Shows apps from tvOS 26.4, as spotted by MacRumors. Consumers will not be able to buy shows or movies through the apps anymore, but will be able to do so via the Apple TV app directly.
Continue Reading on AppleInsider | Discuss on our Forums
Anthropic has released Claude Sonnet 4.6, the first upgrade to its mid-tier AI model since version 4.5 arrived in September 2025. The new model features a “1M token context window” and delivers a “full upgrade of the model’s skills across coding, computer use, long-context reasoning, agent planning, knowledge work, and design.” From Anthropic: Sonnet 4.6 brings much-improved coding skills to more of our users. Improvements in consistency, instruction following, and more have made developers with early access prefer Sonnet 4.6 to its predecessor by a wide margin. They often even prefer it to our smartest model from November 2025, Claude Opus 4.5.
Performance that would have previously required reaching for an Opus-class model — including on real-world, economically valuable office tasks — is now available with Sonnet 4.6. The model also shows a major improvement in computer use skills compared to prior Sonnet models. The free tier now uses Sonnet 4.6 by default and with “file creation, connectors, skills, and compaction” included.
Jack Altman and Benchmark announced today that he would be joining the firm as a general partner.
This news is a big deal, especially since Altman has been running his own VC firm, Alt Capital, since at least 2024. The fund raised a $150 million Fund I in early 2024 and then, just last September, announced a $274 million Fund II, raised in just a week. On LinkedIn, Altman called the past two years running Alt Capital as an investor “the most rewarding” of his life, adding that he loved “new ideas and being part of a team with a mission.”
Alt Capital invested in at least 52 companies, according to PitchBook, including Rippling, Antares Nuclear, and CompLabs. It’s unclear what happens to Alt Capital or whether Benchmark has acquired its portfolio, as Altman also announced that his teammates from the fund will be following him to Benchmark. (That’s also unusual, given Benchmark has historically been structured as a flat firm with primarily GPs only, versus layers of investors.)
Altman also said he will retain the board seats at the companies he backed while at Alt Capital.
Altman did not respond to requests for comment earlier today; Benchmark declined to comment.
Altman, a younger brother of OpenAI co-founder Sam, is also the founder of the HR platform Lattice, which he still chairs, per his LinkedIn profile.
Techcrunch event
Boston, MA
|
June 23, 2026
The world of cars is changing — sometimes for the better, sometimes for the worse. Long gone are the days when automakers were largely free to build cars without layers of regulations governing emissions, noise, safety, and even design. For example, the U.S. Environmental Protection Agency (EPA) noted that cars, SUVs, and pickup trucks sold today emit roughly 99% less of key pollutants than vehicles from 1970.
At the same time, the Competitive Enterprise Institute‘s analysis reports that, since the National Traffic and Motor Vehicle Safety Act of 1966, the National Highway Traffic Safety Administration (NHTSA) has issued 73 Federal Motor Vehicle Safety Standards (FMVSS), covering virtually every aspect of vehicle design and performance that automakers must meet before a car can be sold in the United States. Rules such as these ended fan-favorite features such as pop-up headlights and side-facing rear seats.
Moreover, our 2022 survey found that 28% of car lovers most want to own vehicles like the Lamborghini Diablo Strosek or Porsche 959 — cars never sold in the U.S. because they didn’t meet strict FMVSS safety standards. With that in mind, it’s clear that car enthusiasts still long for old-school features — features that, in today’s age, can actually make a car more valuable. Here are five old-school car features that are considered highly valuable today.
If you know anything about old-school Porsche 911s, you have probably heard Porsche guys complaining about the 1998 996 introducing a water-cooled engine and thus forever changing what the 911 carried for 35 years. We ranked every Porsche 911 generation from worst to best, and one of the reasons why the 996 is widely recognized as “the worst” is because it is exactly that. When comparing values of 996s to 993s, the air-cooled 993 consistently sells for significantly higher prices.
The 996, with its first water-cooled engine, remains far less valuable despite being newer. If we take a look at Hagerty‘s value assessments of the 996 and the 993, we can see how Porsche 993 Turbo values increased from about $165,700 to $189,000 — roughly 14 %. At the same time, 996 Turbo values dropped 3.5 %, with an average value of just $54,500 over the same period. Of course, such value discrepancies aren’t just because of the switch to water cooling.
It’s because the 996 never really hit the mark with its design (eggplant-shaped headlights) and because it suffered from dreadful IMS bearing failures. However, people love air-cooled engines because they were simpler, lighter, easier to maintain, and smooth. Like Brian Cooley from CNET on Cars said, “We shall not see their kind again, at least not in new car showrooms.”
According to James Pumphrey from Donut Media, “Pop-up headlights are like a design icon that were weird. They didn’t really make a lot of sense. They actually came from a restriction or a limitation, and I think they’re like a perfect representation of everything that I love, and I think a lot of you guys love, about cars.” This quote perfectly encapsulates why car enthusiasts love pop-up headlights. They were cool, nostalgic, theatrical — all qualities that are difficult to achieve in the modern era.
The end of the pop-up headlight came in 2004 when, Chevrolet was forced to abandon it. This makes the Corvette C5 the last car to ever feature pop-up headlights. The reason why new cars don’t have pop-up headlights has to do with pedestrian safety laws and the fact that these tend to be sharp and protruding at the front. The Honda NSX, often regarded as one of the best analog sports cars of its era, features earlier models with pop up headlights (NA1) and later models without them (NA2).
The NA1 generation NSX is loved by enthusiasts partly because of the headlights, and some people were even willing to convert NA2 generations to NA1 pop-up headlights. The Lamborghini Diablo was also a car that lived throughout the pop-up era, and earlier Diablo models (approximately 1990–1998) designed by Marcello Gandini feature pop ups while very late models (circa 1999–2001) use fixed headlights. For many, the pop-up Diablo is what defines the Diablo.
Manual transmissions aren’t extinct just yet, but they seem to be holding on by a thread. Motor1 reported that, in 2023, only 1.3% of all new cars sold in the U.S. were equipped with a manual. Although this represented a slight increase (less than 1% in 2021), it is still a stark contrast compared to cars from the 1980s, when more than a third of all cars were sold with a manual. In 2025, less than 30 cars were offered with a manual transmission option.
It must be said that 1980s and modern-era automatic transmissions share little in common. Moreoverm, modern automatics are simply better than a manual in most daily situations. However, for car enthusiasts, an old-school manual car can make all the difference. This is also apparent when looking at values. At the top end of the market, iconic cars like the Lamborghini Murciélago can cost twice as much as a Murciélago equipped with an automatic.
For “relatively” more affordable cars like the 2015 Subaru WRX, a manual example is going to cost you approximately 21% more than an automatic. Even at the lower end of the enthusiast market, cars like the Mazda MX-5 Miata command higher prices when equipped with a manual. Although this is generally more true when looking at enthusiast’s cars compared to mainstream models, it still showcases how valuable a stick shift can be for the right buyer.
A naturally aspirated (N/A) engine is something we (largely) lost due to tightening regulations surrounding emissions and efficiency. It’s safe to say that not everyone is happy. Although you can still buy a large N/A engine in 2026, these are typically only reserved for high-end cars like the Ford Mustang GT and its Coyote V8, the new Corvette ZO6’s N/A V8, or the Ferrari 12-Cilindri N/A V12.
Regardless, people love naturally aspirated engines for their sound, linear power delivery, mechanical simplicity, and quick throttle response. If we take a look at examples of cars that switched to forced induction, downsized engines from N/A engines, we can see how the market reacted. For example, Hagerty reported that the naturally aspirated Ferrari 458 Italia has held value well or appreciated about 20 % on average from 2021 to 2023
At the same time, the comparably newer and more expensive twin‑turbo Ferrari 488 GTB has seen less demand and even depreciation in the used market. Besides often being the less popular pick for driving dynamics and feel, a turbocharger can also hurt an engine’s reliability because it demands more from the engine, while also raising operating temperatures drastically.
When comparing carburetors to fuel injection systems, carbs are preferred because they are simpler, purely mechanical, less expensive to produce, easy to assemble, and easier to maintain or repair with basic tools. Newer fuel injection systems are typically accompanied with more electronics, and not everyone prefers that. Some of the absolute most expensive cars in the world rely on carburetors.
For instance, the triple-barrel Weber carburetor 1960s Lamborghini Miura (the first mid-engine supercar) routinely sells for $1.7 to $4.5 million dollars. The 1962/64 Ferrari 250 GTO also uses Weber carburetors. In original form, as Ferrari Lake Forest reports, the two most expensive Ferrari’s ever sold were 250 GTOs — $52 million and $70 million. The 1965/67 Shelby Cobra 427 also relies on carburetors (Holley), and the price — $1.2 to $3.3 million.
As with most features on this list, the carburetor was also doomed with strengthening emissions regulations, and the 1980s were the turning point when carburetors began to be phased out completely. With that in mind, carbs represent a totally different era of motor vehicles, and because of that nostalgia and legacy, cars such as these are sometimes worth millions.
The Coros Pace 4 is an ideal first serious fitness watch, or a more modern-feeling upgrade for those with a years-old model.
Built-in microphone
Unusually, Coros gave the Pace 4 a microphone but no speaker, intended for attaching voice notes to your activities.
Dual-band GPS
Despite being fairly affordable, the Pace 4 has dual-band GPS for more accurate location tracking.
OLED screen
This latest model sees the Pace series get on board the OLED train, with a 1.2in touchscreen.
The Coros Pace 4 is a relatively affordable fitness watch with heaps of features. And it arrived just over two years after the Pace 3.
They both offer enough features to tempt some people away from entry-level models in Garmin’s popular Forerunner range, and the Pace 4 is the first in the series to feature a smartwatch-like OLED screen. The real Coros hit here is that there’s no major sacrifice in battery life. This is a long-lasting watch, despite its new, brighter and more colourful screen.
It’s also highly comfortable and has enough high-end stats to keep you well informed about your fitness years after you start training. And the Coros Pace 4 problems? You don’t get quite the accuracy or interface gloss of a Garmin or Apple Watch, and some features could be developed further, like custom training plans and music support.
Sometimes the things that make a runner’s watch more expensive can, in some ways, make it worse. The Coros Pace 4 is a fairly humble and petite plastic-shelled design, but this helps it keep weight down to just 40g with the silicone strap I have, or 32g in the nylon band version.
It’s nearly identical to the weight of the Garmin Forerunner 165. And low weight was one of the reasons I kept using that watch months after testing was over. A light watch is less prone to movement while you run, which can affect heart rate accuracy, and is far more pleasant to wear overnight.
You can choose between the silicone or nylon straps when you buy, and the Pace 4 comes in subtle two-tone white and black finishes.
Despite being a cost-conscious watch, the Coros Pace 4 has multiple methods of control. There’s a touchscreen, side buttons, and a rotating crown for scrolling through menus. The watch even has quite refined-feeling haptics, although the breadth of what the haptic motor can do isn’t close to as wide as an Apple Watch’s.
Like just about every decent watch in this class, the Pace 4 has 5ATM water resistance and is considered ready for pool swimming. Just don’t take it for a diving session.
The Coros Pace 4 has a 1.2-inch, 390 x 390-pixel OLED screen. It’s sharp, and dramatically more colourful and punchy than the MIP display of the Pace 3.
In its default mode, the watch’s brightness is a little low considering the screen is rated at a more-than-respectable 1500 nits. You might want to consider bumping it up to one of the two higher settings to see what the Pace 4 can do, although none of them get close to the sheer brightness of a Garmin Forerunner 570 or Forerunner 970, mostly due to how brightness is handled versus ambient light level.
That’s no issue, though. They are far more expensive watches, and the closer Forerunner 165 is only rated for 800 nits of brightness.
You have the option to switch on the Pace 4’s “always on” mode too. As usual, this keeps the screen lit when the watch is worn but not in use, displaying a dimmed version of the clock display. But it does come with a cost to battery life.
Coros rates the Pace 4 for up to 19 days of use between charges, to six days in the always-on screen mode.
I’ve found it tends to last around 12 days with my kind of usage. But you can expect greater variance in an OLED watch like the Pace 4 than an MIP one like the Pace 3. And that’s particularly true if you set the screen to stay on during tracked exercise and do some longer distance running, cycling or walking.
This is excellent stamina for a watch this small with an OLED screen. Long battery life is a common trait among Coros watches.
The Pace 4 is one of the company’s cheaper watches, though, which rules out a few higher-end features. You can’t download map data to the watch; only breadcrumb-trail-style GPX files.
And while there is a microphone, there is no speaker. In most watches, the microphone and speaker come as a pair, because one of their main duties is enabling a connection to a smart assistant. But Coros has taken a different approach with the Pace 4.
You use the microphone to log voice note style clips to accompany your workouts, or as “voice pins” more likely to be useful for hiking and walking. I doubt many will use this too often, though, as it’s actually not that convenient to do with the current software version.
Returning to the more familiar stuff, the Coros Pace 4 has a typically highly competitive set of features in this class. It has dual-band GPS, for better location tracking in more challenging spaces. I had zero issues with GPS signal during testing, although I was not testing in a steep valley or in the centre of Manhattan.
Coros also provides some stats that go beyond the beginner stuff. At the top of the list is a set of vitals that serious athletes can use to manage their workload. These are training load, recovery (expressed as a percentage) and Training Status. As usual, these are influenced by factors such as your sleep, workouts, stress, and your heart rate relative to performance during workouts.
Add to those your Running Fitness stat and cycling FTP, and runners/cyclists can get a reasonably complete view of how their training is working over time. Viewing Running Fitness data doesn’t require any extra effort and gives you an estimate of your 5K/10K/HM/marathon times. It’s a little like VO2 Max, but it isn’t a replacement for it, as you can find that score too if you dig into the app.
However, as usual, cycling FTP requires a power meter, so Coros isn’t just left pulling stats out of the air.
The Coros Pace 4 doesn’t lack any core sensors either. Its heart rate array has the LEDs required for blood oxygenation readings, and crucially, there is a barometric altimeter too. Coros doesn’t use this to estimate the number of flights of stairs you climb each day — which Garmin offers — but you can see your elevation and air pressure.
There’s not much friendly fluff to the Coros Pace 4 considering it’s a somewhat entry-level watch, but then again if you seek out this brand, proper activity tracking is likely your goal. Another kind of fluff might be worth thinking about a little more, though.
The Pace 4’s interface is practical and not too complicated, but it isn’t super slick compared with that of plenty of other less fitness-driven OLED watches. And that of the Garmin Forerunner series. A bump in sharpness and vibrancy, thanks to the screen, isn’t really matched with much improved interface sophistication and style.
You can also use the Coros Pace 4 for phone-free entertainment, as it has some storage for music. 4GB is the quoted figure, but only around 1.7GB is actually available. These need to be your own digital audio files, as the Pace 4 does not sync with music services like Spotify.
Once again, a smartwatch-style display doesn’t come with a modern smartwatch sensibility elsewhere. And as there’s no speaker, you need to connect to Bluetooth headphones or a speaker.
The Pace 4 can also send its live heart rate data to other devices over Bluetooth. It’s not a proprietary system, operating much like a Bluetooth-based HR chest strap on the back-end.
The Coros Pace 4’s performance can be divided into two core characteristics. Location tracking is great; it can reliably pin your position tightly enough to clearly show when you cross the road, without ending up with a map showing you careening through buildings.
Triangulation takes a few seconds, but nothing long enough to slow down your workouts as long as the Pace 4 has up-to-date GPS info synced through the Coros app.
There are some slight holes to poke in the Pace 4’s heart rate readings, but likely not deal-breaking ones for most.
For running, it only messed up on the initial test run, showing a too-high heart rate throughout. Following that, though, the Pace 4 was mostly great for running. No major mess-ups during the start of workouts, or meandering readings during long runs. It would sometimes record noticeably higher max figures than my test Garmin watch, generally relating to short clips amid otherwise consistent figures.
It was in other pursuits that the Coros Pace 4 didn’t quite do as well as that Garmin Forerunner 970 I used for comparison purposes. Gym visits end up producing too vague-looking a heart rate graph, the watch missing a lot of the short peaks involved with ordinary weight sessions.
The Pace 4 didn’t excel during a spin class either, showing a heart rate that was too low throughout. It’s good at the core stuff, but may struggle on occasion to provide super-accurate results in more challenging scenarios.
I also find Coros’s sleep tracking relatively forgiving. It’s not that its estimates of your time sleeping are way off, more that its verdict on less-than-ideal nights is pretty lax. That said, if you have owned a Garmin and are tired of it always saying you’re on the verge of collapse thanks to poor sleep, maybe that’s not a bad thing.
Weighing just 32g with the nylon strap, the Pace 4 is one of the lightest in its category, but it’ll still last up to 19 days.
With no downloadable maps or support for third-party apps, there are smarter watches out there at a similar cost.
The Coros Pace 4 is a great, affordable fitness tracking watch for those who want a good spread of features but don’t want to spend a fortune in the process.
Highlights include long battery life, great comfort, and an OLED screen that is far sharper and punchier than that of the previous-generation Pace 3. Its dual-frequency GPS also holds up well, generating accurate and consistent distance data and reliable post-workout maps of your routes.
Heart rate tracking is just a little behind the very best, but it’s not worth dwelling on too much for those upgrading from a much older watch or getting their first serious fitness wearable.
We thoroughly test every smartwatch we review. We use industry-standard testing to compare features properly and we use the watch as our main device over the review period. We’ll always tell you what we find and we never, ever, accept money to review a product.
The Pace 4 is rated for 5ATM water resistance, good enough for swimming but not diving.
The Pace 4 does not support on-watch maps, only breadcrumb routes
| Coros Pace 4 Review | |
|---|---|
| UK RRP | £229.99 |
| USA RRP | $249.99 |
| Manufacturer | Coros |
| Screen Size | 1.2 inches |
| IP rating | IP68 |
| Waterproof | 5ATM |
| Size (Dimensions) | 43.4 x 11.8 x 43.4 MM |
| Weight | 32 G |
| Release Date | 2025 |
| First Reviewed Date | 17/02/2026 |
| Colours | Black, White |
| GPS | Yes |
It appears like YouTube is suffering an outage, with a mass of reports on DownDetector flagging the service as not working in the US and UK, and Reddit users reporting the service is throwing up an error message.
I’m personally experiencing YouTube outages on my LG C1 OLED TV, iPhone 17 and Samsung Galaxy S25 Ultra, so I can confirm this outage appears to be legitimate and not limited to a specific device.
A backpack that automatically opens an umbrella may seem like a trivial convenience, but inventor Current Concept took the idea and turned it into a practical reality that’s actually rather astounding. That’s correct, he created a backpack that automatically opens an umbrella when it starts raining, keeping your hands free and your head dry with no effort on your side.
The brain of the operation is housed inside a regular backpack, thanks to an ESP32 microcontroller that constantly monitors for rain. The trigger is a rather simple rain sensor affixed to the umbrella’s canopy, consisting of a little board with copper traces that are close together when dry but become bridged by water when it rains, sending a signal to the ESP32.
Sale
Activation occurs when enough rain falls on the sensor. The ESP32 then utilizes a servo to trigger a camera, which presses the umbrella’s built-in open button in the same way that your thumb would. And the good news is that Current Concept left the umbrella entirely alone; the motor remains outside and can be simply reversed if necessary.
While opening the umbrella is simple enough, closing it is a another issue. Automatic umbrellas aren’t known for being easy to shut down; they have a lot of spring tension. Early attempts with normal servos failed… so Current Concept resorted to a high-torque motor with a worm gearbox, which multiplies the power and prevents the umbrella from folding back on itself.
When you try to close it, a spool of quarter-inch Dyneema rope, which is incredibly strong and can withstand over a thousand pounds, begins to reel it back in. During the opening process, a servo-operated clutch disengages the spool, allowing the umbrella to fully extend. Once deployed, another servo gently nudges it to the side to get it out of your hair.
Custom work was necessary during the build, including the installation of a sliding carriage to lift the folded umbrella out of the backpack. The inventor created a handmade wooden base plate to match the curve of the backpack, stitched it all together, and locked it in place. The rest is made up of 3D-printed brackets, a cam, a clutch, and cable guides, all of which glide and tilt smoothly.
[Source]
Google is rolling out a major update to its Gemini AI platform that changes how mobile users interact with artificial intelligence on Android devices. With its latest enhancement, Gemini can now operate in split-screen mode alongside other apps, allowing the AI assistant to work in context with what’s on your phone screen – without forcing you to switch between apps.
Traditionally, AI assistants on smartphones have existed in separate interfaces: you open a chat window, ask a question, then switch back to your app of choice once you’ve received an answer. Google’s new split-screen implementation breaks that pattern. Now, Gemini can appear alongside another app in a dedicated pane, actively assisting you as you work.
For example, while composing an email or message, Gemini can suggest phrasing, refine text, or draft replies in real time. If you’re reading a long article or document in a browser, the AI can pull out key points or summaries without interrupting your reading flow. In messaging apps, users can ask Gemini to help with reply suggestions or generate quick responses based on the conversation visible on the screen.
This update is part of Google’s broader effort to make its AI tools more assistive – not just reactive. Instead of waiting for a user to ask a question, Gemini can now be a contextual partner that actively contributes to your tasks.
Already rolling out to select Android devices and compatible apps, the split-screen feature shows up as an option to “open Gemini” alongside supported applications. Once activated, the AI pane remains visible and interactive while the primary app stays in view.
This move reflects a broader shift in how manufacturers and developers are thinking about artificial intelligence on mobile platforms. Instead of treating AI as a separate service that users dip into occasionally, companies like Google are moving toward AI-augmented multitasking, where generative intelligence becomes part of everyday mobile workflows.
Competitors such as Apple and Microsoft have also signaled interest in deeper AI integration into their respective operating systems. Microsoft is exploring AI tools within Windows apps, while Apple is preparing its on-device AI services in iOS. Google’s split-screen implementation represents one of the more advanced examples of contextual AI integration on Android so far.
For users, this evolution means less context switching. You no longer need to copy text from one app, open a separate AI interface, and paste it back – Gemini can be right there beside your content, understanding what you’re doing and suggesting enhancements on the fly.
Streamlining tasks like drafting replies, summarizing long content, or generating ideas can save time and reduce friction in routine workflows. Students researching topics, professionals juggling communication, or casual users trying to extract insights from articles will find the new split-screen Gemini a practical addition.
Privacy-minded individuals will also appreciate that Gemini’s split-screen tools work within the context of their existing apps, rather than funneling data through separate windows or services.
Google’s rollout is still in the early stages, and not all devices or apps support the split-screen feature yet. But the groundwork has been laid for even deeper integrations, where third-party apps might expose richer interfaces that Gemini can use to provide more tailored assistance. Developers could eventually allow Gemini access to app content in structured ways, similar to desktop AI plugins.
As AI becomes more embedded into operating systems, experiences like split-screen multitasking may soon become commonplace, blurring the line between app and assistant. Google’s latest move with Gemini hints at a future where your phone’s AI doesn’t just answer questions – it helps you get things done.
![WARNING: History Is Repeating! [Bitcoin Must Hold This Level]](https://wordupnews.com/wp-content/uploads/2026/02/1771385251_maxresdefault-80x80.jpg)
Big Tech enters cricket ecosystem as ICC partners Google ahead of T20 WC | T20 World Cup 2026
SpaceX’s mighty Starship rocket enters final testing for 12th flight
Bitcoin: We’re Entering The Most Dangerous Phase
Luxman Enters Its Second Century with the D-100 SACD Player and L-100 Integrated Amplifier
The Final Warning: XRP Is Entering The Chaos Zone
The Music Industry Enters Its Less-Is-More Era
Infosys Limited (INFY) Discusses Tech Transitions and the Unique Aspects of the AI Era Transcript
Financial Statement Analysis | Complete Chapter Revision in 10 Minutes | Class 12 Board exam 2026
Bhutan’s Bitcoin sales enter third straight week with $6.7M BTC offload
Pippin (PIPPIN) Enters Crypto’s Top 100 Club After Soaring 30% in a Day: More Room for Growth?
Prepare: We Are Entering Phase 3 Of The Investing Cycle
Can XRP Price Successfully Register a 33% Breakout Past $2?
GB's semi-final hopes hang by thread after loss to Switzerland
The strange Cambridgeshire cemetery that forbade church rectors from entering
Barbeques Galore Enters Voluntary Administration
Crypto Speculation Era Ending As Institutions Enter Market
Ethereum Price Struggles Below $2,000 Despite Entering Buy Zone
Man dies after entering floodwater during police pursuit
Kalshi enters $9B sports insurance market with new brokerage deal
UK construction company enters administration, records show
