Defense Secretary Pete Hegseth will reportedly give Anthropic until Friday to drop certain guardrails for military use, as reported by Axios. The outlet also reported that CEO Dario Amodei met with Hegseth yesterday as the Pentagon ratcheted up pressure on the AI company to give in to its demands.
The makers of Claude have reportedly been offered an ultimatum: Either yield to the government's demands to remove limits for certain military applications, or potentially be forced to tailor its AI model to the government's needs under the Defense Production Act.
Anthropic, for its part, has said that while it was willing to adopt certain policies for the Pentagon, it would not allow its model to be used for mass surveillance of Americans or for the development of autonomous weapons.
Claude is currently the only AI model employed in some of the government's most sensitive work. "The only reason we're still talking to these people is we need them and we need them now. The problem for these guys is they are that good," a defense official told Axios.
The Pentagon is reportedly ramping up conversations with OpenAI and Google about using their models for classified work. ChatGPT and Gemini are already approved for unclassified government use. Elon Musk's xAI also recently signed with the DoD to use Grok in classified systems.This article originally appeared on Engadget at https://www.engadget.com/ai/the-pentagon-has-reportedly-given-anthropic-until-friday-to-let-it-use-claude-as-it-sees-fit-203549467.html?src=rss
Google sent out an AI-generated news alert that included the N-word, according to reporting by Deadline. The push notification featured a link to a story by The Hollywood Reporter regarding an incident at the recent BAFTA Film Awards. The word appeared in the notification under the link.
This was first spotted by Instagram user Danny Price, who accompanied a screengrab with a caption reading "what an interesting Black History Month this has turned out to be." Google has since apologized and said that it has "removed the offensive notification" and is "working to prevent this from happening again."
This story follows the aforementioned BAFTA incident, in which an audience member with Tourette syndrome shouted the N-word when Michael B. Jordan and Delroy Lindo took to the stage to present an award. Tourette syndrome activist John Davidson, who made the comment, said he was "deeply mortified if anyone considers my involuntary tics to be intention or to carry any meaning." The incident has sparked outrage and a renewed discussion on the realities of living with vocal tics.
Asking for more grace for the person who shouted a racist slur instead of for Michael B. Jordan and Delroy Lindo, who had to push through being embarrassed in front of their peers.But thats often the expectation that Black people are just supposed to be ok with being https://t.co/MqHbC8XwsA Jemele Hill (@jemelehill) February 23, 2026
AI makes lots of high-profile errors and this isn't the first time it has ruined a news alert. Apple actually scrapped its own AI push notifications last year when the tool made a series of embarrassing mistakes, including wrongly telling readers that the man accused of killing UnitedHealthcare CEO Brian Thompson, Luigi Mangione, had shot himself.This article originally appeared on Engadget at https://www.engadget.com/ai/google-sent-an-ai-generated-push-alert-that-included-a-racial-slur-195951493.html?src=rss
Three years after announcing plans to produce a film based on the viral YouTube short, The Backrooms (Found Footage) in 2023, A24 has released the first teaser for its adaptation. Backrooms, as the film is now called, is directed by the short's original creator, Kane Parsons, and will be released on May 29, 2026.The teaser offers little to go on for anyone who hasn't watched the original short or the series of videos Parsons made after it, but it is replete with The Backrooms' hallmark: ominous liminal spaces. Layered over footage of stranger and stranger rooms (or perhaps one room becoming the platonic empty retail spaces popularized by short), the voice of Chiwetel Ejiofor tells the film's other star, Renate Reinsve, about a "place" he discovered that's full of rooms.Parsons' original video is inspired by a creepypasta called "The Backrooms" that originated on the forum 4chan in 2019. The YouTube series expanded on the basic concept of a liminal space that exists outside reality with monsters and a mysterious company researching The Backrooms. It's unclear how much of that larger lore will be incorporated into the feature film adaptation, but since the teaser is missing the digital video filter that gave the YouTube short its distinct look, it seems possible Parsons could be going for something a bit different. Well that, and the fact the film stars two Oscar-nominated actors.This article originally appeared on Engadget at https://www.engadget.com/entertainment/tv-movies/heres-the-first-teaser-for-a24s-adaptation-of-the-backrooms-194300513.html?src=rss
Waymo had set out some big plans for expanding its autonomous vehicle taxi program across the US in 2025 and it appears to be continuing that pace into 2026. Today, the company announced that the first public riders can begin using its fully autonomous ride-hailing service in Dallas, Houston, San Antonio and Orlando. To start, these robotaxi rides will only be available for a small number of people with the Waymo app in those cities. "We will be inviting new riders on a rolling basis to ensure a seamless experience across our initial service areas, as we meaningfully scale our operations ahead of opening our service to everyone later this year," the company said in the blog post announcing the expansion.Google-owned Waymo is now operating in ten commercial metro areas. It announced its plans to start testing its vehicles in these four US cities in November. The company also began a test phase in Miami at that time, and Waymo's robotaxi service began accepting riders in that locale in January.This article originally appeared on Engadget at https://www.engadget.com/transportation/waymo-will-start-offering-robotaxi-rides-in-four-more-cities-192841871.html?src=rss
We're less than a month from the availability of Death Stranding 2: On the Beach on PC, and today Sony released the required specs. Despite designer Hideo Kojima being known for spectacle in his projects, the minimum specs are quite reasonable. The low graphics preset runs on an NVIDIA GeForce GTX 1660 or AMD Radeon RX 5500 XT 8GB, and that will give players average performance of 1080p at 30 frames per second, which is rough but at least makes the game available for players who havent upgraded in awhile. Intel Core i3-10100 or AMD Ryzen 3 3100 are the recommended GPUs for that graphics tier. As is typical for PC gaming, though, the higher end performance options will require beefier internals.This port of Death Stranding 2 will also boast a few firsts. The title will mark the debut of Pico as an upscaling option on PC. This upscaler was made by Guerilla Games and was also used for the Death Stranding sequel on the PlayStation 5. The game will also be adding support for ultrawide views. The cutscenes can be viewed in a 21:9 aspect ration and gameplay can be displayed at 32:9. This option will be available for both PC and the PS5 versions of the game, and an ultrawide monitor won't be required to enable this view option.This article originally appeared on Engadget at https://www.engadget.com/gaming/sony-reveals-the-death-stranding-2-required-pc-specs-184411856.html?src=rss
A new app will notify users if smart glasses are likely nearby. The aptly named Nearby Glasses was developed in response to media coverage outlining how glasses like Meta's Ray-Bans have been used to film people without their consent.
As first reported by 404 Media, the app detects the unique Bluetooth signature emitted by smart glasses and sends a push alert that someone wearing the device may potentially be nearby. I consider it to be a tiny part of resistance against surveillance tech, the app's developer Yves Jeanrenaud told 404 Media.
Smart glasses have sparked increased privacy concerns, especially as Meta is reportedly working to add facial recognition technology to its Meta Ray-Bans. OpenAI is also reported to have a pair of smart glasses in the works. It bears mentioning that false positives may occur, including from VR headsets.
Nearby Glasses is currently available on the Google Play Store and GitHub.This article originally appeared on Engadget at https://www.engadget.com/apps/someone-made-an-app-to-warn-you-if-smart-glasses-are-nearby-183359723.html?src=rss
As spicy as the PlayStation State of Play that took place a couple of weeks ago was overall, there was one major first-party game that was notably absent from the showcase: Marvels Wolverine. Insomniac Games latest superhero blockbuster was already slated for a fall release and now the studio has revealed exactly when youll be able to get your claws on it. Marvels Wolverine is coming to PS5 on September 15.Thats it. Thats the announcement. Theres no new trailer to accompany the news, other than a six-second release date reveal video that popped up on YouTube. Insomniac previously said it would reveal more details about Marvels Wolverine this spring.Technically, this release date means that Marvels Wolverine will debut in the last week of summer rather than in the fall. Still, its one of the relatively few blockbuster games you can expect in the tail end of this year because many major developers and publishers will be staying well clear of GTA VI. Insomniacs game will have a couple of months of breathing room before GTA VI soaks up all of the air in the gaming world when it arrives on November 19 assuming Rockstar doesnt announce another delay. However, parent company Take-Two plans to rev up its marketing machine for the game this summer, so its looking like GTA VIs release date will hold this time.This article originally appeared on Engadget at https://www.engadget.com/gaming/playstation/marvels-wolverine-will-hit-ps5-on-september-15-175500927.html?src=rss
Apple just dropped a full trailer for the fifth season of its hit sci-fi show For All Mankind. This is the first real look at the upcoming batch of episodes, which premiere on March 27. We got an extremely short teaser trailer last month but that only showed a guy on a motorcycle riding across Mars.
This is the first real-deal trailer and it's absolutely stuffed with footage indicating where the next season will take viewers. I'm going to get into some spoilers here, so read at your own risk.
For the uninitiated, For All Mankind is an alternate history show that started with a simple premise. What if Russia landed on the moon before America? That has since ballooned into all kinds of stuff which include, as mentioned above, a potential war on Mars.
For All Mankind is a show famous for its time jumps, and season five takes us all the way to an alternate version of 2012. Many of the show's original surviving characters are still kicking around, but they are old as paste and not exactly fit for high-octane space travel. Remember, the first episode started in the 1960s. Franchise lead Ed Baldwin (Joel Kinnamen) looks particularly dusty.
Much of the footage features newer characters, including the grandson of Baldwin. Season four ended with a Mars colony asserting its independence via asteroid theft. Now it looks like Earth is striking back, which could lead to a full-scale war. This is giving me The Expanse vibes, which is never a bad thing.
The show must be clocking good numbers for Apple TV+, as the streamer recently announced a spinoff called Star City. Details are scant, but it looks to cover similar events of the mainline show from Russia's perspective.
New episodes of For All Mankind air each Friday. This season will feature ten episodes and concludes on May 29.This article originally appeared on Engadget at https://www.engadget.com/entertainment/tv-movies/for-all-mankinds-latest-trailer-teases-a-war-on-mars-174822481.html?src=rss
A common theme in online age verification laws is the tension between user privacy and preventing children from accessing harmful or inappropriate content. Now the UK is sending a not-so-subtle message to Reddit on the subject, to the tune of 14.5m ($19.6 million). The nation's Information Commissioner's Office (ICO) accused the company of using childrens data and potentially exposing them to inappropriate content.Children under 13 had their personal information collected and used in ways they could not understand, consent to or control, UK Information Commissioner John Edwards wrote in a statement. That left them potentially exposed to content they should not have seen. This is unacceptable and has resulted in todays fine.In July 2025, Reddit began requiring age verification to access adult content in the UK, in compliance with the Online Safety Act. However, that's only used to block under-18 users from sexually explicit, violent or other mature posts. The platform also prohibits users under 13 from accessing it altogether and enforcement of that policy is lax. It merely requires users to declare, when signing up, that they're over 13. The ICO (accurately) described the method as "easy to bypass."In its defense, Reddit told the BBC that it "didn't require users to share information about their identities, regardless of age, because we are deeply committed to their privacy and safety." The company said it would appeal the decision. "The ICO's insistence that we collect more private information on every UK user is counterintuitive and at odds with our strong belief in our users' online privacy and safety," the spokesperson added."It's concerning that a company the size of Reddit failed in its legal duty to protect the personal information of UK children," Edwards said. "Companies operating online services likely to be accessed by children have a responsibility to protect those children by ensuring theyre not exposed to risks through the way their data is used. To do this, they need to be confident they know the age of their users and have appropriate, effective age assurance measures in place.Reddit failed to meet these expectations, he added. They must do better, and we are continuing to consider the age assurance controls now implemented by the platform. The ICO also accused Reddit of failing to conduct a data protection impact assessment by January 2025.The Guardian notes that the 14.5m fine is the third-largest handed down by the ICO. It trails only a 20m fine for British Airways involving a data breach disclosure and an 18.4m penalty for Marriott Hotels for exposing over 300 million customer records in a hack.This article originally appeared on Engadget at https://www.engadget.com/social-media/reddit-fined-196-million-over-age-verification-checks-in-the-uk-173705048.html?src=rss
Google is building another data center in Texas and says this one will use "advanced air-cooling technology" to limit water consumption. Google is claiming that water use will be limited to "critical campus operations" like kitchens.
These specifics follow the previously announced two-year $40 billion investment the company has pledged in the Lone Star State. The company is also touting some 7,800 megawatts of net energy generation and capacity it has contracted with utility providers to add to the Texas grid.
The company's resource-conscious commitments come as communities nationwide are pushing back on data center construction, amid concerns that they are raising electric bills, worsening global greenhouse gas emissions and often using gargantuan amounts of fresh water. But don't worry, because OpenAI CEO Sam Altman says the water concerns are "fake" and that "it also takes a lot of energy to train a human."
A growing number of voices in tech, most notably Tesla CEO Elon Musk, have said that building data centers in space would address many of these concerns. But some experts believe the potential environmental downside could be devastating.This article originally appeared on Engadget at https://www.engadget.com/ai/google-claims-its-building-data-centers-that-barely-use-any-water-171411965.html?src=rss
