"Classic First Amendment retaliation." That's how US District Judge Rita Lin described the Department of War's effort to blacklist Anthropic and designate it a supply-chain risk.

By all appearances, "these measures appear designed to punish Anthropic," Lin wrote in an order granting Anthropic's request for a preliminary injunction.

Officials seemingly had no authority to take such extreme actions without considering less restrictive alternatives or offering any evidence that Anthropic posed an urgent risk to national security, Lin said. Instead, "the Department of War’s records show that it designated Anthropic as a supply chain risk because of its 'hostile manner through the press.'"

Read full article

Comments

From Ars Technica - All content via this RSS feed

Drone startup BRINC announced Tuesday a significant upgrade for its law enforcement drones. BRINC’s newest model, Guardian, will have Starlink connectivity on every unit—a first for commercially available drones.

This new model, which will enter production later this year, has a flight time of over an hour and can reach a top speed of over 60 miles per hour. BRINC calls it the “first drone that can pursue vehicles.”

Additionally, Guardian can carry numerous payloads from its charging “nest,” including a floatation device, a defibrillator, epipens, the overdose-reversal drug Narcan, and more. The nest can also robotically swap batteries in about a minute, the company claims.

Read full article

Comments

From Ars Technica - All content via this RSS feed

Amazon is developing a new smartphone over a decade after discontinuing the Fire Phone, Reuters reported today, citing four anonymous “people familiar with the matter.”

Reuters said the phone is codenamed Transformer but couldn’t confirm what it might cost, how much Amazon has invested into development thus far, or how much Amazon expects to make off the device. Like any product reportedly under development, it’s possible that Amazon will never release the phone. Reuters’ sources noted that Transformer could be cancelled over finances or a change in strategy.

When reached for comment by Ars Technica, an Amazon spokesperson declined to comment on Reuters’ report.

Read full article

Comments

From Ars Technica - All content via this RSS feed

Warner Bros. just dropped a broody and haunting extended teaser for Denis Villeneuve's Dune: Part 3, the highly anticipated third film in the director's acclaimed franchise—the last in his planned trilogy.

(Spoilers for first two films in the franchise below.)

In 2021's Dune, we first met Frank Herbert's iconic anti-hero, Paul Atreides (Timothée Chalamet). That film culminated in the brutal defeat of House Atreides by rival House Harkonnen, with Paul and his mother, Lady Jessica (Rebecca Ferguson), fleeing to the desert and taking refuge with the Fremen. Among them is Chani (Zendaya), whom Paul has been seeing in visions all along.

Read full article

Comments

From Ars Technica - All content via this RSS feed

Elon Musk has ordered another round of job cuts at xAI after growing frustrated with the poor performance of its coding product, forcing out several more cofounders and parachuting in “fixers” from SpaceX and Tesla to audit the startup.

The latest overhaul of the 2-year-old startup follows the success of Anthropic and OpenAI, whose AI coding tools have shaken up the software industry, multiple people familiar with the decisions said.

Musk has dialled up the pressure after merging SpaceX with xAI in a $1.25 billion deal, as he attempts to meet a June deadline for what could be the biggest stock market listing in history. The world’s richest man has said his goals are to launch AI data centers into space, build factories on the Moon, and colonize Mars.

Read full article

Comments

From Ars Technica - All content via this RSS feed

The key to working at a place like Ars Technica is solid news judgment. I'm talking about the kind of news judgment that knows whether a pet peeve is merely a pet peeve or whether it is, instead, a meaningful example of the Ways that Technology is Changing our World.

The difference between the two is one of degree: A pet peeve may drive me nuts but does not appear to impact anyone else. A Ways that Technology is Changing our World story must be about something that drives a lot of people nuts.

"But where is the threshold?" I hear you asking plaintively. "It's extremely important that I know when something crosses the line from pet peeve to important, chin-stroking journalism topic!"

Read full article

Comments

From Ars Technica - All content via this RSS feed

At this January's massive NAMM music tech show in Los Angeles, six products won "best of show" awards. Several of them went to major music and electronic brands like Yamaha and Boss, but one of the six went to Neural DSP, a much smaller company started in 2017 by Chilean immigrants to Finland.

From its base in the Helsinki area, Neural has made itself an expert in the use of machine learning, robots, and impulse response technology to automate the construction of incredibly lifelike guitar amp modeling software. It quickly jumped into the top ranks of an industry dominated by brands like Universal Audio, Kemper, Line 6, and Fractal. For a hundred bucks, you could buy one of the company's plugins and sound like a guitar god with a $10,000 recording chain of amps, cabinets, effects pedals, and microphones.

In 2020, Neural branched out into hardware, putting its tech not in your computer but in a floor-based box covered with footswitches and called the Quad Cortex. While the company's plugins could each replace one entire pedalboard of gear—plus a few amps and cabs—the Quad Cortex could replace a Guitar Center-sized warehouse of devices, offering hundreds of amps, cabs, and effects.

Read full article

Comments

From Ars Technica - All content via this RSS feed

COLUMBUS, Ohio—Protestors stood in the snow outside the offices of Ohio’s utility regulator in January to say they were fed up with rising electricity rates.

Even a few years ago, the scene would have been hard to imagine, considering the complexity of utility costs and the obscurity of state regulatory agencies. But rate hikes in Ohio and across the country have provoked frustrated consumers to demand answers.

“It’s just getting harder and harder now to live,” said Steve Van Kuiken, a United Church of Christ pastor in Columbus who is part of a community group opposing rate increases. “The working class is really getting squeezed, and everything’s going up.”

Read full article

Comments

From Ars Technica - All content via this RSS feed

For miles around xAI's makeshift power plant in Southaven, Mississippi, neighbors have endured months of constant roaring, erupting pops, and bursts of high-pitched whining from 27 temporary gas turbines installed without consulting the community.

In a report on Thursday, NBC News interviewed residents fighting to shut down xAI's turbines. They confirmed that xAI operates the turbines day and night, allegedly tormenting residents in order to power xAI founder Elon Musk's unbridled AI ambitions.

Eventually, 41 permanent gas turbines—that supposedly won't be as noisy—will be installed, if xAI can secure the permitting. In the meantime, xAI has erected a $7 million "sound barrier" that's supposed to mitigate some of the noise.

Read full article

Comments

From Ars Technica - All content via this RSS feed

On Friday, a body that advises US judges revised the document it created to help judges grapple with scientific issues. The move came after a group of Republican state attorneys general wrote a letter to complain about the document's chapter on climate change, with one of the letter's criticisms being that it treated human influence on climate as a fact. In response to the letter, the Federal Judicial Center has now deleted the entire chapter.

The Federal Judicial Center has been established by statute as the "research and education agency of the judicial branch of the United States Government." As part of that role, it prepares documents that can serve as reference material for judges unfamiliar with topics that find their way into the courtroom. Among those projects is the "Reference Manual on Scientific Evidence," now in its fourth edition. Prepared in collaboration with the National Academies of Science, the document covers the process of science and specific topics that regularly appear before the courts, like statistical techniques, DNA-based identification, and chemical exposures.

When initially released in December, the fourth edition included material on climate change prepared by two authors at Columbia University. But a group of attorneys general from Republican-leaning states objected to this content. At the end of January, they sent a letter to the leadership of the Federal Judicial Center outlining their issues. Many of them focus on the text that accepts the reality of human-driven climate change as a fact.

Read full article

Comments

From Ars Technica - All content via this RSS feed

Google continues to turn the screws on free YouTube users, expanding a test that restricts access to song lyrics on YouTube Music. Users without a premium subscription have found that Google's streaming music service only shows song lyrics a few times before demanding money.

For as long as YouTube Music has existed, lyrics have been accessible to all users in the mobile app. That started to change over recent months as Google tested a paywall. The lyrics section still appears in the app when playing a song with a free account, but opening it eats into a limited allotment of lyric views. A substantial uptick in user reports, spotted by 9to5Google, suggests this restriction is now rolling out widely.

"You have [x] views remaining," the app now warns free users who access lyrics. It looks like users get five free lyric views before they have to pay up. Google has still neglected to officially announce the addition of this feature to its Premium subscription—there's no mention of lyrics being part of the paid tier on Google's support page.

Read full article

Comments

From Ars Technica - All content via this RSS feed

Donald Trump is so far not stepping in to help Elon Musk end a lawsuit raised by the Securities and Exchange Commission (SEC) over his 2022 Twitter takeover, a US district judge said this week.

Filed by the SEC in the final days of Joe Biden's administration, the lawsuit seeks $150 million in disgorgement, plus interest, as well as civil penalties and an injunction blocking Musk from future wrongdoing.

The complaint alleged that Musk quietly acquired a 9 percent stake in Twitter without filing necessary timely disclosures to alert other investors of a potential change in company control. This allowed Musk to acquire over 70 million shares at an artificially lower price, the SEC alleged, causing substantial economic harm to investors selling Twitter common stock, some of whom have separately sued.

Read full article

Comments

From Ars Technica - All content via this RSS feed

If all goes according to plan Monday, NASA's launch team at Kennedy Space Center in Florida will load 755,000 gallons of super-cold propellants into the rocket built to send the Artemis II mission toward the Moon.

The fuel loading is part of a simulated countdown for the Space Launch System rocket, a final opportunity for engineers to rehearse for the day NASA will send four astronauts on a nearly 10-day voyage around the far side of the Moon and back to Earth. The Artemis II mission will send humans farther from Earth than ever before. The astronauts will be the first to launch on NASA's SLS rocket and the first people to travel to the vicinity of the Moon in more than 53 years.

Charlie Blackwell-Thompson, NASA's launch director for the Artemis II mission, will supervise the practice countdown from a firing room inside the Launch Control Center a few miles away from the SLS rocket at Kennedy Space Center. In a recent briefing with reporters, she called the Wet Dress Rehearsal—"wet" refers to the loading of liquid propellants—the "best risk reduction test" for verifying all is ready to proceed into the real countdown.

Read full article

Comments

From Ars Technica - All content via this RSS feed

RAMP—the predominantly Russian-language online bazaar that billed itself as the “only place ransomware allowed”—had its dark web and clear web sites seized by the FBI as the agency tries to combat the growing scourge threatening critical infrastructure and organizations around the world.

Visits to both sites on Wednesday returned pages that said the FBI had taken control of the RAMP domains, which mirrored each other. RAMP has been among the dwindling number of online crime forums to operate with impunity, following the takedown of other forums such as XSS, which saw its leader arrested last year by Europol. The vacuum left RAMP as one of the leading places for people pushing ransomware and other online threats to buy, sell, or trade products and services.

I regret to inform you

“The Federal Bureau of Investigation has seized RAMP,” a banner carrying the seals of the FBI and the Justice Department said. “This action has been taken in coordination with the United States Attorney’s Office for the Southern District of Florida and the Computer Crime and Intellectual Property Section of the Department of Justice.” The banner included a graphic that appeared on the RAMP site, before it was seized, that billed itself as the “only place ransomware allowed.”

Read full article

Comments

From Ars Technica - All content via this RSS feed

On Friday, OpenAI engineer Michael Bolin published a detailed technical breakdown of how the company's Codex CLI coding agent works internally, offering developers insight into AI coding tools that can write code, run tests, and fix bugs with human supervision. It complements our article in December on how AI agents work by filling in technical details on how OpenAI implements its "agentic loop."

AI coding agents are having something of a "ChatGPT moment," where Claude Code with Opus 4.5 and Codex with GPT-5.2 have reached a new level of usefulness for rapidly coding up prototypes, interfaces, and churning out boilerplate code. The timing of OpenAI's post details the design philosophy behind Codex just as AI agents are becoming more practical tools for everyday work.

These tools aren't perfect and remain controversial for some software developers. While OpenAI has previously told Ars Technica that it uses Codex as a coding tool to help develop the Codex product itself, we also discovered, through hands-on experience, that these tools can be astonishingly fast at simple tasks but remain brittle beyond their training data and require human oversight for production work. The rough framework of a project tends to come fast and feels magical, but filling in the details involves tedious debugging and workarounds for limitations the agent cannot overcome on its own.

Read full article

Comments

From Ars Technica - All content via this RSS feed

The Federal Communications Commission today issued a warning to late-night and daytime talk shows, saying these shows may no longer qualify for an exemption to the FCC's equal-time rule. Because the FCC is chaired by vocal Trump supporter Brendan Carr, changing how the rule is enforced could pressure shows into seeking out more interviews with Republican candidates.

The public notice providing what the FCC calls "guidance on political equal opportunities requirement for broadcast television stations" appears to be part of the Trump administration's campaign against alleged liberal bias on broadcast TV. Carr, who has eroded the FCC's historical independence from the White House, previously pressured ABC to suspend Jimmy Kimmel and threatened ABC’s The View with the equal-time rule.

The Carr FCC's public notice today said that federal rules "prevent broadcast television stations, which have been given access to a valuable public resource (namely, spectrum), from unfairly putting their thumbs on the scale for one political candidate or set of candidates over another." These rules come from "the decision by Congress that broadcast television stations have an obligation to operate in the public interest—not in any narrow partisan, political interest," the Carr FCC said.

Read full article

Comments

From Ars Technica - All content via this RSS feed

Moxie Marlinspike—the pseudonym of an engineer who set a new standard for private messaging with the creation of the Signal Messenger—is now aiming to revolutionize AI chatbots in a similar way.

His latest brainchild is Confer, an open source AI assistant that provides strong assurances that user data is unreadable to the platform operator, hackers, law enforcement, or any other party other than account holders. The service—including its large language models and back-end components—runs entirely on open source software that users can cryptographically verify is in place.

Data and conversations originating from users and the resulting responses from the LLMs are encrypted in a trusted execution environment (TEE) that prevents even server administrators from peeking at or tampering with them. Conversations are stored by Confer in the same encrypted form, which uses a key that remains securely on users’ devices.

Read full article

Comments

From Ars Technica - All content via this RSS feed

A company called GRU Space publicly announced its intent to construct a series of increasingly sophisticated habitats on the Moon, culminating in a hotel inspired by the Palace of the Fine Arts in San Francisco.

On Monday, the company invited those interested in a berth to plunk down a deposit between $250,000 and $1 million, qualifying them for a spot on one of its early lunar surface missions in as little as six years from now.

It sounds crazy, doesn't it? After all, GRU Space had, as of late December when I spoke to founder Skyler Chan, a single full-time employee aside from himself. And Chan, in fact, only recently graduated from the University of California, Berkeley.

Read full article

Comments

From Ars Technica - All content via this RSS feed

On Monday, the US Department of the Interior announced that it was pausing the leases on all five offshore wind sites currently under construction in the US. The move comes despite the fact that these projects already have installed significant hardware in the water and on land; one of them is nearly complete. In what appears to be an attempt to avoid legal scrutiny, the Interior is blaming the decisions on a classified report from the Department of Defense.

The second Trump administration announced its animosity toward offshore wind power literally on day one, issuing an executive order on inauguration day that called for a temporary halt to issuing permits for new projects pending a re-evaluation. Earlier this month, however, a judge vacated that executive order, noting that the government has shown no indication that it was even attempting to start the re-evaluation it said was needed.

But a number of projects have gone through the entire permitting process, and construction has started. Before today, the administration had attempted to stop these in an erratic, halting manner. Empire Wind, an 800 MW farm being built off New York, was stopped by the Department of the Interior, which alleged that it had been rushed through permitting. That hold was lifted following lobbying and negotiations by New York and the project developer Orsted, and the Department of the Interior never revealed why it changed its mind. When the Interior Department blocked a second Orsted project, Revolution Wind offshore of southern New England, the company took the government to court and won a ruling that let it continue construction.

Read full article

Comments

From Ars Technica - All content via this RSS feed

San Francisco was affected by a massive power outage over the weekend. It started with a fire at a substation in the city on Saturday afternoon, causing a blackout that at times affected as much as a third of the city, leaving more than 130,000 homes without power. Among the city's affected critical systems were the traffic lights, which paralyzed Waymo's fleet of robotaxis, stopping them in their tracks and clogging traffic.

Any recent visitor to San Francisco can't help but notice the profusion of sensor-festooned autonomous vehicles on the roads, especially the all-white Jaguar I-Paces that belong to Waymo. The robotaxi company has more than 800 AVs in its Bay Area fleet; that can feel like a conservative estimate when you see five or six at a time—invariably with no occupants—within a block.

The cars navigate the city, combining high-resolution maps with inputs from lidar, optical, and other sensors on the upfitted Jags. The cars drive conservatively, but they can get confused in edge cases—earlier this month, Waymo issued a recall to fix a problem in which its robotaxis would illegally pass stopped school buses.

Read full article

Comments

From Ars Technica - All content via this RSS feed

Before he became chairman of the Federal Communications Commission, Brendan Carr seemed to be a big believer in the agency's role as an independent branch of the federal government. According to the pre-2025 version of Brendan Carr, the White House interfered with the agency's independence when a Democratic president publicly urged the FCC to adopt net neutrality rules.

When the Biden-era FCC reinstated Obama-era net neutrality rules in 2024, Carr alleged that President Biden "took the extraordinary step to pressure the FCC—an independent agency that is designed to operate outside undue political influence from the Executive Branch." As evidence, Carr pointed to a 2021 executive order in which Biden called on agency heads to "consider using their authorities" for various types of pro-competitive policies, including the adoption of net neutrality rules.

Carr said that President Obama similarly "pressure[d] an independent agency into grabbing power that the Legislative Branch never said it had delegated." Obama's intrusion into this independence, according to Carr, came in November 2014 when the president released a two-minute video urging the agency to implement net neutrality rules and reclassify broadband providers as common carriers.

Read full article

Comments

From Ars Technica - All content via this RSS feed

Google is generally happy to see people using generative AI tools to create content, and it’s doubly happy when they publish it on its platforms. But there are limits to everything. Two YouTube channels that attracted millions of subscribers with AI-generated movie trailers have been shuttered.

Screen Culture and KH Studio flooded the site with fake but often believable trailers. The channels, which had a combined audience of more than 2 million subscribers, became a thorn in Google’s side in early 2025 when other YouTubers began griping about their sudden popularity in the age of AI. The channels produced videos with titles like “GTA: San Andreas (2025) Teaser Trailer” and “Malcom In The Middle Reboot (2025) First Trailer.” Of course, neither of those projects exist, but that didn’t stop them from appearing in user feeds.

Google demonetized the channels in early 2025, forcing them to adopt language that made it clear they were not official trailers. The channels were able to monetize again, but the disclaimers were not consistently used. Indeed, many of the most popular videos from those channels in recent months included no “parody” or “concept trailer” disclosures. Now, visiting either channel’s page on YouTube produces an error reading, “This page isn’t available. Sorry about that. Try searching for something else.”

Read full article

Comments

From Ars Technica - All content via this RSS feed

Federal vaccine advisors hand-selected by anti-vaccine Health Secretary Robert F. Kennedy Jr. have voted to eliminate a recommendation that all babies be vaccinated against hepatitis B on the day of birth. The decision was made with no evidence of harm from that dose and no evidence of any benefit from the delay.

Public health experts, medical experts, and even some members of the panel decried the vote, which studies and historical data indicate will lead to more infections in babies that, in turn, will lead to more cases of chronic liver disease, liver cancer, and premature death.

“I will just say we have heard ‘do no harm’ is a moral imperative. We are doing harm by changing this [recommendation],” Cody Meissner, a pediatrician and voting member of ACIP said as he voted against the change.

Read full article

Comments

From Ars Technica - All content via this RSS feed

The panel of federal vaccine advisors hand-selected by anti-vaccine Health Secretary Robert F. Kennedy Jr. has once again punted on whether to strip recommendations for hepatitis B vaccinations for newborns—a move it tried to make in September before realizing it didn’t know what it was doing. The decision to delay the vote today came abruptly this afternoon when the panel realized it still does not understand the topic or what it was voting on.

Prior to today’s 6–3 vote to delay a decision, there was a swirl of confusion over the wording of what a new recommendation would be. Panel members had gotten three different versions of the proposed recommendation in the 72 hours prior to the meeting, one panelist said. And the meeting’s data presentations this morning offered no clarity on the subject—they were delivered entirely by anti-vaccine activists who have no subject matter expertise and who made a dizzying amount of false and absurd claims.

“Completely inappropriate”

Overall, the meeting was disorganized and farcical. Kennedy’s panel has abandoned the evidence-based framework for setting vaccine policy in favor of airing unvetted presentations with misrepresentations, conspiracy theories, and cherry-picked studies. At times, there were tense exchanges, chaos, confusion, and misunderstandings.

Read full article

Comments

From Ars Technica - All content via this RSS feed

Amazon Prime Video has scaled back an experiment that created laughable anime dubs with generative AI.

In March, Amazon announced that its streaming service would start including “AI-aided dubbing on licensed movies and series that would not have been dubbed otherwise.” In late November, some AI-generated English and Spanish dubs of anime popped up, including dubs for the Banana Fish series and the movie No Game No Life: Zero. The dubs appear to be part of a beta launch, and users have been able to select “English (AI beta)” or “Spanish (AI beta)” as an audio language option in supported titles.

“Absolutely disrespectful”

Not everyone likes dubbed content. Some people insist on watching movies and shows in their original language to experience the media more authentically, with the passion and talent of the original actors. But you don’t need to be against dubs to see what’s wrong with the ones Prime Video tested.

Read full article

Comments

From Ars Technica - All content via this RSS feed