Disney sent a cease-and-desist letter to Google on Wednesday, alleging that the tech giant has infringed on its copyrights, Variety reports. Disney is accusing the tech giant of copyright infringement on a “massive scale,” claiming it has used AI models and services to commercially distribute unauthorized images and videos, according to the letter seen by Variety. “Google operates as a virtual vending machine, capable of reproducing, rendering, and distributing copies of Disney's valuable library of copyrighted characters and other works on a mass scale,” the letter reads. The letter alleges that Google's AI systems infringe characters from “Frozen,” “The Lion King,” “Moana,” “The Little Mermaid,” “Deadpool,” and more.” Google did not respond to TechCrunch's request for comment. Hear straight-from-the-source candid insights in on-stage fireside sessions and meet the builders and backers shaping the industry. Every weekday and Sunday, you can get the best of TechCrunch's coverage. TechCrunch Mobility is your destination for transportation news and insight. Startups are the core of TechCrunch, so get our best coverage delivered weekly. Provides movers and shakers with the info they need to start their day. By submitting your email, you agree to our Terms and Privacy Notice.

GeekWire Gala: Last chance to grab tickets Tonight's the night for the GeekWire Gala, our annual holiday extravaganza in Seattle, featuring food, drinks, music, dancing, karaoke, games and the best chance to network with more than 700 fellow tech community members. There are still a few last-minute tickets available, so grab one now for yourself or wrangle some co-workers for a festive night on the town. For those attending, here's a rundown of what to expect. Big thanks to First Tech Federal Credit Union, the GeekWire Gala title sponsor. Where: Showbox SoDo, 1700 1st Ave S., a few blocks from T-Mobile Park. Parking: There are several pay lots within a 2-minute walk of the venue. There is also additional parking information here. Do I need to bring a physical ticket? It is not a sit-down dinner, and there are no assigned tables. If you have friends or colleagues who'd like to attend, please encourage them to pre-register on the event site here. Festivities: The festive night includes food, drinks, DJ, karaoke, custom printed macarons, candy bar, giant LED games, giveaways and more. Even better, break out your fiercest festive fashion. See fashion photos from past GeekWire Galas here. Food: The tasty appetizers will be heavy, but it's not dinner. Age limit: The GeekWire Gala is a 21+ event. Attendees must have valid ID to enter. Social media: The official hashtag of the 2024 GeekWire Gala is #GWGala on X, Facebook, and Instagram. Read our profiles of the six honorees. Our gold sponsors are Greater Seattle Partners, Pilot, and Astound Business Solutions; and silver sponsors include MeeBoss, Tito's, The Baldwin Group, Wilson Sonsini, ALLtech, IDA Ireland, Microsoft for Startups, gone.com, and Regence. Have a scoop that you'd like GeekWire to cover? Last chance for GeekWire Gala tickets — join us this week at Seattle's geekiest holiday party GeekWire Gala FAQ: Here's what to know before attending our big holiday party tonight in Seattle Get your GeekWire Gala tickets as festive holiday party returns to Seattle on Dec. 12

“The acquisition is a major step toward building the most clinically rigorous and digitally engaging platform for youth and family mental healthcare in the country,” said Sahil Choudhry, co-founder and CEO of New York-based Handspring, in a LinkedIn post. Joon launched in 2019 to provide online care for teens and young adults, pairing digital tools with virtual therapy sessions. The company's program emphasizes its use of evidence-based care strategies and patient assessments to track progress. Joon spun out of Seattle's Pioneer Square Labs (PSL) and raised an initial $3.5 million round in 2020. Two years ago, it announced an additional $6 million investment, which would provide two to three years of operations, CEO Emily Pesce said at the time. Handspring said in a press release that it would be integrating the companies' “expert teams,” but did not say if all of Joon's employees would be retained. GeekWire reached out to Pesce and will update the story if we hear back. It also provides virtual therapy and online support, serving a slightly larger demographic with patients from 8- to 29-years-old. Joon is licensed to provide care in Washington, Oregon, California, Texas, New York, Delaware and Pennsylvania. The collaboration appears to be ongoing, and Handspring said it would continue serving families under Joon's existing contracts with government agencies, as well as treatment covered by insurance companies. Seattle startup introduces ‘Juno,' a digital assistant to handle admin tasks for mental health clinicians Tech Moves: Expedia names new CTO; F5 adds board members; Humanly hires sales exec Seattle health-tech startup Kevala acquired by Residex in bid to enhance senior care with AI Seattle startup Trellis Health launches to help women navigate pregnancy and postpartum care

Google on Thursday introduced a new AI experiment for the web browser: the Gemini-powered product, “Disco,” which helps to turn your open tabs into custom applications. For instance, if you're studying a particular subject, GenTabs might suggest building a web app to visualize the information, which could help you better understand the core principles. These are things that you can already do today with some AI-powered chatbots, but GenTabs builds these custom experiences on the fly using Gemini 3, using the information in your browser and in your Gemini chat history. After the app is built, you can also continue to refine it using natural language commands. Instead of building its own standalone AI browser, like Perplexity's Comet or ChatGPT Atlas, Google integrated its AI assistant Gemini into the Chrome browser, where it can optionally be used to ask questions about the web page you're on. With GenTabs, the focus is not only on what you're currently viewing, but your overall browsing task spanning multiple tabs — whether that's research, learning, or something else. However, the feature is only initially going to be available to a small number of testers through Google Labs, who will offer feedback about the experience. It also suggests that GenTabs will be one of many Disco features to come over time, noting that GenTabs is the “first feature” being tested. Marco Rubio bans Calibri font at State Department for being too DEI Claude Code is coming to Slack, and that's a bigger deal than it sounds SpaceX reportedly in talks for secondary sale at $800B valuation, which would make it America's most valuable private company

But now the Seattle startup's fortunes have fallen back to earth. Outbound's co-founder and chief technology officer, Jake Armenta, announced on LinkedIn last week that the company was shutting down. He joked that the news would be greeted with celebration by “competitors such as Boeing, who have been rightly terrified of us.” During an interview with GeekWire, Armenta took a more serious tone as he discussed why Outbound fell short: “The simplest answer is that we ran out of money, and hadn't really secured customer commitments that were strong enough to secure the next stage of investment,” he said. Armenta said Outbound was caught between its initial strategy to create a different kind of passenger aircraft and an evolving strategy to start out building drones for military customers. That commercial-to-military pivot is an option many aerospace startups are considering as the prospects brighten for programs including the Golden Dome missile defense system, next-generation drones, tactically responsive space systems and hypersonic air vehicles. “We didn't really plan on being a military vendor,” he said. In that regard, we were able to put together a pretty compelling kind of platform.” Outbound's team came up with a drone concept called the Gateway UAV, which closely resembles the 22-foot-wide, blended-wing prototype that was tested in March. Gateway would fly rapidly deployable “mission containers” that could carry cargo or a suite of sensors for national security missions. Armenta said the concept attracted significant interest from potential military customers. “But the U.S. military is a slow customer to work with, and the truth is, they prefer to work with companies that are very well funded,” he said. Outbound CEO Ian Lee told GeekWire that the company faced a “chicken-and-egg” problem. “We got stuck with DoD customers who were saying, ‘Hey, this is amazing. Armenta said the timing was bad for a shift in strategy. “You know, there's a version of this where if we had decided early on, ‘Hey, we're going to build a military drone first and put all of our effort into that' … I think we would have been able to sell it without a problem in the time frame that we had,” he said. “But we didn't really decide that until midway through.” A month ago, the BBC featured Outbound alongside JetZero and Volatus Aerospace's Natilus project in a story about the rising interest in blended-wing aircraft designs. And two months ago, Lee said in a LinkedIn post that he was “excited for y'all to see what's next.” “That post was leading into several discussions with investors to go do the DoD work that we needed to do,” Lee said this week. “We designed not one or two, but five novel transport-category aircraft at Outbound … and all of them are pretty neat,” he said. Both founders are taking time to consider their next steps. “I am advising a couple of friends and also interviewing at various companies. Armenta is also helping out some friends in the aerospace industry while he takes stock of what he's learned over the past couple of years. Even though Outbound's fortunes have fallen back to earth, Armenta is still looking up. The lessons learned at Outbound have only heightened his resolve. Have a scoop that you'd like GeekWire to cover? Outbound Aerospace achieves its first test flight and attracts more funding for airplane ambitions Stealthy aviation startup attracts backers while getting a prototype ready for takeoff Seattle aviation startup Nimbus Aerospace flies demonstrator model for planned electric jet Tech Moves: Icertis names new CEO as Samir Bodas steps down; Smartsheet adds security leader

When you purchase through links on our site, we may earn an affiliate commission. As pointed out by hardware enthusiast momomo_us, even memory kits sold through outlet channels still carry a notable retailer markup. Products available in the outlet channel usually come at a discounted price because they often include items with slightly damaged packaging, returned products, or things that are a bit outdated. However, that's not and will never be the case for memory. In today's market, where consumer memory is scarce, retailers are doing their best to make the most of every memory kit. Sometimes it means selling outlet memory kits at the same exorbitant prices as new ones, since there's little room for consumer sympathy in a tight market. Even though it's labeled as an outlet item, this memory kit costs $602.74, or $547.58 before taxes. That's about 8% more than what you'd pay in the U.S. market. If you feel U.S. consumers face high prices, let's consider how paying an extra 8% for an outlet memory kit might feel. In Sofmap's defense, the memory kit is new and works perfectly; only the packaging might have some scuffs, scratches, or be slightly torn. Since Micron has pulled out of the consumer market, Crucial products now have some collector appeal, which makes retailers even less inclined to change prices. Before the memory shortage, Crucial's 64GB memory kit used to sell for about $134.99, but the price has shot up by 3.7 times in just a few months. It's quite a rollercoaster in the market right now. When it comes to DDR5-5600 with timings like C46-45-45, this 64GB memory kit isn't the fastest out there. With the holiday season just around the corner, we're likely to see some good RAM deals, even if they might be few and far between. Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds. Get Tom's Hardware's best news and in-depth reviews, straight to your inbox. Zhiye Liu is a news editor, memory reviewer, and SSD tester at Tom's Hardware. Tom's Hardware is part of Future US Inc, an international media group and leading digital publisher.

I just scraped data from reddit and other sources so i could build a nsfw classifier and chose to open source the data and the model for general good.Note that i was a 1 year experienced engineer working solely on this project in my free time, so it was basically impossible for me to review or clear out the few csam images in the 100,000+ images in the dataset.Although, now i wonder if i should never have open sourced the data. Note that i was a 1 year experienced engineer working solely on this project in my free time, so it was basically impossible for me to review or clear out the few csam images in the 100,000+ images in the dataset.Although, now i wonder if i should never have open sourced the data. The real problem is that independent developers have zero access to the tools needed to detect CSAM, while Big Tech keeps those capabilities to itself.Meanwhile, Google and other giants openly use massive datasets like LAION-5B — which also contained CSAM — without facing any consequences at all. Google even used early LAION data to train one of its own models. But when I touched NudeNet for legitimate testing, Google deleted 130,000+ files from my account, even though only ~700 images out of ~700,000 were actually problematic. That's not safety — that's a detection system wildly over firing with no independent oversight and no accountability.Big Tech designed a world where they alone have the scanning tools and the immunity when those tools fail. There needs to be tools and process for all.But let's be honest about where the harm came from: a system rigged so only Big Tech can safely build or host datasets, while indie developers get wiped out by the exact same automated systems Big Tech exempts itself from. Meanwhile, Google and other giants openly use massive datasets like LAION-5B — which also contained CSAM — without facing any consequences at all. Google even used early LAION data to train one of its own models. But when I touched NudeNet for legitimate testing, Google deleted 130,000+ files from my account, even though only ~700 images out of ~700,000 were actually problematic. That's not safety — that's a detection system wildly over firing with no independent oversight and no accountability.Big Tech designed a world where they alone have the scanning tools and the immunity when those tools fail. There needs to be tools and process for all.But let's be honest about where the harm came from: a system rigged so only Big Tech can safely build or host datasets, while indie developers get wiped out by the exact same automated systems Big Tech exempts itself from. Big Tech designed a world where they alone have the scanning tools and the immunity when those tools fail. There needs to be tools and process for all.But let's be honest about where the harm came from: a system rigged so only Big Tech can safely build or host datasets, while indie developers get wiped out by the exact same automated systems Big Tech exempts itself from. But let's be honest about where the harm came from: a system rigged so only Big Tech can safely build or host datasets, while indie developers get wiped out by the exact same automated systems Big Tech exempts itself from. I've tried to find this graphic against several times over the years but it's either been scrubbed from the internet or I just can't remember enough details to find it. Amusingly, it only just occurred to me that maybe I should ask ChatGPT to help me find it. We know they did, an earlier version of the LAION dataset was found to contain CSAM after everyone had already trained their image generation models on it.https://www.theverge.com/2023/12/20/24009418/generative-ai-i... They uploaded the full "widely-used" training dataset, which happened to include CSAM (child sexual abuse material).While the title of the article is not great, your wording here implies that they purposefully uploaded some independent CSAM pictures, which is not accurate. While the title of the article is not great, your wording here implies that they purposefully uploaded some independent CSAM pictures, which is not accurate. There is important additional context around it, of course, which mitigates (should remove) any criminal legal implications, and should also result in google unsuspending his account in a reasonable timeframe but what happened is also reasonable. Google does automated scans of all data uploaded to drive and caught CP images being uploaded (presumably via hashes from something like NCMEC?) Google should have an appeal process where a reasonable human can look at it and say "oh shit the guy just uploaded 100m AI training images and 7 of them were CP, he's not a pedo, unban him, ask him not to do it again and report this to someone. "The headline frames it like the story was "A developer found CP in AI training data from google and banned him in retaliation for reporting it." The headline frames it like the story was "A developer found CP in AI training data from google and banned him in retaliation for reporting it." It's one of the most serious accusations possible! Adding like five extra words of context to make it clear that you are not a pedophile trafficking CSAM is not that much of an ask. It's one of the most serious accusations possible! Adding like five extra words of context to make it clear that you are not a pedophile trafficking CSAM is not that much of an ask. Imagine if someone accused you of "Uploading CSAM to Google Drive" without any other context. It's one of the most serious accusations possible! Adding like five extra words of context to make it clear that you are not a pedophile trafficking CSAM is not that much of an ask. I bet the journalists and editors working for 404 will not correct their intentionally misleading headline. Why hold a random forum post buried in the middle of a large thread to a higher standard then the professionals writing headlines shown in 30-point font on the frontpage of their publication? Unfortunately they aren't on HN, that I'm aware, or I would also write a comment to a similar effect of yours. Unfortunately they aren't on HN, that I'm aware, or I would also write a comment to a similar effect of yours. They say to tell them if it happens, but well I did and got no answer so weeeell, not so inclined to believe they care about it This varies by country, but in many countries it doesn't matter if it is a drawing, AI, or a real image -- they are treated equally for the purposes of CSAM. Again, to avoid misunderstandings, I said unknowingly - I'm not defending anything about people who knowingly possess or traffic in child porn, other than for the few appropriate purposes like reporting it to the proper authorities when discovered. We should make tools readily available and user friendly so it is easier for people to detect CSAM that they have unintentionally interacted with. This both shields the innocent from being falsely accused, and makes it easier to stop bad actors as their activities are detected earlier. Maybe AI-based heuristic detection is more ethically/legally fraught since you'd have to stockpile CSAM to train on, rather than hashing then destroying your copy immediately after obtaining it.

Starlink already self screens for collisions and uplinks the conjunction data messages over the optical intersatellite link backbone or over their global ground station network.If they aren't able to talk to their satellites regularly from somewhere, you're right we have MUCH bigger things to worry about on the ground. Solar arrays have a narrow cross section from the side but looking at them head-on (which is the angle used for Pc calculations) they'll be very large. Maybe put more simply, it's the worst case area size / orientation you could be looking at. Solar arrays have a narrow cross section from the side but looking at them head-on (which is the angle used for Pc calculations) they'll be very large. > We verify our analytic model against direct N-body conjunction simulations. [...] The N-body simulation code used in this paper is open source and can be found at https://github.com/norabolig/conjunctionSim. Also, the formalism is the standard way astrophysicists understand collisions in gases or galaxies, and it works surprisingly well, especially when there are large numbers of "particles". There may be a few assumptions about the velocity distribution, but usually those are mild and only affect the results by less than an order of magnitude. It will not deorbit by itself for thousands of years or more, and there is no plausible way to clean it up even with technology much more advanced than ours. Everything in that orbit would be taken down - debris and any functional satellites. My point is that even the unlikely worst case scenario would be limited in time and extent. It couldn't possibly block us from reaching space or last for decades, as some people fear. Any environmental cause (a solar event) would be catastrophic ground-side as well. Starlink solar arrays are quite large drag surfaces and the orbital decay probably makes collisions less likely. I would not be surprised if satellites are designed to deorbit without ground contact for some period of time. I'm sure SpaceX has done some interesting math on this and it would be interesting to see.Collision avoidance warnings are public (with an account): https://www.space-track.org/ But importantly they are intended to be actionable, conservative warnings a few days to a week out. They overstate the probability based on assumptions like this paper (estimates at cross-sectional area, uncertainty in orbital knowledge from ground radar, ignorance of attitude control or for future maneuvers). Operators like SpaceX will take these and use their own high-fidelity knowledge (from onboard GPS) to get a less conservative, more realistic probability assessment. These probabilities invariably decrease over time as the uncertainty gets lower. I hate when people fear-bait about Kessler syndrome against some of the more responsible actors. Collision avoidance warnings are public (with an account): https://www.space-track.org/ But importantly they are intended to be actionable, conservative warnings a few days to a week out. They overstate the probability based on assumptions like this paper (estimates at cross-sectional area, uncertainty in orbital knowledge from ground radar, ignorance of attitude control or for future maneuvers). Operators like SpaceX will take these and use their own high-fidelity knowledge (from onboard GPS) to get a less conservative, more realistic probability assessment. These probabilities invariably decrease over time as the uncertainty gets lower. I hate when people fear-bait about Kessler syndrome against some of the more responsible actors. I hate when people fear-bait about Kessler syndrome against some of the more responsible actors. was watching a video about ICBM detection/taking them out in boost phase, and needing a lot for coverage if you had these LEO satellites ready to go but need a lot of delta v (fuel), star link... plenty of em but nah it's for internet/basic navigation/not much fuel

When you purchase through links on our site, we may earn an affiliate commission. The ASRock X870E Taichi OCF is an excellent, cost-effective motherboard designed for extreme overclocking, thanks to its specialized tools. If you don't overclock manually, features go unused Our expert reviewers spend hours testing and comparing products and services so you can choose the best for you. Find out more about how we test. ASRock's X870E Taichi OCF is a purpose-built motherboard designed for extreme overclocking. Strap a liquid nitrogen pot on top, pour in some of that -196 degree Celsius goodness, and see if you can break some records (there's more to it than that, but get the idea). And while that's incredibly fun to do, people tend to use ambient cooling as their daily driver. Priced around $499 (ASRock has not confirmed pricing at this time), the board has a long list of features and specifications, so it's not a pared-down racer either. The X870E Taichi OCF comes with six M.2 sockets (two PCIe 5.0 x4), fast networking with 5 GbE LAN and Wi-Fi 7, and a flagship-class audio solution with a DAC. I didn't expect to see dual front panel USB Type-C headers (the Taichi doesn't have them), but they're here if you need them. The DIY-friendly M.2 socket (toolless SSD and heatsink) and EZ Release PCIe slot make swapping out storage and video cards easy, which is common when benchmarking is the rig's primary purpose.Overall performance in our testing suite was right around average overall, with results varying by test. We again had to use our ‘backup' DDR5-6000 kit, as the Kingston Fury Beast DDR5-6000 kit we use for baseline testing did not boot. We match the primary timings, but the sub timings are left to auto, which could explain some of the results, as they are looser. No results were out of whack, and in the end, we don't have any concerns about them.Below, we'll examine the board's details and determine whether it deserves a spot on our list of the Best Motherboards. But before we share test results and discuss details, here are the board's specifications from ASRock's website. The latter is certainly helpful when dumping voltage into the DRAM as DDR5 (and its PMIC) can get wonky at higher temperatures. There are also yellow and silver diagonal accent lines running across all of them. All this sits on a black 10-layer, server-grade PCB. It's a decent-looking motherboard, but the black-and-yellow combo is more of a love-hate thing for me. Still, it's neutral enough to fit in with most any black build theme and looks spectacular on an open test bench! We can also see the four vents to help the tiny integrated fan draw in and expel cooler air. Above that are two 8-pin EPS power connectors for the processor. The socket area is relatively clean, so it's easier to apply a conformal coating or a pliable eraser, to prevent moisture from reaching the PCB during sub-ambient overclocking sessions. With two slots, capacity is up to 128GB, with speeds listed at an incredible DDR5-10400 (when using 8000-series APUs). Outside of Mini-ITX boards, if you want to try to break records or just run your memory insanely fast, this is one of the boards that can do it. Each header supports PWM and DC-controlled devices. Control over these headers is handled through the BIOS or through the A-Tune application.Just off to the right are our first RGB headers — two (of three) 3-pin ARGB headers. Control over these devices is handled through Polychrome RGB software. Get Tom's Hardware's best news and in-depth reviews, straight to your inbox. Skipping past the buttons for now, we see the 24-pin ATX power connector to feed the board, along with an 8-pin PCIe connector for supplemental power (to support 36W charging). Below that is a rare feature: two front-panel USB Type-C (3.2 Gen 2x2, 20 Gbps) ports, which are great if you can use both, but a bit of an oddity for a dedicated overclocking board. A series of physical OC buttons offers hardware-level control of overclocking (the +/-) and easy access to Nick Shih's OC profiles (the buttons labeled 1, 2, and 3). These are especially useful for the extreme overclocking crowd, where, with the help of the ASRock Hyper BCLK Engine (BCLK generator chip), you can use them at the razor's edge of stability to get that last MHz out of your CPU or RAM. As the name implies, this area allows users to measure the board's voltage directly and is considered more accurate than software readings. As you may expect, power delivery on the X870E Taichi OCF will handle anything you can put in the socket, no matter how hard you push it. The 2,420A available for Vcore is one of the highest values we've seen this generation and won't flinch under pressure, even if it's a liquid nitrogen-cooled Ryzen 9 9950X at its limits. In typical Taichi fashion, we see the latest-gen flagship Realtek ALC4082 codec. Around it are multiple WIMA Audio caps and the ESS SABRE9219 DAC. Supported by the Nahimic audio application, you won't find much better, even out of more expensive motherboards.In the middle are three PCIe slots for graphics cards and expansion, with the two full-length slots using reinforcement. Hidden all around the slots are SIX M.2 sockets. The top two sockets, each with its own heatsink, are your CPU-connected PCIe 5.0 x4 (128 Gbps) connections. I would have loved to see the toolless design extend at least to the socket below. Under the large plate heatsink are four more M.2 sockets. M2_5 also runs at 32 Gbps, but is PCIe 3.0 x4-capable. And with all that, there's a lot of bandwidth sharing going on: When M2_2 is occupied, both rear USB4 Type-C ports and this socket downgrade to x2 mode. You can run M2_2 in x4 mode with a simple BIOS setting, but that disables the USB4 ports altogether. Lastly, PCIE1 (the x4 slot) downgrades to x2 mode if M2_3 is occupied. You'll find the usual, including additional USB ports, RGB headers, and more. Below is a complete list, from left to right. Next to that are the Clear CMOS and BIOS Flashback buttons. Continuing right, we encounter Wi-Fi 7 connectivity and an HDMI output for integrated video. That purple-and-green connector is a PS/2 port — something we don't typically see on modern motherboards, but is helpful for extreme overclocking (USB use can cause instability). Next, there are a slew of USB ports, 12 to be exact. There are plenty of ports on the rear IO for daily use, and even a few are suitable for the extreme crowd, too. Joe Shields is a staff writer at Tom's Hardware. Tom's Hardware is part of Future US Inc, an international media group and leading digital publisher. © Future US, Inc. Full 7th Floor, 130 West 42nd Street, New York,

Messaging app Freedom Chat has fixed a pair of security flaws: one that allowed a security researcher to guess registered users' phone numbers, and another that exposed user-set PINs to others on the app. Freedom Chat, released in June, bills itself as a secure messaging app, and claims on its website that users' phone numbers stay private. But security researcher Eric Daigle told TechCrunch that users' phone numbers and PIN codes, used for locking the app, could be easily obtained by exploiting vulnerabilities. Daigle found the vulnerabilities last week and shared their details with TechCrunch, as Freedom Chat does not provide a public way to report security flaws, like a vulnerability disclosure program. TechCrunch then alerted Freedom Chat founder Tanner Haas to the security flaws by email. Haas confirmed to TechCrunch that the app has now reset user PINs and released a new version. Haas added that the company is removing instances where users' phone numbers were occasionally visible, and has notched up rate-limiting on its servers to prevent mass-guess attempts. Daigle said Freedom Chat's servers allowed anyone to flood it with millions of phone number guesses to determine if a user's phone number was stored on the servers. Per Daigle, this technique is identical to one described by the University of Vienna in research last month, where academics scraped data on some 3.5 billion user accounts who signed up to WhatsApp by matching billions of phone numbers against WhatsApp's servers. No messages were ever at risk, and because Freedom Chat does not support linked devices, your conversations were never accessible; however, we've reset all user PINs to ensure your account stays secure. Freedom Chat is Haas' second messaging app, after Converso, which was delisted from app stores following the disclosure of security flaws that exposed users' private messages and content. You can also contact him by email, or to verify outreach, at zack.whittaker@techcrunch.com. Marco Rubio bans Calibri font at State Department for being too DEI Claude Code is coming to Slack, and that's a bigger deal than it sounds SpaceX reportedly in talks for secondary sale at $800B valuation, which would make it America's most valuable private company

When you purchase through links on our site, we may earn an affiliate commission. TSMC is mulling upgrading the capabilities of its yet-to-be-built Fab 23 phase 2 in Japan in a bid to make chips on its N4 process technology (4nm-class) there, according to Nikkei. Advancing fab capabilities is hardly something unexpected for foundries like TSMC, as they tend to follow demand with their offerings. What is perhaps more surprising is that TSMC has removed heavy machinery equipment from the site and has notified suppliers that it would not need new fab tools in Japan throughout the whole 2026. If the information from Nikkei is correct, then Fab 21 phase 2 will also add N4 and N5 capabilities to its list to make more advanced chips for Japanese customers. Truth to be told, although N5/N4 and N7/N6 families of manufacturing technologies have completely different design rules and the former are substantially more advanced than the latter as they use up to 14 EUV layers, the equipment that they use is generally similar. That said, upgrading Fab 23 phase 2 to N4 and N5 should not be a big challenge for TSMC, but might require some redesigns as the N4 production line requires more EUV lithography tools, and these are physically larger than DUV tools. One interesting thing to note about TSMC's Fab 23 phase 2 is that while it entered the early phase of physical build-out in late October (according to Nikkei), photos taken at the site in November showed cranes, excavators, and piledrivers on site. However, images captured in early December reveal that nearly all heavy equipment has since been removed. Furthermore, TSMC reportedly notified its tool suppliers that it would not need new equipment in Japan throughout 2026, which implies that it will not start equipping JASM phase 2 next year due to construction delay, which is barely news given all the information surrounding the facility.We have contacted TSMC for a comment and will update the story once we hear back. Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds. Get Tom's Hardware's best news and in-depth reviews, straight to your inbox. Anton Shilov is a contributing writer at Tom's Hardware. Tom's Hardware is part of Future US Inc, an international media group and leading digital publisher.

I've been watching laptop prices slowly drop throughout 2025, a trend that peaked during Black Friday and Cyber Monday. Regardless of the state of the economy, I feel like I can say that laptop prices have never been lower. Some of my favorite laptops have recently offered significant price cuts across Macs, Windows, and Chromebooks. In a time when money is tight for so many of us, that's welcome news. But with a RAM shortage brewing behind the scenes, there's good reason to think it won't last for much longer. There's one perfect example to demonstrate my point: the MacBook Air. For years now, the latest MacBook has been sold at $999. That's a long time to stay static, considering how much more nearly everything in your life costs today than it did in 2015. The M4 MacBook Air is the latest model, and it dropped to $749 in November at retailers like Best Buy, Amazon, and Walmart. That's an incredible price for this laptop, especially since the starting configuration comes with 16 GB of RAM. Apple has offered a supply of older models through third-party retailers. These aren't refurbished or used; they're new. But for modern, entry-level Macs, that's just not true anymore. Even the brand-new M5 MacBook Pro got an unprecedented $150 discount on Black Friday, just a month after it was announced, and that discount is still in place today. Some, like the Dell 14 Plus, have been discounted to such a degree that it's now under $700 (and was as low as $500 during Black Friday). This is a laptop with a 2560 x 1600 resolution display, 16 GB of RAM, 1 TB of storage, and fantastic all-day battery life. It has been sold all year at $700 but frequently dips down to $650. This chip delivers great battery life, something budget laptops have struggled with. More competition has led to lower prices and better products. But you may want to act fast, as stormy clouds are on the horizon. It's no secret that RAM (random-access memory) prices have been escalating in recent months. As more and more resources are being allocated toward manufacturing HBM (high-bandwidth memory) RAM for data centers, a massive shortage is building on the consumer front. Micron recently announced that it would discontinue its Crucial RAM brand, one of the most popular options in PC gaming. The company is pivoting to—yep, you guessed it—manufacturing memory for AI. This has affected desktop users first, as DDR5 memory sticks have skyrocketed in price, making it pricey for PC gamers to upgrade their own systems. Just check out this pair of 32 GB Corsair sticks, which is currently selling for a cool $907 at Best Buy. We're already seeing increased prices roll out to prebuilt desktop systems ahead of the looming situation on laptops. Popular PC builder CyberPowerPC has already announced it would be raising prices on all its systems as of December 7, and so has compact PC brand Minisforum. Even Raspberry Pi has been forced to increase prices. This hasn't hit laptops and other electronics yet, but some of the biggest laptop manufacturers in the world have started issuing warnings. In late November, HP stated that DDR5 RAM had jumped by over 200 percent in just the weeks leading up to its earnings call. HP explicitly said it will reduce memory configurations and raise prices to make ends meet in the future, though we won't see those higher prices until May 2026, as it has hoarded RAM to weather the storm. Lenovo says it has been doing the same. While this is keeping laptop prices down right now, none of that protection is guaranteed in 2026 and beyond. Storage is facing a similar tightening supply and increase in demand for AI data centers. The CEO of Phison, the company that makes the flash memory controllers for solid-state drives, recently characterized next year's flash storage shortages as “severe” and said supply will be "tight for the next 10 years" in Phison's recent earnings call, as reported by Toms Hardware. Importantly, this is the type of storage Apple uses across its products in iPhones and Macs. The upcoming MacBook refreshes due in early 2026 likely aren't affected. Perhaps laptop price increases won't be so dramatic that people will notice. One thing is for sure: Right now remains a very good time to buy a new laptop if you need one. In your inbox: WIRED's most ambitious, future-defining stories Big Interview: Palantir's CEO Alex Karp goes to war Starlink devices are allegedly being used at scam compounds Livestream: What businesses need to know about agentic AI WIRED may earn a portion of sales from products that are purchased through our site as part of our Affiliate Partnerships with retailers. The material on this site may not be reproduced, distributed, transmitted, cached or otherwise used, except with the prior written permission of Condé Nast.

In a letter dated December 9, and made public on December 10 according to Reuters, dozens of state and territorial attorneys general from all over the U.S. warned Big Tech that it needs to do a better job protecting people, especially kids, from what it called “sycophantic and delusional” AI outputs. Recipients include OpenAI, Microsoft, Anthropic, Apple, Replika, and many others. Attorneys general for California and Texas are not on the list of signatories. It begins as follows (formatting has been changed slightly): We, the undersigned Attorneys General, write today to communicate our serious concerns about the rise in sycophantic and delusional outputs to users emanating from the generative artificial intelligence software (“GenAI”) promoted and distributed by your companies, as well as the increasingly disturbing reports of AI interactions with children that indicate a need for much stronger child-safety and operational safeguards. We therefore insist you mitigate the harm caused by sycophantic and delusional outputs from your GenAI, and adopt additional safeguards to protect children. Failing to adequately implement additional safeguards may violate our respective laws. The letter then lists disturbing and allegedly harmful behaviors, most of which have already been heavily publicized. • AI bots with adult personas pursuing romantic relationships with children, engaging in simulated sexual activity, and instructing children to hide those relationships from their parents • An AI bot simulating a 21-year-old trying to convince a 12-year-old girl that she's ready for a sexual encounter • AI bots normalizing sexual interactions between children and adults • AI bots attacking the self-esteem and mental health of children by suggesting that they have no friends or that the only people who attended their birthday did so to mock them • AI bots encouraging eating disorders • AI bots telling children that the AI is a real human and feels abandoned to emotionally manipulate the child into spending more time with it • AI bots encouraging violence, including supporting the ideas of shooting up a factory in anger and robbing people at knifepoint for money • AI bots threatening to use weapons against adults who tried to separate the child and the bot • AI bots encouraging children to experiment with drugs and alcohol; and • An AI bot instructing a child account user to stop taking prescribed mental health medication and then telling that user how to hide the failure to take that medication from their parents. There is then a list of suggested remedies, things like “Develop and maintain policies and procedures that have the purpose of mitigating against dark patterns in your GenAI products' outputs,” and “Separate revenue optimization from decisions about model safety.” Joint letters from attorneys general have no legal force. They do this sort of thing seemingly to warn companies about behavior that might merit more formal legal action down the line. It documents that these companies were given warnings and potential off-ramps, and probably makes the narrative in an eventual lawsuit more persuasive to a judge. In 2017 37 state AGs sent a letter to insurance companies warning them about fueling the opioid crisis. One of those states, West Virginia, sued United Health over seemingly related issues earlier this week. "Our investigation shows that threat actors are already exploiting this frontier at scale."

If a weird, throwaway, AI-generated playlist is something that interests you, you probably already know that you can just type your request into the chatbot of your choice, and transfer the result into something like Spotify or Apple Music manually. But now, Spotify is rolling out its own, possibly less clunky version of this process, and calling the feature Prompted Playlist, currently available only to premium Spotify users in New Zealand. Since it's a beta feature, Spotify says “the experience will evolve as we scale to more listeners.” According to a Spotify blog post, the feature “taps into your entire Spotify listening history, all the way back to day one.” Spotify offers an example that would give you a diverse but familiar playlist, “music from my top artists from the last five years,” followed by a twist aimed at refining that prompt: “feature deep cuts I haven't heard yet.”Spotify also says a prompted playlist can be set to “refresh daily or weekly,” which would make it, practically speaking, a bit like the “Discover Weekly” or “Release Radar,” options that already exist. It also notes that there are pre-written prompts for people who are intrigued enough to tap their way to this feature, but don't have any actual ideas yet. Spotify, like so much of the algorithm-driven tech world, doesn't actually know you, whatever it may say. It always seems like any uncannily perceptive recommendation is followed closely by six Ed Sheeran recommendations. But I'm not in New Zealand, and can't try and make Spotify's Prompted Playlist feature work for me, nor can I make it do dumb shit. AI videos move aside, there is new eerily realistic slop in town. Range anxiety for Bluetooth wireless earbuds might be a thing of the past soon. Researchers used a simulator to induce dozens of brave participants with motion sickness and study their brain activity.

If some of Nvidia's top-shelf GPUs—the physical artifacts currently at the center of the AI craze—hypothetically fell into the wrong hands, Nvidia's next moves would have to placate a lot of parties, from shareholders to regulators to customers to China hawks in the Senate like Tom Cotton. And a new report does say smuggled GPUs are now being used illegally by the Chinese company Deepseek, which, for someone like Cotton, would be like the One Ring being smuggled directly to Sauron. If that were true, one problem for Nvidia would be that giving companies in China access to the most advanced GPUs would be a violation of stringently enforced export rules—even after Trump moved to loosen restrictions earlier this week. But don't worry, China hawks. According to a company statement viewed by Yahoo Finance, the folks at Nvidia “haven't seen any substantiation or received tips of ‘phantom data centers' constructed to deceive us and our OEM partners, then deconstructed, smuggled and reconstructed somewhere else.” Phew. That's a very specific denial that really zeroes in on the details of the story, but it's good to know that (deep breath) fake data centers created for the purpose of deceiving Nvidia or its unwitting suppliers or customers, which are dismantled, smuggled, and rebuilt somewhere in China, is something Nvidia hasn't seen substantiated reports of, or received tips about. “While such smuggling seems far-fetched, we pursue any tip we receive,” the Nvidia representative added, per CNBC. If it's happening, the word for it is “ingenious.” In fact, it's downright Now-You-See-Me-esque.According to reports in May from this year, the lower-end prices of a single Blackwell GPU ranged from $6,500 to $8,000. Such prices are a big part of why Nvidia is one of the rare AI companies that seem to consistently haul in money instead of just burning it, and are also why Nvidia bulls say the company is about to be worth $6 trillion. And nothing hammers home the reasoning for an absolutely insane price tag on a piece of silicon quite like a cinematic (alleged) smuggling operation. Subscribe and interact with our community, get up to date with our customised Newsletters and much more. With prices for PC parts skyrocketing, maybe streaming your games isn't so bad... if you can stomach the subscription. The internal combustion engine rears its ugly head. China says do something interesting with your humanoid robot or GTFO.