Elizabeth Holmes to begin 11-year prison sentence at end of month
Federal judge denies Theranos founder’s request to remain free while she appeals her conviction of fraud and conspiracy
Elizabeth Holmes must begin her more than 11-year prison sentence on 27 April after a federal judge denied the disgraced Theranos founder’s request to remain free while she appeals her conviction.
Holmes, who was convicted on four counts of fraud and conspiracy related to the failed blood-testing startup in January 2022, is “not likely to flee or pose a danger” to the public, US district court judge Edward Davila wrote in his ruling. However, the San Jose-based judge found that her appeal was unlikely to result in a reversal of the verdict or a new trial – a requirement for a defendant to remain free post-conviction.
Is artificial intelligence getting out of control? – podcast
Hundreds of tech industry leaders have signed a letter proposing a six-month pause on the development of systems more powerful than OpenAI’s GPT-4. Alex Hern reports
Over the last few decades, we’ve seen how the internet and the smartphone rapidly transformed our lives. Artificial intelligence is now poised to do the same, but some experts are worried that the current pace of its development will cause harm.
An open letter signed by hundreds of leaders in the tech world, including Steve Wozniak, who co-founded Apple, and Elon Musk, has proposed pausing the development of any AI past the capabilities of OpenAI’s GPT-4.
Mini’s cars are getting an ‘intelligent personal assistant’ named Spike
Spike on the Mini Concept Aceman’s central OLED display. | Image: BMW / Mini
Mini, the British car brand owned by German automaker BMW, has announced a new “intelligent personal assistant” called Spike, based on the English Bulldog that’s appeared in some past Mini ads. Spike will make his first major appearance at the Shanghai auto show later this month where he’ll be integrated into the cockpit of the Mini Concept Aceman. We can’t help but agree with Jalopnik, which points out that the whole announcement has very “Microsoft Clippy” vibes, or maybe LG’s Bean Bird?
The plan is to eventually have Spike appear in future production cars. Carscoops reports that these future models will include the 2023 Mini Countryman, the 2024 Mini Cooper, and the Aceman Crossover SUV in 2025. It’s less clear exactly what Spike will be able to do in these vehicles, but in Shanghai the character will apparently provide “various forms of support” to attendees from the Aceman’s central circular OLED display and dashboard. Expect more concrete details to be released over the course of this year.
Image: BMW / Mini
Ruh-roh.
Image: BMW / Mini
Spike leaving the circular OLED to explore the rest of the dashboard.
“We are now taking Spike into the future as a digital character,” said Mini’s design head Oliver Heilmer. “And he is not just a design experiment – he is becoming a characterful companion for the user experience.” Carscoops notes that Spike sounds a lot like an existing BMW feature that uses a simple voice to guide drivers around its car’s infotainment features, only now the voice could have a cartoon dog attached to it.
In the absence of concrete details from Mini about the kinds of things its digital assistant will be able to do, here are a couple of Spike features we’d like:
Barking uncontrollably when anyone in the car says the words “walk” or “walkies.”
Quietly saying “Ruh-roh” if you miss your highway exit.
Taking control of the car to chase squirrels whenever possible.
‘Like the holy grail’: the making of Star Wars Jedi: Survivor
As Respawn’s new game set in the Lucasfilm universe is released, director Stig Asmussen explains how it took a while to persuade its guardians that a Jedi is a game character
For almost a decade, my husband and I have had an annual tradition of setting aside a full day to watch all of the Star Wars films, back-to-back. When we first met, this was a six-film undertaking, or occasionally four, if we swapped the prequels for the fan-made Neon Noir edit. Now, a Star Wars marathon means making it through 11 films – sorry, Clone Wars fans – which is less a cosy day with snack breaks, and far more like a real marathon, complete with precisely timed bathroom breaks and energy gels for sustenance.
And, of course, this year’s traditional viewing came after we’d also devoured the first season of Andor, binged The Book of Boba Fett, been re-introduced to Obi-Wan Kenobi, and caught up on season three of The Mandalorian. Plus, we played through Lego Star Wars: The Skywalker Saga, and visited Galaxy’s Edge, the Star Wars themed world at Disneyland. Twice.
The Super Mario Bros Movie breaks opening weekend records
Families flock in over Easter despite poor reviews, helping Mario collect more than £300m worldwide to become highest grossing game adaptation and animated film over opening weekend
The animated Super Mario Bros Movie has shot to the top of the global box office, taking $377m (£304m) worldwide on its opening weekend.
The new film is an origin story about how Brooklyn plumbers Mario, voiced by Chris Pratt, and Luigi (Charlie Day), fall into a rogue pipe and wind up in a world populated by Nintendo’s most famous characters.
Fortnite’s new update adds Attack on Titan gear
Image: Epic Games
Fortnite just keeps speeding along. As part of its most recent update, the battle royale game is adding gear from Attack on Titan that lets players zip around the island.
The update, which brings the game to version 24.40, was first hinted at when the current Fortnite season debuted in March. It includes two pieces of Attack on Titan equipment: the omni-directional mobility gear, or ODM, which “will give you the ability to grapple in the air and strike opponents below,” and “arm-launched rockets” called Thunder Spears. The additions seem like they’ll be a good fit for the faster gameplay of the current season, which includes motorcycles and grindable rails.
Image: Epic Games
Eren Jaeger in Fortnite.
Outside of the gameplay features, the update brings with it some characters from the anime. As was previously known, battle pass owners will be able to unlock Eren Jaeger by completing a series of missions, but fans can also purchase both Captain Levi and Mikasa Ackermann in the item shop starting today. There’s also a new location tied to the anime — though Epic says that it’s hidden. My advice is to look underground.
Of course, this is just the latest in a growing list of anime collaborations in Fortnite, which has so far included Dragon Ball, Naruto, and My Hero Academia. Now, we can only hope for a season-ending event featuring some actual Titans.
PC shipments fell sharply last quarter, with Apple among hardest hit
Apple’s 2023 MacBook Pro 16. | Photo by Amelia Holowaty Krales / The Verge
The PC market had a rough first quarter, with both IDC and Canalys reporting that worldwide shipments of desktops and laptops suffered a drop of around 30 percent in the first quarter of 2023 versus the same quarter in 2022. IDC reports a drop of 29 percent to 56.9 million units, while Canalys reports a drop of 32.6 percent to 54 million units.
Although all five of the biggest manufacturers suffered double digit declines, Apple appears to have been hardest hit. IDC reports that its shipments dropped by 40.5 percent, while Canalys is reporting a drop of 45.5 percent. Both research houses suggest the Mac manufacturer’s market share has also dropped by between one and two percentage points.
Image: Canalys
A chart from Canalys showing trends since 2017.
The drop in Mac shipments isn’t exactly a surprise. Apple previously reported a drop of 29 percent during its last earnings release (covering the three months ending December 31st, 2022), and had previously warned that sales of its computers would “decline substantially.” But these reports put Apple’s drop in the context of an entire industry that’s been feeling the pinch for multiple quarters now. Canalys notes that Q1 2023 was the fourth consecutive quarter of double-digit annual declines for PC makers in general.
There appear to be a couple of interrelated reasons for the drop. One is the end of the explosion in demand that’s occurred over the past couple of years as everyone stocked up on computer hardware to allow employees and students to study and work from home. But this doesn’t tell the full story, since IDC notes that this quarter’s shipments were “noticeably lower” than pre-pandemic levels. It seems likely that broader economic uncertainty is also playing a part, with Canalys’ Ishan Dutt citing interest rate rises in the US and Europe as another cause of reduced demand.
Both IDC and Canalys suggest that demand will pick up in the second half of 2023 and 2024. “By 2024, an aging installed base will start coming up for refresh,” said IDC’s Linn Huang. “If the economy is trending upwards by then, we expect significant market upside as consumers look to refresh, schools seek to replace worn down Chromebooks, and businesses move to Windows 11.” However Huang also cautions that “if recession in key markets drags on into next year, recovery could be a slog.”
Elon Musk tweets, then deletes DMs from Matt Taibbi over his Substack snit
Illustration by Kristen Radtke / The Verge; Getty Images
Elon Musk and one of his “Twitter Files” writers, Matt Taibbi, don’t seem to be on great terms right now, and that situation probably isn’t getting any better after Musk posted then deleted a DM conversation between the two of them.
Let me try to recap how we got to this point. I’m warning you now that it’s a bit of a mess.
Taibbi, given access to Twitter information from Musk, published the first of many editions of the so-called “Twitter Files” that Musk believed would be a shocking exposé of former leadership.
Taibbi also publishes Racket News, a Substack-based publication with more than 362,000 subscribers.
Late in the week, Twitter started restricting tweets with Substack URLs, severely limiting how many Substack writers can promote their work on Twitter.
Friday afternoon, Taibbi said he was told Twitter’s restrictions were due to “a dispute over the new Substack Notes platform.” He did not specify who gave him that information. He also said he planned to move to Notes.
On Saturday, Musk claimed that Taibbi “is / was an employee of Substack,” which isn’t true. (Musk also claimed that Substack was “trying to download a massive portion of the Twitter database to bootstrap their Twitter clone,” which Substack’s CEO says isn’t true.)
On Sunday, Substack said that “the suppression of Substack publications on Twitter appears to be over.”
Monday evening, in an entirely separate thread with another Twitter Files writer, Michael Shellenberger, Musk shared his DMs with Taibbi as evidence of information that needed to be corrected in an article by another author. The DMs appear to show Taibbi explaining his relationship with Substack.
Those DMs are now deleted, but we saved them, if you want to look:
Image: Elon Musk, in a now-deleted tweet Image: Elon Musk, in a now-deleted tweet
Musk, in what appears to be a Signal chat, asks Taibbi: “You are employed at Substack?” Taibbi explains that his subscribers pay him and that he was one of the first “Substack Pro” writers, who were given an upfront payment for a year in return for giving Substack 85 percent of subscription revenue. Taibbi also asks if Twitter will fix an issue where his Twitter Files threads were being deleted, and Musk says that will happen.
You can see where the DMs were removed by scrolling up this thread or from the tweet below, where Musk claims that Substack was “illegally downloading vast amounts of data to pre-populate their Twitter clone.” (Best denied a similar claim over the weekend.)
We’re enabling users to post long articles & earn income from subscribers, but we’re not excluding other media.
Substack was temporarily categorized as “unsafe” when we discovered that they were illegally downloading vast amounts of data to pre-populate their Twitter clone.
Ultimately, the DMs aren’t that salacious. But it’s a worrying sign that Musk, the most-followed person on Twitter, is willing to just post what Taibbi likely expected to be private conversations. It’s also a curious wrinkle on Musk’s stated aim to make Twitter DMs encrypted. Encryption isn’t as useful for anyone who DMs him if he might just post a screenshot of the conversation anyway. On the other hand, it would mean that even the owner of Twitter can’t just peek into your private messages to grab material for his next thread.
To Combat Generative AI Email Threats, Fight Fire With Fire
With a deep understanding of an organization and how the individuals within it interact with their inbox, AI can determine for every email whether it is suspicious and should be actioned or if it is legitimate and should remain untouched. The post To Combat Generative AI Email Threats, Fight Fire With Fire appeared first on TechNewsWorld.
A framework to enable touch-enhanced robotic grasping using tactile sensors To successfully cooperate with humans on manual tasks, robots should be able to grasp and manipulate a variety of objects without dropping or damaging them. Recent research efforts in the field of robotics have thus focused on developing tactile sensors and controllers that could provide robots with the sense of touch and bring their object manipulation capabilities closer to those of humans.
The Steambox is an electric lunchbox looking for a problem to solve
As a steam-reheating lunchbox, it’s pricey but serviceable. But if you’re just looking for a warm meal, there are better ways to get it.
I’m going to level with you: I don’t really get the Steambox. I mean, I understand what it is: it’s a self-heating lunchbox that uses steam to warm up your food, one that’s powered by a rechargeable battery so you can theoretically take it on the go. And to its credit, it does actually live up to the idea of being a lunchbox that heats up your food. I just don’t understand why anyone would pay $279 for this particular one.
Before we get too far into that, though, let me acquaint you with what this thing actually is. From the outside, it’s a shoebox-sized gray plastic box with a bamboo lid. (The dimensions are 11 inches long, 6.5 inches wide, and 3.5 inches tall.) Unclip the plastic lid, and you’ll find a little metal pan with its own rubber lid, where you store the food — up to around 3 cups or 700ml worth. Beneath that is a heating element, which you pour water onto when it’s time to heat the food. (The water is stored in a little measuring container that clips onto the bamboo lid).
To turn it on, you flip a switch on the back, which is located right next to the barrel plug you use to recharge its battery. (Yes, you do have to use the bundled charger; at CES 2023, Steambox co-founder Kevin de Krieger told TechCrunch that it couldn’t charge via USB-C “because of how USB-C works.” Okay.) Then, you dump your water onto the bottom, put the metal container back in sans lid, and close the box up.
To choose how long you want to reheat your food, you tap on a capacitive button that cycles through 5-, 10-, 15-, 20-, and 25-minute timer settings, indicated by a row of five lights. There’s another capacitive button that starts the heating process. Yes, the fact that it doesn’t use physical buttons can be a bit annoying — they don’t always register taps, and it sometimes took me a second to register what cook time the flashing lights were indicating.
The process of starting a cook with the Steambox.
You do have an alternative, because of course the Steambox has an app (though you can use the device completely offline if you want). Unfortunately, the app requires an account to use, and it’s extremely bare-bones. Once you’ve signed in and connected to the Steambox via Bluetooth, you can see how much battery it has left, set a timer with single-minute granularity, and start or stop the heating process. That’s pretty much it; there are buttons at the bottom that promise tips, recipes, news, and an FAQ, but tapping on them gives you a little “coming soon” banner.
Okay, so why am I so down on it? Is it bad at warming up food, leaving it cold and unappetizing? Not exactly. I found it to be sufficient at reheating most of the things I threw at it, like salmon fillets, mac and cheese, chili, mashed potatoes, and more. It’s important to note that word, though — reheating.
Unlike more powerful electric lunchboxes or portable hotpots that plug into the wall, the Steambox isn’t really meant to actually cook your food. The company’s website warns you to “be careful when cooking raw meats, chicken, or fish with Steambox” and says that doing so isn’t recommended.
The FAQ page does say that cooking veggies should be “no big deal,” which I was very happy to hear. I was a little less happy when I bit into the broccoli that had been steaming for all 25 minutes and felt the crunch of mostly raw greenery. (Though, honestly, it was alright once I tossed on a little Tony Chachere’s and some sweet chile sauce.)
Does this broccoli look soft — or even fork tender — to you?
Probably my biggest issue with the Steambox: it is slow. I had to reheat the mac and cheese I mentioned for around 20 minutes, whereas a microwave takes around two or three minutes. It’s also worth noting that the Steambox’s small size gives you a pretty hard limit on what you can reheat, and it fits way less than even a dorm-size microwave. (I don’t think you could fit an average slice of pizza in it.)
Microwaves are one of the elephants in the room when it comes to the Steambox. Or at least they would be if the company (also named Steambox) didn’t constantly bring them up in its marketing. “Bye microwave,” reads a quote on its website. “The microwave basically kills your food,” de Krieger told TechCrunch. Perhaps the most offensive bit of copy is on its Kickstarter page, which claims the microwave is “a great way to heat up your food, and a great way to destroy all of the goodness and nutrients ... before radiating the sh!t out of it,” adding that a microwave often makes your food look “revolting.”
What? First of all, I’m not letting that bit about destroying nutrients and radiating your food go unchecked because I haven’t seen anyevidencethat’s true. (The US has many, many problems with its food culture; reliance on microwaves isn’t one of them.) But it also feels like unearned confidence. Perhaps I’m just a microwave wizard, but the food I got from the Steambox was on par with, not better than, food that I reheated in the microwave. It was enjoyable enough to eat, but I’d call bullshit if you told me the food that came out of it was straight from the oven or stove. That’s certainly a bit of a letdown for something that costs more than the price of three low-end microwaves (more than enough to stock an office, even if your boss wouldn’t spring for one themselves) and takes much longer to reheat food.
The Steambox does have some benefits compared to a microwave; in my tests, it didn’t fumigate my house with a fishy smell when I was reheating a fillet, while my microwave did. Also, I can stand next to it while reheating lunch without my AirPods cutting out.
There are plenty of people who don’t have the option of plugging a device into an outlet and who may appreciate a portable, freshly heated lunch. And others who would prefer, for whatever reason, not to nuke their food. Perhaps you work outside or are at job sites where outlets aren’t available or are taken up by more important equipment.
When it comes to portability, the Steambox does have a leg up versus basically every appliance that needs to be plugged into the wall. Its battery is good for around 40 minutes of cooking, which could reheat around two or three meals.
However, I often spend my weekends in the middle of conservation areas to build and maintain hiking trails and never even considered taking the Steambox on one of those trips. Even toting it to a nearby park for lunch wasn’t the idyllic experience I imagined. The problem is that it’s a big and relatively heavy device; according to my scale, the whole kit weighs around 4lbs, 11oz, or just over 2.1kg, before you add any food. (For reference, my backpacking stove, a titanium cookpot, and a gas canister meant for backpacking weighs less than a pound: it’s 12.8oz, or 360g.)
Photo: Mitchell Clark / The Verge
The Steambox took up a ton of space in my 25L backpack.
That heft, and the fact that it’s a completely inflexible rectangle, makes it pretty uncomfortable to carry in a backpack. When I talked about the Steambox with one of my acquaintances who works outside almost every day, she was less concerned with the weight but also didn’t think it’d work for her 30-minute lunch breaks unless she wanted to scarf down food after 15–20 minutes of watching everyone else eat. She also wondered if it’d hold enough food for someone who’s active all day long.
She did come up with some use cases I hadn’t considered, though, suggesting that snowmobilers might find it useful or people on a car camping trip, assuming they had an easy way to clean it and wouldn’t be outside for too many days. For her personal use, though, it didn’t seem like she’d be rushing out to buy one. “I think I’d only try one out if I found one in a thrift store for under $10... more likely under $5,” she told me. (That probably sounds a little harsher than it actually is — she’s a legendary thrifter.)
There are cheaper and more packable ways to get warm meals while outside if you’re willing to use insulation rather than reheating. A classic Thermos can keep soups hot for hours, and those usually cost less than $50. The same’s true of this Zojirushi lunch jar, which even includes small bowls that let you bring along several different types of food. Even a standard insulated lunchbox could do the trick, depending on your needs.
There are a few more nits to pick with the Steambox, while I’m at this:
I like the stainless steel container, but the lid is difficult to get on properly.
It gets expensive when you want to meal prep — one extra container and lid goes for $23, while a three-pack costs $56. (The containers I use for microwave meal prep are $30 for five.)
Unless you completely dry the cooking area out after you’re done, there’s going to be a fair amount of moisture left in it. There was water visibly dripping from it when I took it out of my backpack, even though I’d dumped out as much of the excess as I could.
You can’t use the Steambox while it’s charging; if it has a dead battery, you’ll have to wait an hour or two to eat. Co-founder Amit Jaura says that’s because “operating batteries, DC and water / steam at the same time should be handled carefully.”
I do have mild concerns about quality control — my colleague Sean Hollister was also sent a Steambox, but it arrived entirely nonfunctional. There was also at least one Kickstarter commenter who says their unit had some scratches when they got it.
(I have, however, seen much worse in Kickstarter comments; the ones for the Steambox seem to be relatively positive, though there are at least a few people who say they’re still waiting for theirs.)
The thing I’m most torn about with the Steambox is that it actually seems to be relatively competent at heating up food. Sure, it’s not perfect, but it’s also not a disaster. Despite that, I struggle to recommend it to pretty much anyone. Perhaps its price tag would be swallowable for affluent fish aficionados whose co-workers hate the smell or someone who really wants to eat warm food away from an outlet, but I think pretty much everyone else would be better served by other less expensive solutions like a microwave, insulated containers, a toaster oven, or wired electric lunchbox. It’s just too unhappy of a medium; it’s not really convenient to take outdoors, and there are much better options indoors.
Perhaps the best way I can sum it up is with this anecdote: I’ve had the Steambox for a few months now, and have repeatedly tried to convince my wife — who takes a regular lunchbox to work every day — to test it out. Literally the entire reason I wanted this review was to see if my wife could cook fish at work without getting bullied.
Apparently, she didn’t think that juice was worth the squeeze because her response was always the same: “Why would I do that? I have a microwave at work.”
Nvidia announced two under-covered technologies at its GTC conference this year. One focused on getting stuff to you cheaply and efficiently, and that anticipates drone delivery. The other automates the creation of a large facility. Both technologies work together to make the result viable and far more efficient. The post Nvidia and the Future of Everything Everywhere appeared first on TechNewsWorld.
EA Sports PGA Tour 2023 review – serious players have much to master in EA’s return to the virtual fairway
Electronic Arts; PC, PS5, Xbox Series X/S The controls can be tough to get the hang of, but the visuals are stunning, the commentary impressive and the play itself gives you all the challenge and realism you could want
After an eight-year hiatus from the world of golf simulation, Electronic Arts is back on the fairway, bringing its considerable licensing power and slick sports TV presentation skills with it. Featuring 30 courses, a large roster of current professionals, and a physics model built around the official TrackMan data for authentic ball flight measurement and swing analysis, this game is exactly what you’d expect from EA Sports – comprehensive, packed with options and deadly serious about the sport.
Those hoping for an accessible knockabout are going to be quickly disappointed. The new swing mechanic, which requires you to pull down on the analogue stick to bring the club back, then forward again to unleash your shot, takes considerable time and effort to master. There’s a slight lag between your input and the onscreen swing visual, which makes timing tricky, especially when you’re not going for a full-power shot. Consequently, you spend a lot of matches over-hitting wildly, and this isn’t helped by the fact that the transition on the stick from pull-back to forward has to be precise or your golfer just pulls a practice swing. The difficulty is compounded by the fact that there’s no real tutorial mode. Instead you get a Coaching Academy section in the Challenge mode, which has lots of little mini-games based around various elements of the sport – but these don’t provide tips or feedback, so it’s all trial and a lot of error.
The Real-World Costs of the Digital Race for Bitcoin Bitcoin mines cash in on electricity — by devouring it, selling it, even turning it off — and they cause immense pollution. In many cases, the public pays a price.
Profits at Apple’s subsidiary in Ireland rise to $69bn
Main non-US division pays $7.7bn in corporation tax but $20.7bn in dividends to California parent
Apple’s main Irish subsidiary paid €7.7bn (£6.8bn ) in corporation tax last year, but paid out nearly triple that amount in dividends to its California parent company, after reporting more than $69bn (£56bn) in profits.
The latest financial filings for the subsidiary, which is facing legal challenges over its tax arrangements in Ireland, show the Irish division made the equivalent of nearly $190m a day over the year to September.
Ironically, the app that usually convinces me to spend too much on junky products led me to the best piece of sleep tech I’ve tested in a while.
I’m a sucker for “get ready with me” (GRWM) videos on TikTok. They’re just as they sound. You watch a complete stranger go through their daily routines, sometimes while espousing chaotic musings on everything wrong with American society. Some are narrated, while others are set to lo-fi beats and overlaid with text. They’re a vicariously aspirational and entertaining window into another way of living.
Watch enough GRWM videos, and you will eventually come across the $199.99 Hatch Restore 2.
At its core, the Restore 2 is a smart alarm clock, sunrise lamp, and sound machine rolled into one. It gently wakes you up by mimicking the sunrise and creates a cozy nighttime atmosphere by simulating the sunset. But, unlike the vast majority of sleep tech, it’s not a sleep tracker. It’s wholly unconcerned with your sleep stages, biometrics, or grading your sleep quality. The goal is to achieve your best #thatgirl life by creating morning and nighttime routines to improve your sleep habits.
I am a sleepy girlie trying to become a morning person. Sadly, TikTok has made me try a number of things — and most of them have been misses rather than hits. But like a moth to an aesthetic flame, I called in the trendy gadget. You never know. Every once in a while, the influencers are onto something.
#GRWM, Hatch Restore 2 edition
Picture this. You are a tired husk of a human, doomscrolling on your TikTok For You page. A conventionally attractive woman with luminous skin and luscious hair brushes her teeth, does her 10-step skincare routine, and before crawling into bed, taps her Restore 2. Soothing rain sounds emanate from the device, and she falls into a restful slumber. The next morning, she stretches a perfectly manicured hand to turn off her alarm on the Restore 2 before making a matcha latte in a mason jar. (Never mind how ridiculous it must look to film this kind of video.)
Watch enough GRWM videos on TikTok, and you’ll probably come across the Hatch Restore 2.
I watched dozens of videos like this before the Restore 2 ever arrived on my doorstep. I saw dozens more while reviewing this thing. So believe me when I say the reality is not quite as glamorous, but it’s not too far off from the truth, either.
Here is a day in my life using the Restore 2.
I wake up at 6:15AM, even though my alarm is set for 6:30AM. This used to be because my rotund cat Pablo would perch over my head, meowing loudly for his morning kibble. Instead, I’m waking up because there’s a bright-but-not-blinding white light beaming into my face from the Restore 2. For a second, I’m quite pleased that I woke up before the alarm and mildly amazed that this keeps happening. Seeing that I’m awake, now Pablo meows. I stretch a hand (with chipped nail polish) and press the right button to turn off my alarm.
I used to press the same button a second time to start a guided morning stretch in bed. But I’m married to a night owl, and they complained about it waking them up too many times. There’s no way to pair headphones to the Restore 2, so I only get to do my guided stretches when they’re away on business. I am begrudgingly up, my spouse is snoring, and Pablo still begs to be fed. I remind the bugger that he has a fancy automatic feeder that will shower him with kibble at 6:25AM.
I brush my teeth, take my meds, and make some pre-workout — another product that TikTok convinced me to buy. At 6:30AM, cheery retro lo-fi beats emerge from the Restore 2. It turns out, I only pressed the left button instead of holding it down, so I didn’t actually turn the alarm off. I marvel at how crisp it sounds until my thoughts are interrupted by a groan from my spouse. Oops. I apologize and promise to try to turn it off correctly tomorrow. I go do my 30- to 60-minute workout, take a shower, drink a protein shake, and sit down for work for the next eight to 10 hours.
The light is diffused but bright enough to wake you up.
After blindly shoveling dinner into my gob, I lose track of how I spend the hours of 7PM to 9PM. At this point, I set out my workout clothes for the next day, brush my teeth, slather on skincare to fight the ravages of time, and if I’m good, I settle into bed to read. If I’m bad, I doomscroll on TikTok. When my eyes get droopy, I tap the left button on top of the Restore 2, which triggers the nighttime routine I set in the app. The light turns orangey-red, and I hear choo-choo noises. Some lady tells me stories about trains, and I conk out. (I won’t remember much about the story other than she asks imaginary passengers for their tickets a lot.) When Pablo wakes me up at 3AM because cats are gonna cat, the Restore 2 is making soothing rain noises. I fall back asleep. Rinse and repeat.
So yeah, the Restore 2 works well for me. The light consistently wakes me up before my alarm, and the sleep stories / white noise sounds knock me out at night. I may never be #thatgirl, but if I’m honest, I’m too old for that shit anyway. I’m just happy to make progress on my journey to being an early bird.
A well-designed alarm clock
The Restore 2 isn’t just effective. It’s a good lookin’ device that fades into the background but in an aesthetically pleasing way. I mean, look at this thing. It feels like it was plucked straight out of a West Elm catalog while the rest of my bedroom hails from the IKEA or Wayfair bargain bin.
It helps that the front is covered in a natural linen fabric, which gives it a modern, Japandi vibe. The fabric also helps diffuse the light like a lampshade would (as you can see in the above photo), compared to a more traditional sunrise lamp. It comes in three colors: slate, latte, and putty. My review unit is latte, but the others are equally neutral tones that should blend into whatever your bedroom’s color theme or aesthetic is.
These are the two buttons you’ll use the most. The left is your nighttime routine button, while the right controls your morning alarm.
But as pretty as the Restore 2 is, I was skeptical when I first pulled it out of the box. I’ve been burned by plenty of devices that prioritize design over function. My biggest worry with the Restore 2 was the fact that there aren’t obvious controls when you look at it straight on. The controls do take a few nights to master, but they ended up being intuitive to use overall.
Up top, there are two tactile buttons. The left, for your nighttime routines, has a shut-eye design, while the right has an open eye for your morning alarm and moments, Hatch’s term for curated playlists. What I like most is that the nighttime button is concave, while the morning button protrudes a bit. I don’t have the best memory in the morning or when I’m sleepy, so it’s helpful to be able to distinguish them by touch.
On either side of the time, you can tap the device to raise (right side) or lower (left side) the volume.
The back also features the natural linen fabric, but you probably won’t look at it much.
On the left side of the Restore 2, you’ll find a small toggle. That lets you turn the alarm off and on without having to ever look at the app. I never really used it all that much, mostly because I forgot it was there. At the front of the device, on either side of the time display, you’ll find the invisible volume controls. Tapping the left side lowers the volume, while tapping the right side increases it. If at any time you find the clock to be too bright, you can always tap the whole thing to dim it.
A definite upgrade with a bigger library
The Restore 2 is an upgrade over its predecessor in every sense. It has a more appealing design, comes in more colors, and has three speakers, compared to the original Restore’s one. It also has several remastered sounds, 10 new sunrise alarms, new light and sound pairings for sunsets, 21 new sleep sounds, and a new “morning moment” option. The latter is a series of affirmations, stretches, or motivational content to get you going in the morning (provided you don’t have a sleep partner with a different schedule).
The Hatch app is pretty easy to use once you get the hang of it, and I appreciate that you can preview how things will look and sound before you add it to your routines. That said, once you find something that works for you, there’s not much reason to use the app unless you want to switch things up. In the month or so that I’ve been testing the Restore 2, I’ve only changed up my content twice — and that was mostly for review purposes. Left to my own devices, I probably wouldn’t have tried to switch things up for another few months.
There’s a pretty extensive content library in the Hatch app, but these “morning moments” playlists require a membership.
However, if you’re someone who likes to switch things up more often, there’s more than enough content to keep you satisfied. Within the app, there are multiple “channels” with curated playlists of music, dreamscapes, guided exercises, and sleep stories for both nighttime and morning routines. Each channel has multiple tracks, and Hatch also says it regularly updates the content library.
You don’t have to worry about getting bored with the lights, either. Not only are there preprogrammed color palettes to go along with the content, but there are also 18 separate colors to choose from, spanning the entire rainbow. I mostly stuck to soft whites and peachy oranges, but if you love the idea of a mint green morning alarm, go for it.
The only rub is that if you want to access content channels, you’ll need to shell out for a membership. You can either opt for a one-month trial and an annual membership of $49.99 or a one-week trial and a monthly membership of $4.99. But you don’t need to shell out. If all you want is sunrise alarm sounds and white noise options, those are available for free.
Is it worth it?
Buying stuff you see on TikTok is a game of roulette. On the one hand, I’ve found my favorite sunscreens, protein powder, and backpack on that app. But generally speaking, I’ve found most “Tiktok made me buy it” products haven’t lived up to the hype.
There’s an toggle on the side in case you just want to quickly turn the alarm on or off.
In the case of the Hatch Restore 2, I went in fairly skeptical. As much as I enjoy #GRWM videos, they have the potential to be the most insidious kind of advertising. It’s genius, really. Hatch stuck its products into the routines and habits of popular influencers, thereby subtly teaching millions that this device is the key to getting a good night’s rest. (This is literally how Febreze became a household item.)
Knowing all this, I’ve been pleasantly surprised at how much I like the Restore 2. The only things I really don’t like are the $200 price tag and the subscription. But, if the price isn’t a turnoff and you’re struggling to build good nighttime and morning routines, this is one of the most effective sleep gadgets I’ve used in a good long while. For once, I don’t regret that TikTok made me try it.
The List of Absolutely Everything That Might Kill You
Widely available, episodes weekly Fireworks, bees, volcanoes, theme parks … there are many dangers in life. DJ Matt Edmondson and This Is Going to Hurt author Adam Kay’s new podcast explores just how likely some things are to kill you. While Kay has seen a lot of death as a doctor, his co-host has only dealt with a deceased pet rabbit, so the contrast between the two offers up many moments of comedy. Hannah Verdier
Cybercrime: be careful what you tell your chatbot helper…
Alluring and useful they may be, but the AI interfaces’ potential as gateways for fraud and intrusive data gathering is huge – and is only set to grow
Concerns about the growing abilities of chatbots trained on large language models, such as OpenAI’s GPT-4, Google’s Bard and Microsoft’s Bing Chat, are making headlines. Experts warn of their ability to spread misinformation on a monumental scale, as well as the existential risk their development may pose to humanity. As if this isn’t worrying enough, a third area of concern has opened up – illustrated by Italy’s recent ban of ChatGPT on privacy grounds.
The Italian data regulator has voiced concerns over the model used by ChatGPT owner OpenAI and announced it would investigate whether the firm had broken strict European data protection laws.
Creative tech firm Talenthouse is close to failure as debts mount
Company that matches artists with brands is understood to have laid off most of its staff and faces legal action by creditors
A tech company that claimed to “democratise creativity” by matching artists with design briefs for major brands is on the brink of collapse after being issued with a winding-up petition over unpaid debts.
Talenthouse, whose clients have included Netflix, Coca-Cola, Nike and the UN, is facing legal action by creditors in the UK and is understood to have laid off most of its workforce, with top executives also departing its parent company in recent days.
Substack CEO pushes back at Elon, says Twitter situation is “very frustrating”
Quick recap: Substack, the popular subscription newsletter platform for writers, launched a new feature in its app called Notes last week, which is fairly similar to Twitter.
Twitter, the world-historical clown car of a company currently operated by Elon Musk, responded by blocking the ability to like or retweet any posts containing the word “Substack,” throwing up a warning message if users clicked on Substack links, and finally blocking even the wordSubstack from being searched. Substack users, the vast majority of whom are independent small business owners who depend on Substack as an enterprise software provider, mostly responded to this by saying they would leave Twitter and use Substack Notes. (There’s more here, but it’s a combo platter of hilarious and tiresome given some of the personalities involved. Just click the links.)
Musk sort-of explained his decision in a reply this week by claiming that Substack was “trying to download a massive portion of the Twitter database to bootstrap their Twitter clone” which is very funny — just imagine starting any sort of Twitter competitor and prepopulating itwith the garbage of Twitter — and also a pretty huge claim to make without any evidence.
The Substack Notes post from CEO Chris Best.
Anyhow, Substack CEO Chris Best responded to Elon with a post on Substack Notes, of course. The feature in in beta so there’s no permalink as yet, but he passed it along to us and we’re posting a screenshot so people can refer to it.
Best says Substack has used the Twitter API for years and thinks they’re in compliance with the terms of service; the implication is that Twitter never told the company about any alleged violation. And of course, Best says the whole situation is “very frustrating,” with a nod to the fact that Substack writers are actually customers of his enterprise software product; messing with Substack links has more impact on individual small business owners than it does on the platform company.
It is worth noting here that Andreesen Horowitz is an investor in both Substack and neo-Twitter, so that’s going great. And if you’re reading this you can probably just react to the phrases “free speech,” “Twitter,” and “Elon Musk” in whatever way kicks your Saturday night off right. I trust you. Party hard.
TikTok to a Heinz jar: vodka pasta sauce’s journey from fad to supermarket shelves
Whether it’s celebrity fads or wacky creations such as lasagne-filled yorkshire pudding, how food plays online is now key to brands
Sometimes you just need some vodka pasta sauce – especially when you’ve been grounded for trying to charter a helicopter on your dad’s credit card.
When Sofia Coppola’s 16-year-old daughter, Romy Mars, set the internet alight last month after doing just that – and then recorded a TikTok video to complain about her punishment while making the sauce – there was plenty of discussion about her family dynamics. Others, though, were more interested in her food choices.
It is the home of dance tutorial videos and viral comedy sketches. But it is also host to self-harm and eating disorder content, with an algorithm that has been called the “crack cocaine of social media”.
Now, the information commissioner has concluded that up to 1.4 million children under the age of 13 have been allowed access to TikTok, with the watchdog accusing the Chinese firm of not doing enough to check underage children were not using the app.
See How Real AI-Generated Images Have Become Human eyes — and even technology — often struggle to identify images created by artificial intelligence. Experts fear that may hasten an erosion of trust in media, in government and in society.
Why Pope Francis Is the Star of AI-Generated Photos Francis has become a recurring favorite to show in incongruous situations, such as riding a motorcycle and attending Burning Man, in A.I.-generated images.
‘War of attrition’: why union victories for US workers at Amazon have stalled
A year after a ‘historic’ victory in Staten Island, New York, hope for a wave of union victories is looking less momentous
A year ago, Amazon workers in Staten Island, New York won a “historic” victory – overcoming a multimillion-dollar campaign by the multibillion-dollar corporation to win the right to organize Amazon’s first-ever union.
A year on from that victory – which labor leaders had hoped would trigger a wave of union victories – is looking less momentous and another union election win at Amazon has remained elusive.
Twitter is now marking Substack links as unsafe
Illustration: Alex Castro / The Verge
Twitter has started marking links to Substack as unsafe. If you click on a link on Twitter with substack.com in the URL, Twitter will show a separate notice warning you that “the link you are trying to access has been identified by Twitter or our partners as being potentially spammy or unsafe.”
Don’t be alarmed — the links we’ve checked appear to be perfectly safe. This notice seems instead to be Twitter’s latest hostile move toward Substack after the email newsletter platform announced its own Twitter-like “Notes” feature on Wednesday.
On Thursday, Twitter blocked people who use Substack from embedding tweets into their stories. Then, late Thursday or early Friday, Twitter started blocking engagement on tweets containing links to Substack; users weren’t able to like or retweet them, but could quote retweet them. Then, on Friday morning, Twitter applied those same restrictions to tweets from the official Substack account.
The warning you get when trying to follow a Substack link from Twitter.
While those moves are clearly unfriendly, I’d classify the unsafe warning as downright aggressive. Twitter’s URL policy hasn’t changed since 2020, according to the last updated note on the page, and Substack’s site appears to be working as it always has, so it’s hard to say what justification Twitter’s using for the warning. CEO Elon Musk hasn’t tweeted anything about Substack recently and, when reached for comment, Twitter’s press email auto-replied with a poop emoji, which it has been doing since mid-March.
The elephant in the room is Substack’s Notes feature, which adds very Twitter-like elements to the newsletter platform. Matt Taibbi, a journalist who has history reporting stories for Musk, said Friday that he was told by an unspecified party that “Twitter is upset about the new Substack Notes feature, which they see as a hostile rival.” He also notes that he was “given the option of posting my articles on Twitter instead of Substack.”
In December, Musk said he considered “relentless advertising of competitors” to be a policy violation, and blocked sharing links to Instagram, Mastodon, Facebook, and others. However, those restrictions were later removed. It’s possible that he’s decided to bring it back now that he feels like Substack is trying to compete with Twitter, though it is worth noting that, for now at least, you still are able to tweet a Substack link. Anyone trying to follow it will just have to click through the warning to get to the content.
Substack didn’t immediately respond to The Verge’s request for comment on the warning. However, its founders have recently spoken out about how Twitter is treating their platform.
Microsoft locks down game emulation on the Xbox Series X and Series S
Illustration: Alex Castro / The Verge
Microsoft is barring users from running game emulations on the Xbox Series X | S. On Thursday, Twitter user @gamr12, who’s involved with the distribution of the RetroArch emulation software on Xbox, posted the error message they received when attempting to launch emulated content.
“Unable to launch this game or app,” the message reads. “The game or app you’re trying to launch violates Microsoft Store policy and is not supported.” Other userswith emulation software on the Xbox Series X | S report running into the same issue.
As pointed out by @gamr12, you can still emulate games on the Xbox Series X | S, but only if you put the device in Developer Mode, which you have to pay for. Microsoft seems to have only gotten rid of the option when the console’s put in Retail Mode, something all users can switch on for free with a little technical know-how.
While it’s still not clear what prompted the change, Alyanna, an active emulator fan who says she is a Microsoft Azure developer, claims she contacted an unnamed “friend at the Xbox QA team” about the issue weeks ago, who said the reason for the ban is Nintendo.
While the source and claims of the message haven’t been verified or confirmed, it reads, “The primary reason for the ban is related to legal issues with Nintendo.” “While emulating itself is not illegal, it can be used to play games from consoles that are still under copyright protection without permission, which can create issues with Nintendo and its affiliates.”
Microsoft’s rules technically don’t allow emulations, but the company typically looked the other way in the past, according to Kotaku. “We continually evolve our mechanisms for reviewing and taking enforcement actions on content distributed to the Store to ensure alignment with our Microsoft Store Policies,” Microsoft tells Kotaku. “Per 10.13.10, Products that emulate a game system or game platform are not allowed on any device family,” Microsoft says. The company didn’t immediately respond to The Verge’s request for comment.
Social media analyst Emily Hund: ‘We can never know the truth behind an influencer’s seeming authenticity’
Today influencers sell ideas about science and medicine as well as products. But the integrity on which their status rests, says the US author, is as unknowable as the algorithms that push their content
In the early 00s, Emily Hund dreamed of a career as a journalist at a glossy fashion magazine. But after internships with New York media companies and having witnessed falling circulations and redundancies, she switched to studying one of the catalysts for these changes: social media and the influencers whose YouTube, TikTok and Instagram posts sell ideas, lifestyles and products to their followers. The influencer industry ranges from global stars such as the Kardashians to micro-influencers who post on niche interests. What they have in common is that they work with brands to promote or sell to an audience. Hund is now a research affiliate at Pennsylvania University’s Centre on Digital Culture and Society and her first book on influencers is published in the UK this month.
How did social media take hold in people’s lives? There was a lot of optimism about social media in the 00s when technology first made it easier to share opinions. During the economic crisis of 2008, when people were out of work or looking for ways to make money, it really took off. Bloggers found loyal audiences, so advertisers got interested. This was all happening against a crisis in traditional media and they were looking for new ways to promote products.
Samsung’s having a terrible financial year
Samsung's chip business is killing profits. | Photo by JUNG YEON-JE/AFP via Getty Images
Samsung Electronics is having another bad quarter according to the company’s own preliminary estimates. This time, it’s warning that quarterly operating profit decreased 96 per cent compared to the same period last year, worse than the two-thirds plunge it suffered the three months prior. Not even robust sales of the new Galaxy S23 series could overcome a global plummet in chip demand.
Semiconductor demand has waned since peak covid as everyone splurged on lockdown toys and tools. Now the global economic slowdown has caused consumers to think twice about buying that next gadget resulting in swollen chip inventories for companies like Samsung. And according to my ECON101 professor, prices will fall when there’s more supply than demand putting the hurt on corporate profits. And you can’t get a pardon for that.
But don’t feel too badly for Samsung. It still made a ton of money. Operating profit is estimated at 600 billion won (about $456 million) — the lowest it’s been in 14 years, reports the Financial Times. It’s just not the staggering 14.12 trillion Korean won (about $10.7 billion) raked in last year at this time.
As a result, Samsung says it’s cutting memory chip production to a “meaningful level” to address the oversupply. “We have cut short-term production plans, but as we project solid demand for the mid-to-long term, we will continue to invest in infrastructure to secure essential cleanrooms and to expand R&D investment to solidify tech leadership,” Samsung said in a statement.
One bright spot for the quarter was the new Galaxy S23 series, with sales of about 11 million units. That’s up by 50 percent compared to its predecessor, according to Hanwha Investment & Securities analyst Kim Kwangjin as reported by Bloomberg.
Samsung will issue its final financial statement in a few weeks, but it typically diverges very little, if at all, from preliminary guidance.
This economist won every bet he made on the future. Then he tested ChatGPT
Bryan Caplan was skeptical after AI struggled on his midterm exam. But within months, it had aced the test
The economist Bryan Caplan was sure the artificial intelligence baked into ChatGPT wasn’t as smart as it was cracked up to be. The question: could the AI ace his undergraduate class’s 2022 midterm exam?
Caplan, of George Mason University in Virginia, seemed in a good position to judge. He has made a name for himself by placing bets on a range of newsworthy topics, from Donald Trump’s electoral chances in 2016 to future US college attendance rates. And he nearly always wins, often by betting against predictions he views as hyperbolic.
Senior Twitter Lawyer Resigns, the Latest in a Series of Executive Departures The lawyer had been closely involved in the company’s talks with the Federal Trade Commission over promises it had made on privacy.
The impossibility of logging off
Image: Mengxin Li / The Verge
The logout button has become practically defunct.
On Twitter, the phrase “log off” is often uttered as a contemptuous command, directed toward a perennially online person who needs to put down their phone and go for a walk. Logging off, by this logic, is not a task to be done on one’s device but a mental state to inhabit and idealize. Still, there’s a deep irony to this sentiment: a person who manages to mentally log off, whether it be for an hour or a week, might still remain technically logged in. It’s common practice for users to leave open browser tabs and apps, for devices to ambiently operate in the background, collecting data. Most people don’t spare this fact a second thought. We’re accustomed to staying logged in so that we can continue scrolling exactly where we left off.
The interface of consumer websites and apps, in turn, has reflected this shift. People once prioritized logging in as much as logging out, but now, according to freelance UI designer Jesse Showalter, access to content is of utmost importance, even at the cost of constantly sharing our data. Logging out, by contrast, carries little value for companies or consumers.
The logout button seems to have been rendered practically defunct. I only purposefully sign out of certain accounts when I’m trying to curb my usage of a site or app (usually it’s Twitter or Amazon). Even then, that process isn’t always straightforward. A few months ago, I was using a friend’s laptop to send some emails and realized that I was prohibited from individually signing out of my Gmail account. Doing so would also log my friend out. This is a desktop-specific nuisance that Google has maintained for many years. Instead, I had to use a separate device, like a mobile phone, to revoke account access.
This unexpected logout hurdle only affirmed my conspiratorial suspicion: websites and apps have a profit-bearing incentive to keep users logged in, reflected in mobile and desktop interface design. The logout button, as a result, has been consigned to the depths of the settings menu, as is the case with Discord and YouTube, or even removed as a function entirely, like on the mobile apps for Facebook Messenger and WhatsApp. My theory aligns with the history of dark interface patterns quietly pushed by major tech companies. These UI decisions are made with corporate interests in mind “to trick users into doing things they might not otherwise do.” Concealing the logout function by a few clicks is admittedly not as manipulative as tricking a user to agree to be tracked. The logic seemed similar, though. People are less likely to consider signing out if the option isn’t quite so front and center. Otherwise, why would there be a genre of YouTube tutorials with thousands, if not millions, of views detailing the logout process of certain apps?
I reached out to some interface designers to see if my theory held any weight, only to learn that the reality is not as nefarious as I’ve made it out to be. User behavior informs design decisions and vice versa. Since our devices are treated as extensions of our cognitive selves, people shift constantly between mobile and desktop interfaces, even simultaneously using both. These days, users expect their session data to be seamlessly transferred across devices: “This productivity handover results in people needing to stay constantly logged in,” said Tom McLean, the lead UX designer at technology and design consultancy Door3. The login experience, too, becomes less of a hassle, and service providers like Google or Meta can gather browsing data and habits on their users, even when they’re not using the app.
This relationship is more symbiotic than antagonistic, McLean added. Most interfaces are designed to be user-friendly. Businesses benefit from customers engaging with their product. A positive interface experience increases the amount of time spent on an app, which translates to increased advertising revenue. In some cases, a user doesn’t necessarily need to make an account to be “logged in.” They simply have to permit a site to create a session cookie (usually by checking a box) that gathers data on the user’s activity. These cookies are often not cleared when the browser is closed, so the person can technically stay logged in for an extended period of time. Few users manually clear out their cache. In the end, convenience is always something the user pays for, said Sara Vienna, vice president of design at the interface design firm MetaLab, in the form of money or personal data for targeted ads.
Most sites aren’t purposefully sending users on a wild goose chase before logging out.
The sign-out process might seem more confusing or challenging because users are less familiar with contemporary interface designs, Vienna said, which have only grown more customized and sophisticated in the past decade. Designers are not as often relying on “cookie-cutter, bootstrap-like interfaces,” Vienna added. “When people pull from familiar design patterns, there’s a benefit from a UX perspective because familiarity means people know where to find things.”
Developers simply don’t think about logging out as much, unless they’re building platforms that contain sensitive user information. With healthcare or financial services apps, for example, logging out is an automated, built-in feature. User sessions are programmed to quickly time out or expire for privacy. Or in the case of news sites with paywalls, like The New Yorker and New York Magazine, users complain about being routinely booted out.
Vienna imagines an inevitable future where users will be more thoughtful about their data, demanding more control over how it’s used and collected. The side effect of that? An internet that allows for more states than the binary of being logged in or logged out. “When we think of states, it’s the way the interface should change and evolve depending on the user’s location, time, actions, and anticipated needs,” Vienna said.
Until then, when we take a break from our screens, leave our homes and go for a walk, to go touch grass even, our devices at home will remain logged in, with the promise of a seamless and continuous experience for whenever we might return — the expectation that, no matter what, we always will.
OneNote is getting Microsoft’s new AI Copilot to help you write your notes
Illustration: The Verge
Microsoft is planning to add its new Microsoft 365 Copilot assistant to OneNote. The software giant originally announced Copilot for apps like Teams, Word, Excel, and PowerPoint last month, and it’s now ready to demo how note taking will be impacted by an AI-powered assistant.
“As your notetaking partner, Copilot uses your prompts to draft plans, generate ideas, create lists, organize information, and more,” explains Greg Mace, a product manager for OneNote.
Image: Microsoft
How the Copilot integration works in OneNote.
Much like Copilot’s integration in Word, the AI assistant can rewrite, format, or summarize existing text in OneNote based on prompts in a sidebar interface. You can also ask the AI assistant to create a plan for an event or generate topics and talking points for meetings.
Microsoft hasn’t shown off Copilot’s inline integration in OneNote yet, though. In the Word version you can highlight paragraphs and the Copilot assistant will appear when you mouse over a section of the paragraph to offer up rewritten text automatically.
currently playing around with Microsoft's AI Copilot for Word, thanks to @XenoPanther. It's only supposed to be available in "private preview" for "select Microsoft 365 enterprise customers," but a few registry keys here and there and you can get the basics working pic.twitter.com/TSO1KgnqwH
Microsoft also hasn’t announced exactly when Copilot will be available in OneNote. A small number of Microsoft 365 enterprise customers are currently testing a preview version of Copilot in apps like Word and Outlook, and some of these early versions can be enabled in beta copies of Word right now.
I’ve been playing around with the private preview of Copilot in Word recently, using it to generate text. It’s a very early implementation right now, so there’s lots of missing functionality that needs sorting before it’s ready for public preview. Microsoft is testing this heavily before rolling it more broadly, and the company hasn’t committed to any dates for Copilot just yet.
Microsoft brings its Bing chatbot to your fingertips with SwiftKey on Android
Illustration: The Verge
Microsoft has added Bing Chat AI to its popular SwiftKey keyboard for Android. The integration started appearing on Wednesday and allows SwiftKey Beta users to quickly access Microsoft’s Bing chatbot at the tap of a button in any app.
The Bing Chat integration inside SwiftKey includes the usual chat mode that offers access to the chatbot, or a tone mode that will rewrite any text right within the keyboard. If you were only using the Bing mobile app to access the chatbot, this new integration makes that app rather redundant if, instead, you’re willing to install and use SwiftKey everywhere.
To access this new Bing Chat integration you’ll need to download the beta version of SwiftKey on the Google Play Store. Pedram Rezaei, Microsoft’s CTO of mobile and commerce, confirmed on Twitter that the feature is “slowly rolling out,” but there’s no word on when SwiftKey iOS users will be able to try it out.
Bing Chat is coming to SwiftKey. Download the latest SwiftKey beta and sign in to your MSA.
Reposted with cleaner screenshots. pic.twitter.com/dBss7gnOzn
Microsoft discontinued support for SwiftKey on iOS last year, before quickly returning the app to the App Store. At the time Rezaei said Microsoft will be “investing heavily in the keyboard,” just as the company was preparing to publicly launch an AI-powered version of Bing.
While the iOS version of SwiftKey hasn’t been updated regularly, Microsoft’s AI push will likely see this Bing integration appear on the iPhone keyboard side soon. Microsoft might also be responding to competition from a number of AI-powered keyboards on iOS that include ChatGPT integration. Paragraph AI includes an interface that lets you generate text everywhere and is already rising in popularity on iOS.
Microsoft originally acquired SwiftKey in 2016, after the third-party keyboard soared in popularity on Android and iOS. It was the second best Android keyboard in The Verge’s roundup in 2015, thanks to its tracing, autocorrect, and theme support.
/cdn.vox-cdn.com/uploads/chorus_asset/file/24576567/P90500155_highRes.jpg)
Image: BMW / Mini
/cdn.vox-cdn.com/uploads/chorus_asset/file/24576568/P90500156_highRes.jpg)
Image: BMW / Mini
/cdn.vox-cdn.com/uploads/chorus_asset/file/24575228/Eren_Jaeger_Outfit_and_Accessories.jpg)
Image: Epic Games
/cdn.vox-cdn.com/uploads/chorus_asset/file/24576515/8n88Ao1LaRfbfxOQuN88h279EruOXkeT.jpeg)
Image: Canalys
/cdn.vox-cdn.com/uploads/chorus_asset/file/24575935/download.jpeg)
Image: Elon Musk, in a now-deleted tweet/cdn.vox-cdn.com/uploads/chorus_asset/file/24575936/download__1_.jpeg)
Image: Elon Musk, in a now-deleted tweet
/cdn.vox-cdn.com/uploads/chorus_asset/file/24552171/20230328_162723.jpg)
/cdn.vox-cdn.com/uploads/chorus_asset/file/24552164/20230328_160428.jpg)
Photo: Mitchell Clark / The Verge
/cdn.vox-cdn.com/uploads/chorus_asset/file/24539793/236581_hatch_restore_2_AKrales_0062.jpg)
/cdn.vox-cdn.com/uploads/chorus_asset/file/24539799/236581_hatch_restore_2_AKrales_0121.jpg)
/cdn.vox-cdn.com/uploads/chorus_asset/file/24539794/236581_hatch_restore_2_AKrales_0073.jpg)
/cdn.vox-cdn.com/uploads/chorus_asset/file/24539792/236581_hatch_restore_2_AKrales_0051.jpg)
/cdn.vox-cdn.com/uploads/chorus_asset/file/24539798/236581_hatch_restore_2_AKrales_0108.jpg)
/cdn.vox-cdn.com/uploads/chorus_asset/file/24539801/236581_hatch_restore_2_AKrales_0151.jpg)
/cdn.vox-cdn.com/uploads/chorus_asset/file/24539796/236581_hatch_restore_2_AKrales_0095.jpg)
/cdn.vox-cdn.com/uploads/chorus_asset/file/24571807/IMG_0152.jpeg)
/cdn.vox-cdn.com/uploads/chorus_asset/file/24570431/Image_from_iOS.jpg)
/cdn.vox-cdn.com/uploads/chorus_asset/file/24565876/Copilot_SidePane_Blog2.gif)
Image: Microsoft
