Beyond qubits: Meet the qutrit (and ququart)

The world of computers is dominated by binary. Silicon transistors are either conducting or they’re not, and so we’ve developed a whole world of math and logical operations around those binary capabilities. And, for the most part, quantum computing has been developing along similar lines, using qubits that, when measured, will be found in one of two states.

In some cases, the use of binary values is a feature of the object being used to hold the qubit. For example, a technology called dual-rail qubits takes its value from which of two linked resonators holds a photon. But there are many other quantum objects that have access to far more than two states—think of something like all the possible energy states an electron could occupy when orbiting an atom. We can use things like this as qubits by only relying on the lowest two energy levels. But there’s nothing stopping us from using more than two.

In Wednesday’s issue of Nature, researchers describe creating qudits, the generic term for systems that hold quantum information—it’s short for quantum digits. Using a system that can be in three or four possible states (qutrits and ququarts, respectively), they demonstrate the first error correction of higher-order quantum memory.

Read full article

Comments

Krypto steals the show (again) in Superman trailer

David Corenswet stars as Clark Kent in director James Gunn’s Superman.

We’re about to enter a new era for DC Studios with the July release of Superman, writer/director James Gunn’s fresh take on one of the most iconic superheroes. And after months of tantalizing teases, we finally have the first official trailer, featuring a bickering Clark Kent and Lois Lane, plenty of action, villains being villains, kaiju, and of course, our favorite flying super-pup, Krypto.

As previously reported, Gunn has described his take as less of an origin story and more of a journey, with Superman struggling to reconcile his Kryptonian heritage and aristocratic origins with his small-town adoptive human family. Gunn tapped David Corenswet to play Clark Kent/Superman at 25, a bit more established than the young cub reporter of Smallville, for instance. Rachel Brosnahan plays Lois Lane, Skyler Gisondo plays Jimmy Olsen, and Nicholas Hoult is arch-nemesis Lex Luthor. Luthor’s sidekicks are played by Sara Sampaio as Eve Teschmacher and Terence Rosemore as Otis.

The cast also includes Nathan Fillion as Guy Gardner/Green Lantern (sporting a disastrous bowl haircut); Anthony Carrigan as Rex Mason/Metamorpho, who can transmute elements in his body to change forms; Isabela Merced as Hawkgirl; Edi Gathegi as Michael Holt/Mister Terrific, an inventor turned superhero; Maria Gabriela de Faria as Angela Spica/The Engineer, whose abilities stem from embedded nanotechnology; and Pruitt Taylor Vince and Neva Howell as Clark’s parents, Jonathan and Martha Kent, respectively. We’ll also see Frank Grillo reprise his role as Rick Flag Sr. from the animated series Creature Commandos; Sean Gunn as Maxwell Lord; and Milly Alcock as Superman’s cousin, Kara Zor-El/Supergirl.

Read full article

Comments

Netflix is bringing back ‘Star Search’ as it continues to expand into live TV

Star Search, a reality competition series that’s featured a shocking number of musicians and comedians before they were famous, is coming back as a Netflix show. The streaming platform announced that it’s bringing Star Search back as a live show with interactive elements, building on Netflix’s expansion into live TV and events.

Like the original, the new Star Search will focus on “emerging performers in music, dance, comedy and kids’ acts,” Netflix says. The company didn’t share who would host or judge these acts, and doesn’t have a premiere data as of yet, but Netflix did claim that the new Star Search will be “more interactive than ever.” Past seasons of the show allowed the at-home audience to rate each performance before their scores were read live on air. It’s not hard to imagine Netflix rigging something similar inside its own apps.

Reality competition shows are a dime a dozen in 2025, particularly on Netflix, but Star Search is notable for the sheer number of famous people who competed on it. Like American Idol, not many of them actually won, but it’s rare to see a competition show that’s featured everyone from Christina Aguilera to Sinbad.

Star Search is just one part of Netflix’s growing ambitions for live TV, which now includes everything thing from alternative talk shows to WWE. The company streamed its first NFL game on Christmas Day 2024, and has continued to expand into other live events, like award shows. At the same time, Netflix’s plans for gaming have seemingly diminished — the company closed its AAA game studio and has removed interactive shows like Bandersnatch from its catalog.

Casting has already begun for the new version of Star Search. You can head to Netflix’s website for more information and to apply to be on the show.

This article originally appeared on Engadget at https://www.engadget.com/entertainment/streaming/netflix-is-bringing-back-star-search-as-it-continues-to-expand-into-live-tv-214022695.html?src=rss

Uber To Introduce Fixed-Route Shuttles In Major US Cities

Uber is launching a fixed-route shuttle service in major U.S. cities that offers commuters up to 50% off UberX fares during weekday peak hours. Called “Route Share,” the service aims to provide a more affordable, predictable alternative to standard ride-hailing. TechCrunch reports: The commuter shuttles will drive between pre-set stops every 20 minutes, according to Sachin Kansal, Uber’s chief product officer. He noted that there will be dozens of routes in each launch city — like between Williamsburg and Midtown in NYC. The routes, which are selected based on Uber’s extensive data on popular travel patterns, might have one or two additional stops to pick up other passengers. To start, riders will only ever have to share the route with up to two other co-riders.

Riders can book a seat anywhere from seven days to 10 minutes before a scheduled pickup, and the app will provide them with turn-by-turn directions to get them from their house to the corner where they’ll be picked up. Uber is relying on the same underlying technology that it uses for Uber Share, its shared rides offering where riders can get 15% to 30% off the cost of an UberX ride by pooling with others. Kansal told TechCrunch that Uber completes millions of shared trips annually and has been seeing more traction lately as riders look for more ways to save. Hence, Route Share.

Uber envisions a future where Route Share could qualify for pre-tax commuter benefits. However, as a spokesperson noted, the company would need to find a way to match those trips with Uber XL vehicles. That’s because only six-seater vehicles would meet the eligibility requirements. A potential progression of Route Share would involve autonomous vehicles, particularly in chaotic cities like New York City, where no self-driving car companies have deigned to test.


Read more of this story at Slashdot.

Now’s a good time to check in on your Steam account security

Update May 14, 6:38PM ET: Valve has confirmed that Steam systems were not breached and identifying user data has not been stolen or accessed by hackers. The company provided clarification in a Steam blog post:

“We’re still digging into the source of the leak, which is compounded by the fact that any SMS messages are unencrypted in transit, and routed through multiple providers on the way to your phone. The leak consisted of older text messages that included one-time codes that were only valid for 15-minute time frames and the phone numbers they were sent to. The leaked data did not associate the phone numbers with a Steam account, password information, payment information or other personal data. Old text messages cannot be used to breach the security of your Steam account.”

Our original story follows.


Steam has allegedly suffered a data breach in the past week. Details are scant and difficult to confirm, but a known hacker has claimed to be selling a database of more than 89 million user records for the gaming platform with one-time access codes obtained from a third-party vendor used by Steam. If accurate, that would include information about more than two-thirds of Steam’s audience. 

The original LinkedIn post identifying a breach suggested that the leaked information came from cloud communication company Twilio. However, a Steam rep said the platform doesn’t use Twilio, so if there has been a breach, it may be through a different vendor providing SMS codes for access.

While we’re genuinely not sure what’s happening at this stage, the whole kerfuffle is a timely reminder to check in on your online security practices. In the case of Steam, Valve has a mobile authentication program called Steam Guard that can help keep your account secure. It’s also a good practice to make sure you’re regularly changing your pass codes, especially when it’s possible that some component of Steam Guard was at the root of this week’s security drama. A password manager can streamline that process. Since phone numbers appear to have been compromised, be extra alert to possible phishing attempts via text. 

This article originally appeared on Engadget at https://www.engadget.com/gaming/nows-a-good-time-to-check-in-on-your-steam-account-security-205807479.html?src=rss

World Record For Most Playing Cards Knocked Over Like Dominos

This is a video (including some stunning shots through the glass roof) of the world record for most playing cards being toppled like dominos, with a very impressive 51,725 slightly bent cards getting knocked over. Damn, and I thought standing dominos was nerve-racking, this must take a REALLY special kind of person. Which I am not, just to be clear. I am very average. Well, except in the…you know. “Idiot department.” Exactly.

SoundCloud backtracks on ‘too broad’ AI terms of service

SoundCloud is updating its Terms of Use again after angering users with language around AI that even the company now describes as “too broad.” The details of the change were shared in an open letter from SoundCloud CEO Eliah Seton affirming the company’s commitment to artists.

Specifically, SoundCloud’s Terms of Use now forbids the company from using content uploaded to SoundCloud to train generative AI that replicates an artist without their consent. As it’s phrased in the new terms SoundCloud is rolling out in the next few weeks:

We will not use Your Content to train generative AI models that aim to replicate or synthesize your voice, music, or likeness without your explicit consent, which must be affirmatively provided through an opt-in mechanism.

Seton also reiterated that Soundcloud has never used “artist content” to train AI. “Not for music creation. Not for large language models. Not for anything that tries to mimic or replace your work,” Seton writes.

The conflict over SoundCloud’s approach to AI started when users noticed that the company had updated its Terms of Use in February 2024 to allow SoundCloud to use content to “inform, train, develop or serve as input to artificial intelligence or machine intelligence technologies or services as part of and for providing the services.” 

SoundCloud claims that it would only ever use artist’s content to train AI music discovery tools. The company acquired Musiio in 2022 with that exact idea in mind. Still, it’s understandable that SoundCloud users would be sensitive to any AI-related changes the company makes. Like most other content stored online, AI companies have scraped music without artists’ consent and been fairly open about it.

This article originally appeared on Engadget at https://www.engadget.com/entertainment/music/soundcloud-backtracks-on-too-broad-ai-terms-of-service-201023013.html?src=rss

OpenAI rolls out GPT-4.1 to all ChatGPT paying subscribers

OpenAI is making GPT-4.1, the latest addition to its collection of AI models, available to all of its paid users. The rollout to ChatGPT Plus, Pro and Team members begins today, while Enterprise and Edu subscriptions will have to wait a few weeks to get access. Once available, GPT-4.1 can be chosen from the “more models” dropdown menu in the model picker.

This version of the ChatGPT AI chatbot rolled out to the company’s developer API last month, offering upgrades in speed and efficiency over the GPT-4.5 model that it replaced (yes, that is a slightly convoluted numbering system). OpenAI is iterating models at a breakneck speed this year; GPT-4.5 just debuted in February and is already being outpaced in stats by a new version.

The company also has a GPT-4.1 mini option that will replace the current small model of GPT-4o mini. That upgrade will happen for all tiers, including the free one. GPT-4.1 mini will be the fallback option for free users once they reach the cap on their access to the heftier GPT-4o model.

This article originally appeared on Engadget at https://www.engadget.com/ai/openai-rolls-out-gpt-41-to-all-chatgpt-paying-subscribers-200049384.html?src=rss

Mission: Impossible should never have gone full sci-fi

The Mission: Impossible film franchise has always dabbled in the, well, impossible. We’ve seen Tom Cruise’s Ethan Hunt climb his way up the Burj Khalifa, have a motorcycle joust to prevent the spread of a bioweapon and hang off the side of an airplane. Even the most grounded entry, Brian DePalma’s 1996 Mission: Impossible, featured Cruise leaping off of an exploding helicopter onto a train in the Chunnel. But with the previous film, Dead Reckoning, and this year’s follow-up Final Reckoning, the series has jumped completely into science-fiction territory with an AI villain called The Entity. It has the power to control anything that touches the internet, manipulate digital information to suit its needs and potentially wipe out humanity through a global nuclear annihilation.

Stopping the Entity is a mission Ethan Hunt has no choice but to accept. But as a fan of this series from the start — hell, I even like John Woo’s gloriously operatic Mission: Impossible 2 — I can’t help but see the move into true sci-fi as a huge mistake. It makes both Reckoning films far too plot-heavy and impenetrable (Final Reckoning clocks in at three hours!), and they also just don’t have much to say about AI beyond a Terminator-esque extinction scenario. But perhaps worst of all, the shift towards sci-fi inadvertently (or perhaps purposefully) turns Ethan Hunt into some sort of Messiah. Apparently, only Scientology’s greatest son can save us.

The best M:I films are the ones that don’t get bogged down in the intricacies of plot mechanics. That’s a trend that truly kicked off with the JJ Abrams-directed Mission: Impossible 3, which relied on a standard MacGuffin (the “Rabbit’s Foot”) and a powerhouse Philip Seymour Hoffman villain performance to send Ethan and his team gallivanting around the world. With Ghost Protocol, director Brad Bird used his experience in animation and love of silent film to turn Cruise into a modern-day Buster Keaton, hopping from one elaborate set-piece to another.

Mission: Impossible - The Final Reckoning
Paramount Pictures

The series found a new life when writer/director Christopher McQuarrie hopped aboard for Rogue Nation, which introduced Rebecca Ferguson’s enigmatic Ilsa Faust. McQuarrie has previously likened his approach to the franchise as something like action film jazz, wherein he and Cruise would develop some set piece ideas and build a narrative around that. At the same time, they also sought to develop Hunt’s inner-life and team dynamics more than previous films. Plot, once again, was mostly a vehicle to reach those spectacular action set pieces and character-defining moments (which were often one and the same).

McQuarrie mostly repeated his formula for success with 2018’s Mission: Impossible – Fallout, which was notable for featuring a real-time high altitude skydiving sequence. But with 2023’s Dead Reckoning, he faced the limits of trying to improvise a movie as it was being shot. Production was significantly delayed by the pandemic, and the film also had to go through several reshoots. Perhaps not surprisingly, it also became increasingly more complex and plot-heavy.

Mission: Impossible - The Final Reckoning
Paramount Pictures

That movie couldn’t just treat The Entity’s AI like another plot MacGuffin, instead it practically became an anchor for the film’s momentum. We had to learn what the Entity was, why it could be bad and also introduce new characters who were devoted to its ambitions. The final film feels like a hodgepodge of ideas trying to string together a few notable action sequences, like that aforementioned motorcycle jump. The prolonged production also led to the departure of Ilsa Faust, who was immediately replaced by Hayley Atwell’s Grace, an expert thief who’s so thinly sketched she doesn’t even get a last name.

I had hoped that McQuarrie, Cruise and co-writer Erik Jendresen would learn from the sloppiness of the last film and refocus on the characters and action we love in The Final Reckoning, but unfortunately things get even more convoluted. We’re presented with a world where the Entity has already taken over most information systems, can easily reshape digital reality at will and is in the process of taking over nuclear weapons systems around the world. There is no hope but Ethan Hunt, who must seek out the Entity’s source code in a sunken Russian submarine and try to stop it from annihilating humanity (while also trying to survive the apocalypse in an underground data bunker). And if that all sounds tiring as you read it, it’s even harder to swallow as you sit through the film’s three-hour runtime.

Mission: Impossible - The Final Reckoning
Paramount Pictures

Once the film actually starts moving around the half-way point, it delivers some of the most complex set pieces we’ve seen yet. Hunt’s dive into Arctic waters feels as claustrophobic as some of the best scenes from The Abyss, and it’s still thrilling to see Cruise hang onto bi-planes during the climactic chase. I just wish it actually did something interesting with the AI at the center of the story, instead of giving us a basic-ass Terminator/Wargames scenario. We’re told that the Entity has inspired a cult-like following, and that it can completely reshape the idea of truth, but we don’t actually see how it affects people around the world.

That’s a particular shame since the Mission: Impossible series’ has always been about genuine human effort, you’d think McQuarrie and crew would actually have more to say about the impact of AI. Fans want to see practical stunt work being accomplished by a movie star who’s desperate for attention. Now with real-world AI threatening to dumb down the act of creativity and recycle existing content, turning the film’s AI into a simplistic villain just seems like a total waste.

Mission: Impossible - The Final Reckoning
Paramount Pictures

The Final Reckoning also wastes far too much time extolling Ethan Hunt’s virtues as humanity’s savior. No government can be trusted, no elected leaders — just one man who never follows orders. The one man who has given up love and bled for an ungrateful world. Even the people whose lives he has ostensibly ruined can’t help but love him. 

The Mission: Impossible franchise has always been a vanity project for Cruise, but he also balanced out his ego by working with talented directors who pushed him and the series in new directions. Now, in his fourth film with McQuarrie, and possibly his last as the main character, Cruise can’t help but remind us how much he’s suffered. And it’s as dull as yet another world-ending AI villain.

This article originally appeared on Engadget at https://www.engadget.com/entertainment/tv-movies/mission-impossible-should-never-have-gone-full-sci-fi-200043337.html?src=rss

VPN owner shows buyer’s remorse; claims was unaware of lifetime subscriptions

The CEO of the company that purchased VPNSecure in 2023 and claimed to not know that the VPN service provider had previously sold lifetime subscriptions has some regrets.

Earlier this week, Ars Technica reported on VPNSecure canceling thousands of lifetime subscriptions, starting in March. In an email to customers, VPNSecure said that it couldn’t afford to maintain the subscriptions and that the current owners, InfiniteQuant Ltd, weren’t told about the subscriptions when they bought VPNSecure. The sudden deactivation of accounts resulted in customer backlash online, including, as of this writing, 24 pages of one-star reviews on Trustpilot.

“… maybe, honestly, we should have just walked away from this ‘opportunity,’” Romain Brabant, the CEO of InfiniteQuant Ltd, told Ars Technica when asked if he would have handled things differently in hindsight.

Read full article

Comments

A privately developed Australian rocket is ready for a historic launch

Gilmour Space, a venture-backed startup based in Australia, is about to launch a small rocket from its privately owned spaceport on a remote stretch of the country’s northeastern coastline.

It’s the first time anyone has attempted to reach orbit with a rocket designed and built in Australia. Gilmour’s three-stage rocket, named Eris, could launch at any time during a 10-hour window as soon as Friday, local time. In the United States, the launch window runs from 5:30 pm EDT Thursday until 3:30 am EDT Friday.

This comes after a 24-hour delay due to an “issue in the ground support system” for the launch, Gilmour said Wednesday.

Read full article

Comments

Judge admits nearly being persuaded by AI hallucinations in court filing

A plaintiff’s law firms were sanctioned and ordered to pay $31,100 after submitting fake AI citations that nearly ended up in a court ruling. Michael Wilner, a retired US magistrate judge serving as special master in US District Court for the Central District of California, admitted that he initially thought the citations were real and “almost” put them into an order.

These aren’t the first lawyers caught submitting briefs with fake citations generated by AI. In some cases, opposing attorneys figure out what happened and notify the judge. In this instance, the judge noticed that some citations were unverifiable but was troubled by how close he came to including the bogus citations in an order.

“Directly put, Plaintiff’s use of AI affirmatively misled me,” Judge Wilner wrote in a May 5 order. “I read their brief, was persuaded (or at least intrigued) by the authorities that they cited, and looked up the decisions to learn more about them—only to find that they didn’t exist. That’s scary. It almost led to the scarier outcome (from my perspective) of including those bogus materials in a judicial order. Strong deterrence is needed to make sure that attorneys don’t succumb to this easy shortcut.”

Read full article

Comments

Valve takes another step toward making SteamOS a true Windows competitor

We’ve known for months now that Valve is expanding its Linux-based SteamOS operating system beyond the Steam Deck to other handheld PCs, starting with some versions of the Asus ROG Ally. This week, Valve began making some changes to its Steam storefront to prepare for a future when the Deck isn’t the only hardware running SteamOS.

A new “SteamOS Compatible” label will begin rolling out “over the next few weeks” to denote “whether a game and all of its middleware is supported on SteamOS,” including “game functionality, launcher functionality, and anti-cheat support.” Games that don’t meet this requirement will be marked as “SteamOS Unsupported.” As with current games and the Steam Deck, this label doesn’t mean these games won’t run, but it does mean there may be some serious compatibility issues that keep the game from running as intended.

Valve says that “over 18,000 titles on Steam [will] be marked SteamOS compatible out of the gate,” and that game developers won’t need to do anything extra to earn the label if their titles already support the Steam Deck.

Read full article

Comments

ProtonVPN subscriptions are up to 66 percent off right now

ProtonVPN subscriptions are available at a steep discount right now as part of an exclusive sale for Engadget readers. A 12-month subscription is down to $48, which is a discount of around $72 and works out to $4 per month. A 24-month plan now costs just $81. This represents a massive discount of $158 and works out to $3.39 per month.

Proton topped our list of the best VPN services, and with good reason. It’s incredibly powerful and easy to use, which is a boon for those new to the space. The end-to-end encryption is solid and everything’s based on an open-source framework. This lets the company offer an official vulnerability disclosure program.

A subscription includes an IP-masker, so websites can’t track you online, and a built-in ad blocker. We found in our tests that browsing the web and watching streaming content were both speedy while using this VPN, which isn’t always the case with this type of service.

The only caveat? The company will automatically bill you at the normal price when the discounted subscription runs out. Be sure to cancel before that if you aren’t vibing with the platform.

Follow @EngadgetDeals on X for the latest tech deals and buying advice.

This article originally appeared on Engadget at https://www.engadget.com/deals/protonvpn-subscriptions-are-up-to-66-percent-off-right-now-191045586.html?src=rss

I Love This Speaker That Turns Any Song Into a Karaoke Track, and It’s $50 Off Right Now

We may earn a commission from links on this page. Deal pricing and availability subject to change after time of publication.

Finally, AI is being put to good use, without taking anyone’s job (I think?). The Soundcore Rave 3S uses a unique AI feature to turn any regular song you play over Bluetooth into a karaoke track by digitally removing the vocals. I praised it highly in my review, and I’ve had a lot of fun making it a centerpiece at my parties this year.

Right now, the speaker has gotten its first significant discount—you can pick one up for $299.99, $50 of the usual price.

If you’ve ever experienced that feeling of disappointment when looking for a specific karaoke song on YouTube and coming up empty-handed, this speaker is for you. It has happened to me plenty, mainly because my taste in music is fairly niche. If the songs you’re searching for are easily found on YouTube and you’d prefer a smaller, more portable speaker, I recommend getting the Ikarao Shell S1 portable speaker instead.

Otherwise, the Soundcore Rave 3S is pretty sweet. The AI vocal remover works well. You can trigger it by pressing a button on one of the two Bluetooth microphones or through the companion app. You can also choose the degree to which you want the vocals to be removed.

The companion app is intuitive and useful and includes fun Vocal Enhancer features that change the sound of the mic (with options for neutral, warm, presence, and bright). The EQ options are also very detailed and change in real time. The battery will give you up to 12 hours of juice, so you can take the party outdoors (it’s splashproof, with an IPX4 rating).

Warner Bros. Discovery Just Reversed That Terrible ‘HBO Max’ Branding Decision

I love movies and TV shows of all kinds, but I have to admit I have a soft spot for HBO. The channel is responsible for some of my favorite TV experiences of all time: Game of Thrones, Barry, Silicon Valley, and now The Last of Us all come to mind. I’ll watch anything on any platform that looks intriguing enough, but when HBO is attached, I’m even more likely to watch.

That’s why it made absolutely no sense at all for Warner Bros. Discovery to remove the name of one of the most respected brands in show business, truncating HBO Max to just Max. HBO hasn’t gone away: The channel is still a massive part of the streaming service, offering its past catalogue as well as producing new shows as it always has. But there are also “Max Originals,” which are not HBO shows but available on the platform formerly known as HBO Max. Not confusing at all.

It’s been two years since the name change, and it seems like the decision has been a success. Oh, sorry, did I say success? I don’t have access to Warner Bros. Discovery’s internal data, but I imagine it hasn’t been a success, seeing as the company is now bringing back HBO Max from the dead.

Somehow, HBO Max returned

It’s true: Warner Bros. Discovery is returning HBO to its platforms’ name this summer. No longer will you need to tell your friends to open Max to watch the latest episode of Hacks or The White Lotus (were there many of us still calling this anything but HBO anyway?)

The company won’t necessarily admit the name change was a bad idea, but it does offer a corporate explanation. JB Perrette, the company’s President and CEO of Streaming, said that “This evolution has also been influenced by changing consumer needs…No consumer today is saying they want more content, but most consumers are saying they want better content.”

Perrette also said they’ve been “iterating” on the idea for the past two years, which sounds to me like the company had immediate regrets after ditching the HBO brand. Perrette references Warner Bros. Discovery CEO David Zaslav’s, who reportedly championed the idea that the company should “go back to including HBO in the brand, because nothing stands for distinction and quality more than HBO.” Well, duh.

At least the official HBO X account is having a fun time with the news:


This Tweet is currently unavailable. It might be loading or has been removed.

OpenAI promises greater transparency on model hallucinations and harmful content

OpenAI has launched a new web page called the safety evaluations hub to publicly share information related to things like the hallucination rates of its models. The hub will also highlight if a model produces harmful content, how well it behaves as instructed and attempted jailbreaks. 

The tech company claims this new page will provide additional transparency on OpenAI, a company that, for context, has faced multiple lawsuits alleging it illegally used copyrighted material to train its AI models. Oh, yeah, and it’s worth mentioning that The New York Times claims the tech company accidentally deleted evidence in the newspaper’s plagiarism case against it.

The safety evaluations hub is meant to expand on OpenAI’s system cards. They only outline a development’s safety measures at launch, whereas the hub should provide ongoing updates. 

“As the science of AI evaluation evolves, we aim to share our progress on developing more scalable ways to measure model capability and safety,” OpenAI states in its announcement. “By sharing a subset of our safety evaluation results here, we hope this will not only make it easier to understand the safety performance of OpenAI systems over time, but also support community efforts⁠ to increase transparency across the field.” OpenAI adds that its working to have more proactive communication in this area throughout the company. 

Introducing the Safety Evaluations Hub—a resource to explore safety results for our models.

While system cards share safety metrics at launch, the Hub will be updated periodically as part of our efforts to communicate proactively about safety.https://t.co/c8NgmXlC2Y

— OpenAI (@OpenAI) May 14, 2025

Interested parties can look at each of the hub’s sections and see information on relevant models, such as GPT-4.1 through 4.5. OpenAI notes that the information provided in this hub is only a “snapshot” and that interested parties should look at its system cards. assessments and other releases for further details. 

One of the big buts to the entire safety evaluation hub is that OpenAI is the entity doing these tests and choosing what information to share publicly. As a result, there isn’t any way to guarantee that the company will share all its issues or concerns with the public.

This article originally appeared on Engadget at https://www.engadget.com/ai/openai-promises-greater-transparency-on-model-hallucinations-and-harmful-content-184545691.html?src=rss

Palantir CEO Slams Europe’s AI Ambitions

Palantir CEO Alex Karp criticized Europe’s AI adoption while praising Saudi Arabia’s engineering talent at Tuesday’s Saudi-US Investment Forum in Riyadh. “It’s like people have given up,” Karp said of Europe, while commending Saudi engineers for their “meritocracy and patriotism” and “deep tradition in engineering excellence.”


Read more of this story at Slashdot.

Don’t Fall for These Eight Deceptive Marketing Terms

Corporations make a lot of stuff they want to sell you, and invest a lot of money, effort, and time into convincing you to not only to buy a product, but to buy a specific version of it. While there’s nothing wrong with pointing out the advantages of one product over another, sometimes marketers use specifically vague or deceptive phrases in order to convince you to choose the one they’re offering up.

The problem with these phrases is that they exist in a kind of gray area—they’re not outright lies, but they’re also not particularly transparent. When you see one of these eight terms printed on a product’s label, it’s a good reminder to engage in some critical thinking about what it’s actually telling you.

FDA-approved

Seeing the phrase “FDA Approved” on a medicine’s product label probably gives you a certain sense of confidence. After all, the Food and Drug Administration doesn’t just approve everything! Getting that FDA stamp of approval must mean it’s safe and high-quality, right?

But that’s not actually what “FDA Approved” means at all. The phrase specifically means “the drug is determined to provide benefits that outweigh its known and potential risks for the intended population.” It has absolutely nothing to do with quality, and it doesn’t even mean it’s low-risk—just that the benefits outweigh the risks. That’s useful information, but in marketing, the phrase is used as an indicator that you’re getting a superior product that you can trust to be safe, when all it really means is that it works, and the downsides are (probably) worth it.

Genuine leather

The word “genuine” is doing a lot of work here. You probably think it must be some kind of industry term, with all kinds of meaningful grading and quality testing behind it. But in the words of Lifehacker editor Beth Skwarecki, “‘genuine leather’ just means it’s…leather.” It literally just means the thing you’re holding in your hand is, in fact, made of leather. It says nothing about the quality of that leather—good or bad.

Patented

If you watch Shark Tank, you probably have a high opinion of the patent process, and assume that if a product is patented—and shouts that information everywhere in its advertising—it must be innovative and unique. You’d better buy this version, is the implication, because no one else can replicate its patented magic.

Eh, not necessarily. Sure, some patents do, in fact, protect innovative ideas. But patents can be issued for a lot of reasons—sometimes minor technical improvements, or new ways of combining ingredients or components. The U.S. issues hundreds of thousands of patents every year, and not all of them are meaningful in the sense of describing amazing breakthroughs or even unique applications. But the term conveys a certain gravitas to a product’s other claims around effectiveness, utility, and value, so marketers use it any time they can.

Maximum strength

The “it goes to 11” of marketing copy. This phrase always looks convincing— it’s usually written in all caps at the top of the label to let you know that the product is not messing around, and that it is the most product you can buy.

The problem is, what does “maximum” mean? Maximum compared to what? If it means compared to the other versions of the product, that doesn’t necessarily mean competing brands won’t be stronger. If it refers to some sort of legal or physical limitation on how powerful a drug or other product can be, then that same restriction applies to competing products as well. It’s a relative term that doesn’t mean anything unless you put in some research to understand what the reference points actually are.

All-natural

Every now and then, someone falls for the classic “dihydrogen monoxide (DHMO)” prank, agreeing that a dangerous chemical like DHMO should be banned after hearing about all the potential hazards it causes, like death if accidentally inhaled. DHMO is, of course, water (H2O)—dangerous under the right conditions, but also necessary for life. The point being, many chemicals are all-natural, technically speaking, and there’s no clear guidance from the FDA on what the term means. Most products are processed to such a degree that it’s impossible to determine what “natural” even could mean, so labeling something as “all-natural” doesn’t mean much.

Doctor-approved

When shopping for health-related products of any kind, it can be persuasive to see one is “doctor approved” in some way. You might think this means that a major medical association has come together to recommend that product, or at least conducted some sort of study and determined that this product did the job best. It probably doesn’t mean that, though. While it often means that at least one doctor tested or reviewed the product, that doctor may have been on the company payroll, and either way the phrase doesn’t mean a more formal or official approval process of any kind took place.

Clinically proven

Similarly, the phrase “clinically proven” can be deceptive because the clinical studies being referenced are often paid for by the manufacturer itself. While that doesn’t necessarily mean these studies are fake, it does call into question how objective the study and its designers were, and whether any independent bodies corroborate its findings.

Worse, sometimes the studies being cited for the “clinically proven” claim don’t actually prove that the product works. Sometimes fine print hidden somewhere on a label or website will clarify this, but not always. Even if a study was reasonably well-conducted and independently organized, you need more than one study to have confidence that an active ingredient or specific formulation of something actually works as advertised.

Recyclable

If you’re worried about the future of the planet, you might be seeking out products that lower your carbon footprint. Seeing that a product’s packaging is recyclable makes you feel better about using it, because you know that when you toss it away you’re not adding to a landfill.

Except, often, you are. Companies sometimes use materials in their packaging that are technically recyclable, but practically not recyclable. In other words, the specific plastic or other material can be recycled, but aspects of the packaging itself—shape, size, and the stuff it contains—mean that it will be separated out and tossed into a landfill regardless. And sometimes the specific material used isn’t recycled universally—for example, HDPE plastic can be recycled, but not all recycling facilities accept it.

Five Hidden Windows App Switcher Secrets

Alt-Tab was among the first Windows keyboard shortcuts I learned when I first used a PC over two decades ago, right after Ctrl-C (copy) and Ctrl-V (paste). Alt-Tab opens the app switcher, which lets you quickly bring a different app to the foreground. You can use it to quickly swap between two open apps, or to cycle between all of your open apps. But what you might not know is that the app switcher can so a lot more than that. I’m here to walk you through the best tricks hidden in the commonly used Windows feature.

Cycle through open apps

Hold down the Alt key and keep pressing Tab to open the app switcher and cycle through all your open apps. Once you release the shortcut, the selected app will come to the foreground. You can also cycle through this list in reverse order by holding Alt-Shift and pressing the Tab key repeatedly.

Stop the app switcher from auto-hiding

The app switcher’s temporary nature is a bit annoying sometimes. The moment you release Alt-Tab, the app switcher disappears. You can get around this by pressing Alt-Ctrl-Tab. Now, you’re free to release the keyboard shortcut and keep the app switcher happily floating above all your open apps. Use arrow keys to cycle through your list of open apps, or use the mouse to directly pick the app you need. To dismiss the app switcher, simply click outside of it.

If this shortcut is too difficult to press, try using the Alt key to the right of the spacebar along with Tab. So, press Right Alt-Tab and it’ll also stop the app switcher from automatically hiding.

Use the app switcher to quit apps

Once you have the app switcher open, you can use it to quit apps, too. Press Alt-Ctrl-Tab to open the app switcher, then move the cursor to any of the thumbnails on screen. You’ll see a small X button in the top-right corner of each thumbnail. Click the X to quit that app. Alternatively, you can use the arrow keys to select any app and press the Delete key to quit the selected app.

Enjoy a full-screen view

The full-screen Task View in Windows 11, against a gradient background.

Credit: Pranay Parab

If a small floating window is not sufficient for you, you can make the app switcher full-screen as well. Press Windows-Tab to open the Task View, which shows a list of all open apps in full-screen. You can use this to switch to any app or to quit apps. Press Esc to leave this view.

Try a third-party customization tool

While Windows’ defaults work well, you can also consider using third-party apps to customize the app switcher’s theme. Both Winaero Tweaker and AltPlusTab let you change the look and feel of the app switcher. You can use these apps to change the switcher’s opacity, fonts, appearance, and more.