We might be skeptical of some Nvidia’s claims, like whether a $549 RTX 5070 will truly deliver the performance of a $1599 RTX 4090. But it’s almost impossible not to be impressed by the RTX 5090 Founder’s Edition, where Nvidia fit 575 watts of graphics power, including 21,760 CUDA cores and 32GB of GDDR7 memory, into a video card just two slots wide.
It almost has to be seen to be believed, and we sent my colleague Antonio around the CES show floor in Las Vegas in what was initially a fruitless search. No PC manufacturer seemed to have an interactive game demo running on a 5090, much less the two-slot card.
But on Wednesday, we finally spotted the real deal at Nvidia’s offsite event — and then some. It’s heavy, and the uniquely desirable $2,000 card may wind up being rare, but it’s here, and it works.
Below, find our pictures of the relatively compact 5090; its incredibly compact PCB with the Blackwell chip on top; a game demo running on the 5090; a picture of the 5090, 5080, and 5070 Founder’s Editions side by side; and some examples of just how bulky every other partner’s cards can be compared to Nvidia’s own.
Google is testing an experimental AI feature that creates a personalized podcast using your Search and Discover feed history. The Daily Listen feature in Google’s Search Labs is rolling out to Android and iOS users in the US according to 9to5Google, and works similarly to the Audio Overviews feature for Google’s NotebookLM project.
Daily Listen references search data and Discover feed interactions to assess which news articles would be of most interest to the user, and then summarizes those stories and topics into a roughly five-minute audio overview. It provides a text transcript, alongside audio scrubber controls that allow users to play, pause, mute, rewind, or skip to the next story.
Image: Google (Via 9to5Google)
Here’s a few screengrabs of the Daily Listen UI, showing where the feature is located and how it presents related topics.
Search Labs users who opt-in to try this experiment will find Daily Listen in the personalized widget carousel underneath the Search bar at the top of the Google app. A “Related stories” tab will appear at the bottom of the audio player that corresponds to content within the overview that allows users to like or disapprove of each story with a thumbs up/down, and explore new topics.
There’s currently no word on when/if this feature will be rolled out to the wider public. Google’s AI Search Overviews were tested in a similar way via Search Labs before launching to a general audience last year, so it’s possible that Daily Listen may one day become a persistent feature — hopefully one that’s easier to opt out of than AI Overviews.
Apple is refuting rumors that it ever let advertisers target users based on Siri recordings in a statement published Wednesday evening describing how Siri works and what it does with data.
The section specifically responding to the rumors reads:
Apple has never used Siri data to build marketing profiles, never made it available for advertising, and never sold it to anyone for any purpose. We are constantly developing technologies to make Siri even more private, and will continue to do so.
The conspiracy theory the company is responding to resurfaced last week after Apple agreed to pay $95 million to settle a lawsuit over users whose conversations were captured by its Siri voice assistant and potentially overheard by human employees.
While Apple’s settlement addresses an issue that The Guardian reported in 2019. The report showed human contractors tasked with reviewing anonymized recordings and grading whether the trigger was activated intentionally, would sometimes receive recordings of people discussing sensitive information. But it doesn’t include any reference to selling data for marketing purposes.
After The Guardian’s report in 2019, Apple apologized and changed its policy, making the default setting not to retain audio recordings from Siri interactions and saying that for users who opt-in to sharing recordings, those recordings would not be shared with third-party contractors.
However, reports about the settlement noted that in earlier filings like this one from 2021, some of the plaintiffs claimed that after they mentioned brand names like “Olive Garden,” “Easton bats,” “Pit Viper sunglasses,” and “Air Jordans,” they were served ads for corresponding products, which they attributed to Siri data.
Apple’s statement tonight says it “does not retain audio recordings of Siri interactions unless users explicitly opt in to help improve Siri, and even then, the recordings are used solely for that purpose. Users can easily opt-out at any time.”
Facebook responded to similar theories in 2014 and 2016 before Mark Zuckerberg addressed it directly, saying “no” to the question while being grilled by Congress over the Cambridge Analytica scandal in 2018.
So, if Apple (and Facebook, Google, etc.) is telling the truth, then why would you see an ad later for something you only talked about?
There are other explanations, and attempts to check the rumors out include an investigation in 2018 that didn’t find evidence of microphone spying but did discover that some apps secretly recorded on-screen user activity that they shipped to third parties.
Ad targeting networks also track data from people logged onto the same network or who have spent time in the same locations, so even if one person didn’t type in that search term, maybe someone else did. They can buy data from brokers who collect reams of detailed location tracking and other info from the apps on your phone, and both Google and Facebook pull in data from other companies to build out profiles based on your purchasing habits and other information.
This robot vacuum from Dreame has a robotic arm that can pick up items and use tools from its toolbox (pictured in the background) to clean more areas of your home. | Image: The Verge
I’ll see your arm and raise you an arm and two legs. It was the battle of the bots on the CES show floor as robot vacuum manufacturers Dreame and Roborock each added limbs to their rival robovacs.
Dreame launched its X50 Ultra at the show earlier this week, debuting the first robovac that can use its legs to navigate steps and room transitions up to 6cm high. But elsewhere at the show, competitor Roborock was showing off its latest flagship, the Saros Z70, which has an arm that can pick up items like socks.
Not to be outdone, Dreame then showed off a soon-to-be-released model at its CES booth, combining those two step-climbing legs with a robotic arm of its own.
Dreame’s model has a chunkier-looking arm than the Roborock’s, and it says it can pick up items up to 500 grams, whereas Roborock’s can only tackle items up to 300 grams. Dreame says its arm can pick up sneakers as large as men’s size 42 (a size 9 in the US) and take them to a designated spot in your home. The concept could apply to small toys and other items, and you’ll be able to designate specific areas for the robot to take certain items, such as toys to the playroom and shoes to the front door.
However, I didn’t see the robot picking up a sneaker — or anything at all — apparently, the infamous CES show floor Wi-Fi couldn't hack it. Instead, they showed the robust-looking arm moving up, down, and around while the robot lifted itself up on its two small legs. It looked like a tiny horse.
Photo by Jennifer Pattison Tuohy / The Verge
The robot vacuum lifting itself up on its legs while its arm stretches out.
Another interesting invention is a separate toolbox with various brushes that Dreame says the arm can connect to, enabling it to reach into corners and tight spaces where the bot itself cannot go and sweep out the dirt and dust. They also had a new base station that dispenses multiple mopping pads. This allows the robot to choose different pads for different jobs around your home — one for the kitchen another for the bathroom — to help avoid cross-contamination around your home.
Dreame’s Longdong Chen told The Verge that the step-climbing, tool-using, arm-touting bot should be available later this year. A price hasn’t been announced, but it’s a safe bet that it’ll cost an arm and a leg.
Microsoft is rolling back a model upgrade to its AI-powered Bing Image Creator, reports TechCrunch. The rollback came after weeks of complaints by users that the tool just didn’t work as well after Microsoft “upgraded” to a new version of the DALL-E 3 model on December 18th.
Microsoft declined to comment on its decision to roll things back or offer specifics on what may be causing the gap between user’s expectations and its output.
Today, Microsoft’s head of search, Jordi Ribas, tweeted that they could reproduce “some of the issues reported” and are reverting to an older version of the DALL-E model for now, although it could take a few weeks until it’s complete.
Since the launch of Bing Image Creator last spring, users have generated billions of images with text prompts. I'm pleased to share our latest updates to enhance your creative experience. Starting today, we’re rolling out the latest DALL-E 3 model PR16, which will create images… pic.twitter.com/3p9HsYMlr6
Thanks again for the feedback and patience. We've been able to repro some of the issues reported and plan to revert to PR13 until we can fix them. All Pro users and about 25% of the requests using boosts are now on PR13. The deployment process is very slow unfortunately. It…
As soon as Ribas posted about the change in December, there were complaints that Bing Image Creator was producing less-detailed results or images that didn’t accurately reflect their prompts. In his initial replies, Ribas said the model’s output quality “should be a bit better on average” than before.
It was the same story in posts and comments on Reddit and OpenAI’s community forums. On OpenAI’s forums, a person complained about the model’s handling of fabric on an anime-style character’s dress. The person who posted the below images says the one on the left is “perfect quality” while the one on the right is “over-lit.”
All of these things are subjective, and I can’t claim to think any of them look better than any others. If anything, it feels like an indication that Microsoft doesn’t just have to deal with complaints about bugs or people upset about feature changes — now it has to deal with AI art critics comparing the machine’s output to what they imagine it should create. Perhaps they should ask the artists whose work the generators were trained on for tips about managing a client’s expectations?
The Surface Laptop 7 with Qualcomm’s Snapdragon X Elite processor. | Photo by Chris Welch / The Verge
Microsoft is preparing for a “major” Surface announcement later this month. The software giant has started teasing “a major announcement from Surface for Business” this week in a LinkedIn post spotted by Windows Central. The announcement will be made during Microsoft’s AI tour in New York City on January 30th.
Microsoft is rumored to be launching Intel-powered variants of its Surface Laptop 7 and Surface Pro 11 devices soon, and a prototype of a Surface Laptop 7 with Intel’s Lunar Lake chips appeared on a Chinese second-hand marketplace in October. It’s likely that Microsoft will announce Lunar Lake versions of the Surface Laptop 7 and Surface Pro 11 that are designed for businesses.
The Surface Pro 10 and Surface Laptop 6 both shipped with Intel’s Meteor Lake processors earlier in 2024, before Microsoft went on to launch refreshed Surface Copilot Plus devices with new designs and Qualcomm chips.
The Surface Laptop Studio is also due for an upgrade from the current model’s 13th Gen Intel chips, but that’s less of a business-focused device so it’s unlikely to be part of this event. There are also rumors of an 11-inch Surface Go / Surface Laptop Go hybrid device powered by a Snapdragon X Plus processor, but again that’s unlikely to be targeted to businesses.
We’ll be following Microsoft’s AI tour live later this month, so stay tuned for details on the Surface “major announcement.”
Tonal on Wednesday launched the Tonal 2, an upgraded version of its smart in-home strength trainer.
Launched in 2018, the Tonal is a wall-mounted strength training machine with electromagnetically adjustable resistance and a touchscreen display. It can monitor lifting form, offer virtual coaching,predict and set the optimal weight for each exercise, and responsively adjust as you lift for a personalized workout. The company also offers on-demand workouts.
Along with sporting a new all-black design and chrome accents, the Tonal 2 comes with a host of meaningful internal improvements, including a new adaptive weight system with up to 250 pounds of digital resistance, up from the 200 pounds in the original trainer.
The latest version of the trainer now offers drop sets via a software update also available on the Tonal 1, with the option toautomatically reduce weight during a set as muscle fatigue builds up. The Tonal 2 also comes with a new “Aero HIIT” mode, which lets you combine cardio and strength training into a single session.
Tonal also improved its optional Smart View feature, which uses cameras to watch your form and give real-time coaching cues. On the Tonal 1, Smart View required you to use your phone camera, and used the on-device camera for a second angle in some exercises; on the Tonal 2, it uses the built-in camera, with an option to use your smartphone for a second angle. The company says the camera also can now use AI to “analyze 500 data points per second.”
The Tonal 2 is available now, starting at $4,295, not including delivery and installation, plus a $495 “smart accessories bundle” that the company’s website says is “required to unlock the best Tonal experience.” It also requires a $60/month subscription, with a one-year commitment to start. Tonal noted on Instagram it will be introducing a trade-in program for existing Tonal 1 owners looking to upgrade, but didn’t give specifics. Tonal told The Verge the Tonal 2 can be mounted on the original Tonal’s existing wall mount for easier installation.
Awesome Games Done Quick, the annual charity speedrunning marathon is in full swing, already amassing over $500,000 for the Prevent Cancer Foundation. With three more days to go, there’s a great many runs of popular games coming up. But relying on the schedule or your own preferences to determine when to tune in is a surefire way to miss the best AGDQ has to offer.
Take today, for example. I’ve never played Batman Forever: The Arcade Game, but after watching the speedrun, I might have to. It’s a Batman-flavored beat-’em-up that bears precious little resemblance to the movie it’s based on. The runner, LRock617, chose to play as Robin and within moments of starting the game, I was hooked. The graphics were reminiscent of the early Mortal Kombat games, appropriate since it was published by Acclaim Entertainment. But there was a surreality to them that — when combined with Robin’s overpowered abilities (like one that has him essentially go Super Saiyan), the runner’s charisma in explaining what’s happening, and the audience getting in on the fun with funny donation messages — gave the run an absurd goofiness that was a sorely needed shot in the arm at 9AM on a Wednesday morning....
Nintendo isn’t officially at CES, but it might have stolen the show anyway: accessories-maker Genki brought a 3D-printed mockup of what it says is Nintendo’s Switch successor, and we got to hold it and take some high-resolution pictures.
Based on the mockup of what we’ll call the Switch 2, Nintendo’s next console appears to be wider than the original, with slightly larger Joy-Con controllers that seem like they’ll be more comfortable to hold. Compared to a Steam Deck OLED, the Switch 2 mockup still feels noticeably smaller, in part because the Joy-Cons are not as pronounced and ergonomic as the Deck’s grips. But the mockup still feels closer in size to Valve’s handheld than the original Switch.
As previous rumors have indicated, the Switch 2’s Joy-Cons will attach to the console via magnets instead of a sliding rail, Genki cofounder and CEO Eddie Tsai tells The Verge. To detach a Joy-Con from the Switch 2, you press a big button at the top of the backside of each controller, Tsai says, and that button apparently pushes out a pin that nudges against the chassis of the console, disconnecting the magnets.
You can remove the Joy-Cons with brute force if you really wanted to, according to Tsai, but he says, overall, they feel secure for regular use and that the big release button detaches the Joy-Cons with ease. Tsai declined to share where he’d learned details of the new console.
Tsai also tells The Verge that housed within the mounting channel of the Joy-Cons is an optical sensor, and by using another attachment the new Joy-Cons may offer mouse-like functionality. It sounds a bit like what Lenovo does with its Legion Go handheld.
Nintendo has promised that it will announce the Switch’s successor before April 2025, and as that deadline creeps closer, there have been waves of leaks and rumors about the new hardware. In December, YouTube channel SwitchUp posted a video showing a 3D-printed, non-functional Switch 2 mockup provided by a Chinese case manufacturer. That mockup basically looked like the current Switch but bigger, and it revealed a few other potential changes like a new USB-C port on the top of the device and a mysterious new square button under the Home button on the right Joy-Con.
Days later, accessories-maker Dbrand announced its “Killswitch 2” case, and CEO Adam Ijaz told The Verge that it was designed based on “actual dimensions” based on a “3D scan of the real hardware.” Ijaz also said that it was his “understanding” that the console’s Joy-Cons are magnetically attached. And Dbrand’s imagery showed that the new square button had a “C” printed on it, though Ijaz didn’t know what it was. Days after that, our colleague Sean Hollister spoke with a Redditor who shared apparent photos of the Switch 2’s dock and the inside of what appears to be a Switch 2 Joy-Con.
With that April 2025 deadline inching closer every day, it seems like it won’t be long until Nintendo officially, finally reveals the Switch 2. But until that happens, at least we have these pictures to look at.
Photography by Antonio G. Di Benedetto / The Verge
The OnePlus 13 is an excellent alternative to phones from Samsung and Google. | Photo: Allison Johnson / The Verge
Samsung’s Galaxy S25 might be just around the corner, but if you’re looking for a solid Android alternative, OnePlus is already offering the new OnePlus 13 with a free storage upgrade, dropping the price of the 512GB model with 16GB of RAM to $899 ($100 off). It’s also throwing in a OnePlus Watch 2R (a $229.99 value) or a free pair of the OnePlus Buds Pro 3 (normally $179.99) with each purchase. Amazon and Best Buy, meanwhile, are bundling it with a $100 gift card, though only the latter is doing so with the free storage upgrade.
The OnePlus 13 is a stunner of a phone, one that impressed us in a number of ways. It boasts a 6.82-inch 120Hz display and runs on Qualcomm’s new Snapdragon 8 Elite, making it even faster than its predecessor. Its massive 6,000mAh battery also stands out amongst a sea of 5,000mAh Android phones, allowing you to use it for multiple days on a single charge. The triple-array camera — which consists of a 50-megapixel main shooter, an ultrawide, and a telephoto — is also on par with the best, letting you produce sharp photos in dim lighting.
Even more notable, though, is that the OnePlus 13 finally offers all the flagship features you’d expect from a phone at this price. That includes support from all three major US wireless carriers and an IP68 rating for dust and water resistance. OnePlus also promises four years of OS updates and six years of security support, rendering it a better investment than the company’s previous smartphones. Sure, the OnePlus ecosystem still isn’t as rich as that of Samsung or Google, but if that doesn’t bother you, it’s otherwise an excellent phone.
The 2024 Asus Zenbook Duo ison sale for an all-time low of $1,499.99 ($200 off)at Best Buy with an Intel Core Ultra 9 H-series processor, 1TB of storage, and 32GB of memory. The Duo is a great option if you need a lot of screen real estate, as the laptop offers a pair of 14-inch 120Hz OLED screens, one of which is hidden beneath a trackpad and a removable full-size keyboard. The unique laptop also features a stylus and a solid port selection, including a pair of USB-C ports, an HDMI port, a USB-A port, and a 3.5mm audio jack. Read our comparison with the Lenovo Yoga Book 9i.
Costco members can buy thestainless steel Ember Mug 2 for $69.99 ($30 off) through January 10th (it’s also available to non-members who pay a $5 surcharge). The 14-ounce, app-controlled smart mug can keep your coffee or tea heated to your preferred temperature for up to 90 minutes using the built-in battery — or indefinitely when docked on the included charging coaster.
The second-gen Bose Soundlink Flex is on sale at Amazon starting at $111.34 (about $38 off). The original model was one of our favorite Bluetooth speakers thanks to its travel-friendly size and detailed sound, which the latest model expands upon with support for AAC and aptX Bluetooth audio codecs. It also retains anIP67 rating for dust and water resistance — meaning it makes for a great pool companion — along with up to 12 hours of battery life.
As I write this, there are a lot of social network users who are wondering if they should look for a new home. Over at X, Elon Musk has essentially become part of the incoming Trump administration, while various changes have made the formerly popular social network a dark and forbidding forest for many of its former inhabitants.
Meanwhile, Meta’s announcement that it was abandoning third-party fact-checkers and moving its trust and safety teams from California to Texas is making some Facebook and Instagram members nervous. So nervous, in fact, that while we previously included Meta’s Threads social network in this article as a possible alternative to X, we’ve pulled it — at least for now.
So, if you’re no longer feeling safe at your current social network, where do you go?
We’ve been looking into the various possibilities and have put together what is admittedly an incomplete list of some of the current alternatives to X, Facebook, and Instagram that you may want to check out if you’re thinking about leaving your current hangout.
Probably nothing will become the combined news / gossip / conversation / spam source that Twitter once was, and it may be difficult to leave the kind...
That’s a sharp turn from his statements in 2023 laying out the goal of a “less angry place for conversations” that wouldn’t do anything to encourage politics or hard news. However, under Meta’s new approach to moderation — and new rules about what users can say on its platforms — that goal is going out the window just as the Trump administration prepares to take over.
Until now, users have had to opt-in to seeing recommendations of content deemed political, but the change rolling out this week in the US and to the rest of the world next week will turn on the recommendations and a content control setting available with options for less, standard (the default setting), and more.
In a series of Threads posts, Mosseri reiterated, “I’ve maintained very publicly and for a long time that it not our place to show people political content from accounts they don’t follow,” and that “it’s proven impractical to draw a red line around what is and is not political content.”
In a video on Instagram, he said that the push for political content — particularly from users on Threads — is “by the way, very different from the feedback we were getting only a few years ago about people feeling that they were overly exposed to political content on our platforms.” Of course, according to the Wall Street Journal, that was before Mark Zuckerberg experienced the effects of filters cutting down the reach of his post about recovering from a torn ACL and before Meta’s new and friendlier-to-Trump policy chief took over.
Two Apex 300 units with several expansion batteries. | Image: Bluetti
Bluetti just took the wraps off its EnergyPro 6K and Apex 300 energy storage systems. The EnergyPro 6K is a fixed whole-home backup solution, while the modular Apex 300 is designed to provide portable power when at home, at an off-grid cabin, at a job site, or during an RV trip. Both systems can scale in capacity and power to support your individual needs.
The launch is all part of Bluetti’s rebranding around three series of power solutions: EnergyPro products that require professional installation, Apex systems for advanced energy demands, and the smaller Elite lineup of portable power stations and solar generators for more casual use.
The Apex 300 features 3840W of max AC output and 3072Wh of LFP battery storage. It can be combined with the company’s B300K, B300, and B300S batteries, including future expansion batteries yet to be announced. The Apex 300 has an idle power draw of 20W, according to Bluetti, which is pretty efficient for such a large inverter that can simultaneously support both 120V and 240V loads.
Image: Bluetti
The Bluetti Apex 300.
Multiple Apex 300 units can be linked in parallel to increase the maximum output to 11.52kW — enough to power just about any home device or EV. Storage capacity can be expanded to 58kWh with three units and 18 battery packs to keep an entire home running for a few days in the event of a blackout, or much longer if you’re only powering critical devices like the fridge, HVAC systems, and water pumps. It supports up to 30,720W of solar input and is expected to be available in April in the US.
Image: Bluetti
Two worry-free owners of a pair of EnergyPro 6K units connected to an AT1 Smart Distribution Box.
Bluetti bills the EnergyPro 6K as “a reliable, affordable and cost-effective home energy solution ideal for small to medium-sized homes.” It can be configured with 5.8kW to 29kW of power output and between 7.68kWh and 38.4kWh of LFP battery storage. It’s designed to integrate seamlessly into existing rooftop solar systems. It can be paired with a Bluetti AT1 Smart Distribution Box to provide whole-home backup to critical circuits and automatic cutover to a standby generator when needed. Bluetti is also teasing an EnergyPro 13K system, offering 13.2kW to 39.6kW of output and between 9.6kWh and 57.6kWh of storage capacity. The company says that the EnergyPro 6K is expected to be released in Q2 of 2025.
No prices were given as part of today’s announcement, but you can expect both systems to start in the low thousands, or even tens of thousands if you’re looking at a fully specced installation.
One of the biggest announcements in Nvidia CEO Jensen Huang’s CES keynote was the small “Project Digits” AI supercomputer, and if you want to get an idea of just how tiny the $3,000 machine is in real life, we snapped a couple photos of the device under glass today at the show.
Take a look: we’ve captured the front of a Digits computer in the photo at the top of this post, and below this paragraph is a photo of the back featuring the computer’s ports. I really like the textured design.
Photo by Sean Hollister / The Verge
The Digits computers will come with Nvidia’s GB10 Grace Blackwell Superchip, which offers “a petaflop of AI computing performance for prototyping, fine-tuning and running large AI models,” according to Nvidia’s press release. It also includes a GPU built with Nvidia’s Blackwell architecture, 128GB of unified memory, and up to 4TB of NVMe SSD storage.
This isn’t a computer for most people; Nvidia says that Project Digits is intended to provide “AI researchers, data scientists and students worldwide with access to the power of the NVIDIA Grace Blackwell platform.” It definitely isn’t something I will ever buy.
But it is impressively tiny given its capabilities — small computers have been on a tear lately!
Meta is ending its third-party fact checks and making sweeping changes to its content moderation policies.
Meta is making sweeping changes to its content moderation policies, including abandoning third-party fact checks in favor of crowd-sourced “Community Notes” and loosening restrictions on topics like immigration and gender identity. Under the updated Hateful Conduct policy, for example, calling gay and trans people “mentally ill” is now allowed, while an explicit ban on referring to women as “household objects” has been removed.
New policy lead Joel Kaplan said that in pursuit of “More Speech and Fewer Mistakes,” Meta will focus more on preventing over-enforcement of its content policies and less on mediating potentially harmful but technically legal discussions on its platform.
It comes just two weeks before President-elect Donald Trump is set to return to the White House, and CEO Mark Zuckerberg’s announcement appealed to many of the incoming administration’s talking points. Zuckerberg has promised to move US content review from California to Texas, where he says there’s “less concern about the bias of our teams,” and said Meta would work with Trump to “push back on governments around the world that are going after American companies and pushing to censor more.”
Nvidia CEO Jensen Huang made a bold claim onstage at CES earlier this week when he was introducing the next-generation RTX 50-series GPUs. “The RTX 5070, 4090 performance at $549,” said Huang. It’s a claim that’s been echoed on YouTube, TikTok, and social media networks and has generated a debate over the RTX 50 series and DLSS 4’s Multi Frame Generation.
So, can a $549 RTX 5070 really deliver the same level of performance as a $1,599 RTX 4090? The answer is yes and no, and it all comes down to a “fake frames” argument about DLSS Frame Generation that might not even be a big problem for a lot of PC gamers.
Nvidia’s big RTX 5070 claim is all based on its latest generation of DLSS. “Impossible without artificial intelligence,” admits Huang after promising that the RTX 5070 can deliver RTX 4090 levels of performance. DLSS 4 has a new Multi Frame Generation technique that can generate up to three additional frames per every traditionally rendered frame.
Some PC gamers have long argued that this technique, which Nvidia introduced originally with DLSS 3, is simply “fake frames” and not reflective of the true performance of GPUs that have been using rasterization for decades.
Seriously: would you wear something like this on your face if it could really read your mind? | Image: Omi
Nik Shevchenko closes his eyes and starts to focus intently. He’s spent the last half hour or so telling me about his new product, an $89 wearable called Omi that can listen to, summarize, and get information out of your conversations. Now he wants to show me the future. So his eyes are closed, and he’s focusing all his attention on the round white puck stuck to his left temple with medical tape. (Did I mention he’s had this thing on his face the whole time? It’s very distracting.)
“Hey, what do you think about The Verge, like as a news media website?” Shevchenko asks, to no one in particular. Then he waits. Fifteen or so seconds later, a notification pops up on his phone, with some AI-generated information about how reputable and terrific a news source The Verge is. Shevchenko is thrilled, and maybe a little relieved. The device read his brain waves to understand he was talking to it, and not to me, and answered his question without any prompting or switching.
So far, that’s all the brain-computer-interface stuff Omi can do. And it seems pretty fragile. “It just understands one channel,” he says, “it’s one electrode.” What he’s trying to build is a device that understands when you’re talking to it and when you’re not. And then eventually understands and saves your thoughts, which Shevchenko both waves off as total science fiction and says will probably be possible in two years. Whenever it happens, he thinks it might change the way you use your AI devices.
Image: Omi
This is the (more normal) way most people will wear devices like Omi.
For now, the Omi’s actual purpose is much simpler: it’s an always-listening device (the battery apparently lasts three days on a charge) that you wear on a lanyard around your neck that can help you make sense of your day-to-day life. There’s no wake word, but you can still talk to it directly because it’s always on. Think of it as 80 percent companion and 20 percent Alexa assistant.
Omi can summarize a meeting or conversation and give you action items. It can give you information — Shevchenko offhandedly wondered about the price of Bitcoin during our conversation and got a notification from the Omi companion app a few seconds later with the answer. There’s also an Omi app store, which developers are already using to plug the audio input into things like Zapier and Google Drive.
For Shevchenko himself, though, Omi is a personal mentor above all else. “I was born in the middle of nowhere on an island near Japan,” he tells me, and always wanted access to the tech visionaries he grew up admiring. For years, he says he cold-emailed people like Mark Zuckerberg and Elon Musk asking for advice and mentorship on how to make it in tech but never got much response. With no real-life options, Shevchenko decided to build his own.
Omi already has a product called “Personas,” which allows you to plug in anyone’s X handle and create a bot that assumes their social network persona. When Shevchenko shares his screen with me, it shows he’s been chatting with an AI Elon Musk for a long time. “It helps me to understand what I should be working on tomorrow,” Shevchenko says. “Or when I’m talking to someone and I don’t know an answer to the question, it will give me a small nudge — it sometimes tells me I’m wrong!” His wearable heard him say he was sick a few days ago and has been reminding him ever since to get more rest. He asks it every month to give him feedback and tell him how to do better.
He gets a lot of notifications from the Omi app, including during our call, and not all of them make much sense — one was just a transcription of a sentence he’d said a minute earlier. Shevchenko acknowledges it’s early, but he doesn’t seem bothered by the system’s misses. The communication works for him.
Image: Omi
Omi’s tech is actually pretty simple — it’s mostly just a microphone. The AI is the trick.
Most people won’t use Omi this way, though. The product will ship widely in the second quarter of this year, but Shevchenko says the 5,000 people with an early version of the device are using it to help remember things, look up information, and perform other tasks common to AI assistants.
In that sense, Omi has a lot in common with devices like the Limitless Pendant and bears a striking resemblance to another wearable called Friend. When Friend launched last year, Shevchenko claimed Friend CEO Avi Schiffmann was stealing his work, and the subsequent beef included everything from sniping on X to a freestyle rap diss track. Omi was actually called Friend for a while, and Shevchenko says he changed the name both to avoid confusion and because Schiffmann dropped $1.8 million on Friend.com and subsequently dominated search results.
Shevchenko is confident that Omi can improve on those other devices. All of Omi’s code is open source, and there are already 250 apps in the store. Omi’s plan is to be a big, broad platform, rather than a specific device or app — the device itself is only one piece of the puzzle. The company is using models from OpenAI and Meta to power Omi, so it can iterate more quickly on the product itself.
For all their issues and underlying concerns, it’s clear that AI models are already good enough to feel like a true companion to millions of people. You can feel about that however you’d like, but from Omi and Friend to Character.AI and Replika, bot friends are quickly becoming real friends. What they need, then, is both more information about you and more ways to help you. Omi thinks the first answer is an always-on microphone, and the second is an app store. Then, I guess, comes the brain.
Samsung announced its first Galaxy Unpacked event of 2025, which it’s historically used to introduce the latest flagship Galaxy smartphones. That would be the Galaxy S25 family if our count is right, including the base model, the expected Galaxy S25 Ultra, and whatever else falls between. We’ll know for sure before too long with an unveiling scheduled for January 22nd.
But before the company even confirms what we’re getting, you can already place a reservation to preorder the device. If you sign up at Samsung’s website or the Shop Samsung app by January 22nd at 1 PM ET (which only entails submitting your name and email address) and later place your pre-order, you’ll get a $50 credit that’s good toward any additional devices or accessories you’re purchasing alongside it. That sadly means you can’t use the $50 toward the device itself. The fine print also mentions that you can get a $100 credit toward your pre-order of a qualifying 2025 TV (perhaps some of the models it announced at CES) or audio device using the same email address, hinting that we may see more than just smartphones at the event.
As for what to expect from the Galaxy S25 family, they could be some of the first “Qi2 Ready” devices, which would mean they’ll support the wireless charging standard, but require cases to use them with magnetic Qi2 chargers. A recent video leak suggests the Galaxy S25 Ultra may get a slight redesign with a curvier chassis. We may even get a Galaxy S25 “Slim” to turn the de facto trio of flagships into a quartet.
And we didn’t need to see the letters “AI” in Samsung’s teaser to know that would be a central selling point for its latest devices. We’ve already seen Galaxy AI deployed in several of its devices over the last couple of years, which includes features like Circle to Search, computational photography and video tricks with generative photo manipulation, summarized notes and transcriptions, and live translations. Samsung will presumably continue building that experience out as part of One UI 7.
Mythic Quest’s third season ended on a hopeful note as the Playpen team hunkered down to start developing an all-new expansion, but all that hard work looks like it’s just going to lead to more headaches judging from the show’s latest season 4 trailer.
While Mythic Quest’s new season will find Dana (Imani Hakim), Jo (Jessie Ennis), and Brad (Dani Pudi) basking in the success of their Cozy Galaxy project, things are going to be a bit tougher for Ian (Rob McElhenney) and Poppy (Charlotte Nicdao) as they try to figure out how keep players coming back to Playpen. In the trailer, David (David Hornsby) says that he wants the pair to take their time to come up with some genuinely good ideas. But it’s clear that he really wants those solutions quickly, which feels like part of why one of Ian’s first moves is to incorporate an AI duplicate of himself into his workflow.
Poppy’s idea to let Playpen players create their own in-game content seems a bit more on the level and like something that might actually get people logging on regularly. But of course, the first thing players want to do with the features is find ways to make their characters pantomime sex, which is exactly the sort of thing that gets Congress wondering who the game is actually for. No one actually says “Roblox” in the trailer, but it’s obvious that Mythic Quest’s writers room has been reading the news and getting the sense that something smells a bit off about games with these kinds of business models. Which is probably why the show won’t pull any punches when it returns on January 29th (and its spin-off debuts on March 26th).
Meta is testing eBay listings on Facebook Marketplace in the US, Germany, and France. The company said in a post on Wednesday that it’s making the change to comply with last year’s antitrust order by the European Union, though it continues to appeal the decision.
With the test, Facebook Marketplace users can browse eBay listings on Facebook Marketplace and then check out on eBay. Meta said it could “benefit” both platforms, as it exposes eBay sellers to Facebook’s audience, while giving Marketplace users access to “a broader array of listings from the eBay community.” The news was reported earlier by Bloomberg.
Image: Meta
“It’s a small test for now, so not all people will see it, but we hope to expand soon,” Meta spokesperson Jocelyn Jones said in an emailed statement to The Verge. Facebook Marketplace users will see both local eBay listings and items that can be shipped, according to eBay’s FAQ.
Last year, the EU fined Meta $840 million over claims it forcibly exposed Facebook users to Marketplace by linking its selling and social platforms. It also accused Meta of imposing “unfair trading conditions” on rival classified ads services to benefit Marketplace, and ordered Meta to stop engaging in this allegedly illegal behavior. At the time, Meta said it would work on a solution while appealing the fine, saying that the EU’s decision relies “on a hypothetical potential to harm competition.”
Update, January 8th: Added a statement and more information from Meta.
