SoylentNews

"Fnord666" writes:

Civil liberty concerns spur FAA to revise drone no-fly zones near ICE vehicles:

In January 2026, during the height of protests against immigration raids in Minneapolis, federal agents shot and killed 37-year-old Renee Good . Before even gathering all the facts , the Department of Homeland Security labeled the mother of three an "anti-ICE rioter" who "weaponized her vehicle against law enforcement" in an "act of domestic terrorism."

Days later, the feds announced a major expansion of "no-fly zones" in the name of national security. While such no-fly zones used to be about controlling aircraft, they now often focus on small drones. The expanded no-fly zones announced on January 16 prohibited such drones from flying within 3,000 lateral feet and 1,000 vertical feet of federal facilities.

But for the first time, the order extended no-fly zones to ground vehicles belonging to the Department of Homeland Security. Even while the vehicles were in motion. Even if they were unmarked. And even if their routes had not been announced.

This exceptionally ambiguous policy posed real danger to people like Rob Levine , a freelance photojournalist and commercial photographer in Minneapolis for nearly four decades. Since Levine got his remote-pilot certification and bought his first drone in 2016, he has flown a small fleet of DJI quadcopter drones to take aerial photographs and videos of Minnesota's rivers, bridges, and cities, along with crowds gathered for outdoor concerts and parades. More recently, he has documented Twin City residents protesting the increased presence of federal agents in their community.

Levine immediately stopped flying when he saw the no-fly notice. The notice said government agencies could shoot down or seize drones "deemed to pose a credible safety or security threat," and it warned of civil and even criminal penalties for drone operators.

"I saw what these federal agents were willing to do, the violence they were willing to visit upon even constitutional observers here in the Twin Cities who were just photographing what they were doing," Levine told Ars.

Good's killing had occurred just six blocks from his home. "It didn't take much imagination to think what they would do to somebody with a drone, and so for weeks I didn't go fly," he said.

A week after the no-fly zone warning, the situation in Minneapolis escalated further when Customs and Border Protection officers killed Alex Pretti , a 37-year-old intensive care nurse, after wrestling him to the ground and shooting him multiple times.

Levine wanted his drones back in the air. But when he sought guidance from the Federal Aviation Administration, the agency candidly acknowledged that the no-fly zone warning was "ambiguous" and "therefore, any flight carries the risk of inadvertent violation."

Could such a policy possibly be legal?

Linux cryptographic code flaw offers fast route to root

An Anonymous Coward writes:

Patches land for authencesn flaw enabling local privilege escalation

https://hackread.com/linux-kernel-vulnerability-copy-fail-full-root-access/

Developers of major Linux distributions have begun shipping patches to address a local privilege escalation (LPE) vulnerability arising from a logic flaw.

The newly disclosed LPE, dubbed Copy Fail (CVE-2026-31431), comes from a vulnerability in the Linux kernel's authencesn cryptographic template.

"An unprivileged local user can write four controlled bytes into the page cache of any readable file on a Linux system, and use that to gain root," the writeup from security biz Theori explains.

The kernel reads the page cache when it loads a binary, so modifying the cached copy amounts to altering the binary for the purpose of program execution. But doing so doesn't trigger any defenses focused on file system events like inotify.

The proof of concept exploit is a 10-line, 732-byte Python script capable of editing a setuid binary to gain root on almost all Linux distributions released since 2017.

Copy Fail is similar to other LPE bugs such as Dirty Cow and Dirty Pipe, but its finders claim it doesn't require winning a race condition and it's more broadly applicable.

It's not remotely exploitable on its own – hence LPE – but if chained with a web RCE, malicious CI runner, or SSH compromise, it could be relevant to an external attacker. The bug is of most immediate concern to those using multi-tenant Linux systems, shared-kernel containers, or CI runners that execute untrusted code.

According to Theori, the vulnerability also represents a potential container escape primitive that could affect Kubernetes nodes, because the page cache is shared across the host.

Linux distros Debian, Ubuntu, and SUSE have issued patches for the problem, as have overseers of other distros.

Red Hat initially said it was going to defer the fix but later changed its

guidance to indicate it will go along with other distros and patch promptly.

The CVE has been rated High severity, 7.8 out of 10.

Theori researcher Taeyang Lee identified the vulnerability, with the help of the company's AI security scanning software, Xint Code.

The number of bug reports has surged in recent months, helped by AI-powered flaw-finders. Microsoft just reported the second largest number of patches ever.

Dustin Childs, head of threat awareness for Trend Micro's Zero Day Initiative, expects this is due to security teams using AI to hunt bugs. "There are many things we could speculate on to justify the size, but if Microsoft is like the other programs out there (including ours), they are likely seeing a rise in submissions found by AI tools," he wrote earlier this month.

AI-assisted vulnerability research recently prompted the Internet Bug Bounty (IBB) program to suspend awards until it can understand how to manage the growing volume of reports.

Original Submission #1  Original Submission #2 

An Anonymous Coward writes:

Apple wants to kill your Time Capsule, but they run NetBSD so they can't:

It seems like Apple is finally going to remove support for AFP from macOS, twelve years after first moving from AFP to SMB for its default network file-sharing technology. This change shouldn't impact most people, as it's highly unlikely you're using AFP for anything in 2026. Still, there is one small group of people to whom this change has an actual impact: owners of Apple's Time Capsule devices. Time Capsules only support AFP and SMB1, and with SMB1 being removed from macOS ages ago, and now AFP being on the chopping block as well, macOS 27 would render your Time Capsule more or less unusable.

It's important to note that the last Time Capsule sold by Apple, the fifth generation, was released in 2013, and the product line as a whole was discontinued in 2018. If you bought a Time Capsule in the twilight years of the line's availability, I think you have a genuine reason to be perturbed by Apple cutting you off from your product if you upgrade to macOS 27, but at least you have the option of keeping an older version of macOS around so you can keep interacting with your time Capsule. It still feels like a bit of a shitty move though, as those fifth generation models came with up to 3TB of storage, which can still serve as a solid NAS solution.

Thank your lucky stars, then, that open source can, as usual, come to the rescue when proprietary software vendors do what they always do and screw over their customers. Did you know every generation of Time Capsule actually runs NetBSD, and that it's trivially easy to add support for Samba 4 and SMB3 authentication to your Time Capsule, thereby extending its life expectancy considerably? TimeCapsuleSMB does exactly that.

If the setup completes successfully, your Time Capsule will run its own Samba 4 server, advertise itself over Bonjour (show up automatically in the "Network" folder on macOS), and accept authenticated SMB3 connections from macOS. You should then be able to open Finder, choose Connect to Server, and use a normal SMB URL instead of relying on Apple's legacy stack. You should also be able to use the disk for Time Machine backups.
        ↫ TimeCapsuleSMB

It's compatible with both NetBSD 4 and NetBSD 6-based Time Capsules, although you'll need to run a single SMB activation command every time a NetBSD 4-based Time Capsule reboots. This will also disable any AFP and SMB1 support, but that is kind of moot since those are exactly the technologies that don't and won't work anymore once macOS 27 is released. The installation is also entirely reversible if, for whatever reason, you want to undo the addition of Samba 4.

This whole saga is such an excellent example of why open source software protects users' rights, by design.

Original Submission

An Anonymous Coward writes:

The classified deal apparently doesn't allow Google to veto how the government will use its AI models:

Google has signed a classified deal that allows the US Department of Defense to use its AI models for "any lawful government purpose," The Information reports. The agreement was reported less than a day after Google employees demanded CEO Sundar Pichai block the Pentagon from using its AI amid concerns that it would be used in "inhumane or extremely harmful ways."

If the agreement is confirmed, it would place Google alongside OpenAI and xAI, which have also made classified AI deals with the US government. Anthropic was also among that list until it was blacklisted by the Pentagon for refusing the Department of Defense's demands to remove weapon and surveillance-related guardrails from its AI models.

Citing a single anonymous source "with knowledge of the situation," The Information reports that the deal states that both parties have agreed that the search giant's AI systems shouldn't be used for domestic mass surveillance or autonomous weapons "without appropriate human oversight and control." But the contract also says it doesn't give Google "any right to control or veto lawful government operational decision-making," which would suggest the agreed restrictions are more of a pinky promise than legally binding obligations. The deal also requires Google to assist with making adjustments to its AI safety settings and filters at the government's request.

"We are proud to be part of a broad consortium of leading AI labs and technology and cloud companies providing AI services and infrastructure in support of national security," a Google spokesperson said in a statement to The Information, adding that the new agreement is an amendment to its existing government deal. "We remain committed to the private and public sector consensus that AI should not be used for domestic mass surveillance or autonomous weaponry without appropriate human oversight."

Original Submission

owl writes:

https://read.thecoder.cafe/p/linux-broke-postgresql

On April 3, 2026, Salvatore Dipietro, an engineer at AWS, posted a patch to the Linux kernel mailing list. The reason: on a 96-vCPU Graviton4 machine running Linux 7.0, PostgreSQL throughput had dropped to roughly half of what it produced on Linux 6.x. In this post, we will trace what changed in Linux 7.0, how PostgreSQL manages memory, and what role memory pages play in making the problem appear (or disappear). Get cozy, grab a coffee, and let's begin!

The Problem

Salvatore Dipietro ran pgbench (PostgreSQL's standard benchmarking tool) on a Graviton4 processor with 96 vCPUs. The workload was a benchmark doing simple updates at scale factor 8,470 (i.e., roughly a 847 million row table), simulating 1,024 clients and 96 threads. A serious, high-parallelism load designed to stress the system.

The results were striking. Linux 7.0 delivered roughly half the throughput of Linux 6.x on the same hardware and workload:

Linux 6.x: 98,565 transactions per second

Linux 7.0: 50,751 transactions per second

Original Submission

TheDeadDontDie writes:

Colorado has led the US on legislation that ensures people can fix their stuff. Manufacturers tried to claw back that control but ultimately failed—for now:

A controversial bill in Colorado that would have undone some repair protections in the state has failed. The bill had been the target of right-to-repair advocates, who saw it as a bellwether for how tech companies might try to undo repair legislation more broadly in the US.

Colorado's landmark 2024 repair law, the Consumer Right to Repair Digital Electronic Equipment , went into effect in January 2026 and ensured access to tools and documentation people needed to modify and fix digital electronics such as phones, computers, and Wi-Fi routers. The new bill, SB26-090 , would have carved out an exception to those repair protections for "critical infrastructure," a loosely defined term that repair advocates worried could be applied to just about any technology.

SB26-090 was introduced during a Colorado Senate hearing on April 2 and was supported by lobbying efforts from companies such as Cisco and IBM. It passed that hearing unanimously. The bill then passed in the Colorado Senate on April 16. On Monday evening, the bill was discussed in a long, delayed hearing in the Colorado House's State, Civic, Military, and Veterans Affairs Committee. Dozens of supporters and detractors gave public comments. Finally, the bill was shot down in a 7-to-4 vote and classified as postponed indefinitely.

Danny Katz, executive director of the local nonprofit consumer advocacy group CoPIRG, says the battle was a group effort. Speaking against the bill were a cohort of repair advocates from organizations such as PIRG , Repair.org , iFixit , Consumer Reports , and local businesses and environmental groups like Blue Star Recyclers , Recycle Colorado , Environment Colorado , and GreenLatinos .

[...] Supporters of the bill, backed by companies like Cisco, had pointed to the potential for cybersecurity risks as their motivation for altering the law's language. If companies were required to make repair tools available to anyone, the theory goes, what's to stop bad actors from using those tools to reverse engineer critical technology like Internet routers? Withholding those tools, they posited, would make them less available to hackers who could misuse them. Advocates of the bill said that companies should be allowed to keep their secrets if it ensured security, though that argument starts to fall apart with a little scrutiny.

AnonTechie writes:

An interesting essay about the issues with vibe coding ...

A marketing manager with no engineering background opens Cursor on Monday morning. By Wednesday afternoon, she has a working customer-facing app. It looks polished. It performs the core task. She demos it to her VP, who forwards it to their CMO, who then shows it in the executive staff meeting as evidence that the team is "moving at AI speed."

By Friday, it is in front of customers.

No one asked who owned the decision to ship it. No one tested it against the conditions it would actually face. No one had the cultural standing to say this looks great, and we are not putting it into production. The prototype became a product because the organization had no system for telling the difference.

I watched a version of this scenario play out recently in a boardroom. A senior executive demoed an AI-built internal tool. The room admired the speed. What received less attention were the harder questions: Who would own it after launch? Who would maintain it? And what would happen when it produced an answer that was confidently wrong?

This is what vibe coding is about to expose across businesses. The companies that think the story is about software are going to lose to the companies that understand the story is about judgment.

The Real Trend Is Decision Compression

Andrej Karpathy coined the term "vibe coding" in early 2025 to describe an AI-assisted style of building software through natural-language prompting, often without close inspection of the underlying code. Google Cloud describes vibe coding as a software development practice that makes app building more accessible, especially for people with limited programming experience. Tools like Cursor, Replit, Lovable, Bolt, GitHub Copilot Workspace, v0 by Vercel and Claude Code have moved the practice from novelty to workplace reality with stunning speed.

All of that is true. None of it is the point.

The point is that vibe coding collapses the distance between idea and artifact from months to hours. When that distance collapses, every quality-control mechanism your organization developed over the last 30 years gets bypassed by default. Design review. Security review. Legal review. Brand review. The simple friction of having to convince an engineer your idea was worth building. That is a governance story, not a software story. It is happening at every level of the org chart simultaneously.

[Source]: Forbes

Original Submission

JoeMerchant writes:

Google has a price for you. Proton found it. The company analyzed over 54,000 demographic profiles using 2025 ad auction data to see what advertisers pay to reach different Americans. The average American generates about $1,605 a year in advertising value. The median is $760. The gap between those two numbers tells the story. A small number of high-value users pull the average up. The business runs on outliers.

The spread is stark. A 35- to 44-year-old man in Bozeman, Montana — no children, desktop user, making high-value corporate searches — is worth an estimated $17,929 per year. An 18- to 24-year-old father in Fort Smith, Arkansas — Android phone, low-value searches — is worth $31.05. That is a 577x difference between two people using the same free service. Device matters. A desktop user is worth 4.9 times more than the same person on Android. An iPhone user is worth 2.7 times more than Android. Having children costs you roughly 17% of your ad value. Advertiser value peaks between ages 35 and 44. By 65, average value drops to $511.

Where you live sets a floor on your price. Local service providers — lawyers, real estate agents, financial planners — bid against each other for local clicks. The more competitive the local market, the higher the floor price for everyone in it. The top markets are Edmond, Oklahoma and Bozeman, Montana, followed by Naperville, Illinois, Santa Fe, New Mexico, and Durham, North Carolina. The least valuable markets are concentrated in the Rust Belt and Appalachia — Wheeling and Parkersburg in West Virginia, Toledo, Ohio, and Buffalo, New York — where lower median incomes and fewer competing advertisers mean less bidding pressure. Over a decade, the average American represents roughly $16,050 in ad value. The most monetized profiles approach $180,000. Most people would not hand a corporation that much money over a lifetime. But that is what the system collects.

-----

Google, while big, is only one internet advertiser - and all that collected advertising income actually comes from consumers of the goods and services being advertised, as a premium on the price of the products. One particular medical device I worked on cost $600 to make, and $14,400 to sell at a net price to the patient of $15,000 for the device and another $15,000 to the hospital for the implantation procedure. Yes, the company was operating at break-even, spending 24x what the physical device cost to make and deliver on nothing but sales and marketing - hoping that some day they could get those sales costs down... didn't happen during the 2 years I worked there.

Original Submission

fliptop writes:

Microsoft, long a symbol of American innovation, is now offering a voluntary early retirement program that targets thousands of its most seasoned U.S. employees. Framed as a generous opportunity for longtime workers, the move instead reveals a deeper corporate calculus: trimming payroll of experienced Americans to redirect resources toward artificial intelligence infrastructure and, likely, a younger, often less expensive workforce:

This is not mere cost-cutting in response to market pressures. It is a strategic thinning of the ranks amid hundreds of billions committed to AI development, at a time when the company has already shed thousands of jobs in recent years. By dangling buyouts before employees whose age plus years of service equal 70 or more—primarily those at senior director level and below—Microsoft aims to reduce its 125,000-strong U.S. workforce by up to 7 percent, or roughly 8,750 people, without the public backlash of outright layoffs.

The program, announced in an internal memo from Chief People Officer Amy Coleman, marks the first such voluntary retirement initiative in the company's 51-year history. Eligible workers will receive notification beginning May 7 and have 30 days to decide. While presented as support for those "considering their next chapter," the timing aligns precisely with Microsoft's voracious appetite for AI spending, projected near $100 billion in capital expenditures this year alone.

[...] Recent history underscores the trend. Microsoft has conducted multiple rounds of job cuts, even as it competes fiercely with Google and others in the AI race. Similar moves at Meta, which recently slashed 10 percent of its workforce to fund infrastructure, reveal an industry-wide willingness to sacrifice people for processors. The human element—wisdom forged through years of problem-solving—receives polite acknowledgment before being shown the door with a severance package and extended healthcare.

Previously: Tech Industry Lays Off Nearly 80,000 Employees in the First Quarter of 2026 (Almost 50% Due to AI)

Original Submission

anubi writes:

The ability to Side load Android may be going away

I just ran across this while bringing up another Android phone:

It is linked from the F-Droid website:

125 days until lockdown.

Starting September 2026, a silent update, nonconsensually pushed by Google, will block every Android app whose developer hasn't registered with Google, signed their contract, paid up, and handed over government ID.

Every app and every device, worldwide, with no opt-out.

( I have an interest in developing an Android apk for using cellphones as an HMI for Arduinos. )

Also, An Anonymous Coward writes:

Starting September 2026, a silent update, nonconsensually pushed by Google, will block every Android app whose developer hasn't registered with Google, signed their contract, paid up, and handed over government ID.

Every app and every device, worldwide, with no opt-out.

In August 2025, Google announced a new requirement: starting September 2026, every Android app developer must register centrally with Google before their software can be installed on any device. Not just Play Store apps: all apps. This includes apps shared between friends, distributed through F-Droid, built by hobbyists for personal use. Independent developers, church and community groups, and hobbyists alike will all be frozen out of being able to develop and distribute their software.

Registration requires:

• Paying a fee to Google
• Agreeing to Google's Terms and Conditions
• Surrendering your government-issued identification
• Providing evidence of your private signing key
• Listing all current and all future application identifiers

If a developer does not comply, their apps get silently blocked on every Android device worldwide.

Continued here.

Original Submission #1  Original Submission #2 

anubi writes:

I thought you guys might like this ..

Somebody has some 'splainin' to do!

https://www.tomshardware.com/tech-industry/artificial-intelligence/claude-powered-ai-coding-agent-deletes-entire-company-database-in-9-seconds-backups-zapped-after-cursor-tool-powered-by-anthropics-claude-goes-rogue

The founder of PocketOS has penned a social media post to warn others about the "systemic failures" of flagship AI and digital services providers. Jer Crane was inspired to write a public response after an AI coding agent deleted his firm's entire production database. The AI agent's misdemeanors were then hugely amplified by a cloud infrastructure provider's API wiping all backups after the main database was zapped. This tag team of digital trouble has wiped out months of consumer data essential to the firm's, and its customers, businesses.

[...] "Yesterday afternoon, an AI coding agent — Cursor running Anthropic's flagship Claude Opus 4.6 — deleted our production database and all volume-level backups in a single API call to Railway, our infrastructure provider," sums up the PocketOS boss. "It took 9 seconds."

[...] The PocketOS boss puts greater blame on Railway's architecture than on the deranged AI agent for the database's irretrievable destruction. Briefly, the cloud provider's API allows for destructive action without confirmation, it stores backups on the same volume as the source data, and "wiping a volume deletes all backups." Crane also points out that CLI tokens have blanket permissions across environments.

It was also observed by the irate SaaS founder that Railway is actively promoting the use of AI-coding agents by its customers. Crane's use of an AI coding agent on the Railway platform wasn't exploring new frontiers, or wasn't supposed to be. Meanwhile, Crane has been provided no recovery solution, and Railway has apparently been hedging carefully regarding any such possibility.

[...] Thankfully, PocketOS had a full 3-month-old backup, which was restorable from, so the deletion gaps are all limited to the interim period.

There are lessons to be learned from mistakes, as usual. Crane bullet points five things that need to change as the AI industry scales faster than it builds a worthwhile safety architecture. Specifics he calls for include; stricter confirmations, scopable API tokens, proper backups, simple recovery procedures, and AI agents existing within proper guardrails.

In the meantime, please follow a thorough backup regimen and be careful out there. This isn't the first time we've seen an AI go rogue and start deleting important databases.

HowdyDoody writes:

University of Oregon scientists repurposed battery-testing tool to better measure coffee's flavor profile:

University of Oregon chemist Christopher Hendon loves his coffee—so much so that studying all the factors that go into creating the perfect cuppa constitutes a significant area of research for him. His latest project: discovering a novel means of measuring the flavor profile of coffee simply by sending an electrical current through a sample beverage. The results appear in a new paper published in the journal Nature Communications.

We've been following Hendon's work for several years now. For instance, in 2020, Hendon's lab helped devise a mathematical model for brewing the perfect cup of espresso, over and over, while minimizing waste. The flavors in espresso derive from roughly 2,000 different compounds that are extracted from the coffee grounds during brewing. So it can be challenging for baristas to reproduce the same perfect cup over and over again.

That's why Hendon and his colleagues built their model for a more easily measurable property known as the extraction yield (EY): the fraction of coffee that dissolves into the final beverage. That, in turn, depends on controlling water flow and pressure as the liquid percolates through the coffee grounds. The model is based on how lithium ions propagate through a battery's electrodes, similar to how caffeine molecules dissolve from coffee grounds.

[...] There are existing methods for collecting information on coffee's chemical composition, most notably liquid or gas chromatography combined with mass spectrometry. But these kinds of analyses are expensive and time-consuming, and predictive results are limited. There are also electrochemical techniques for measuring the concentration of caffeine and other molecules, but these have not taken into account coffee strength—a property determined by all the variables that go into preparing a cup of coffee, such as coffee and water masses, grind settings, water temperature and pressure, roast color, and so forth. That's the information likely to be most helpful to baristas.

Arthur T Knackerbracket writes:

https://mashable.com/article/nasa-nancy-grace-roman-space-telescope-explained

About a quarter-century after the Hubble Telescope reshaped astronomy, and a few years into the era of the James Webb Space Telescope, NASA's Nancy Grace Roman Space Telescope will join them not as a replacement, but as a big-picture partner. Where Hubble and Webb zoom in for close‑ups, Roman will capture Hubble‑like detail across areas about 100 times larger, turning isolated snapshots into sweeping surveys that show the very scaffolding of the universe.

At NASA's Goddard Space Flight Center in Greenbelt, Maryland, engineers are wrapping up prelaunch testing on the cutting-edge telescope. Next, the observatory will travel 900 miles to Kennedy Space Center in Cape Canaveral, Florida, where teams will prepare it for launch. 

That could happen as early as this September, about eight months ahead of schedule, NASA managers said at a news conference on Tuesday, April 21. Once in space, Roman will head to a stable orbit about 1 million miles from Earth, near the same region where Webb orbits the sun, and begin a years‑long campaign of deep space imaging. 

"We didn't want to wait to launch the Nancy Grace Roman. We're eight months ahead of schedule," said Nicky Fox, NASA's associate administrator of science. "Everybody felt the urgency. Everybody was sprinting towards this."

Named for Nancy Grace Roman, who became the agency's first chief of astronomy and one of its earliest female executives, the telescope reflects a legacy of opening new windows on the universe from above Earth's atmosphere. Nicknamed the "mother of Hubble," Roman helped lay the groundwork in the 1960s for a whole fleet of space telescopes.

At the heart of the mission is Roman's eight-foot-wide mirror, the same size as Hubble's, paired with a powerful camera that sees in infrared light, like Webb. That camera's field of view is Roman's superpower. In a single shot, it can image vast swaths of sky that Hubble simply can't match. 

Because a space telescope can only see one patch of sky at a time, it has to take many separate "pointings" — individual shots aimed at slightly different spots — and stitch them together into a mosaic.

In 2023, Ami Choi, an astrophysicist and scientist for Roman's wide field camera, contrasted the difference between Hubble and the new telescope. To photograph the Andromeda Galaxy, Hubble has to take 400 smaller images and stitch them together. For Roman's camera, that should only take two pointings, she said. 

This wide, sharp vision is what scientists need to study the so-called "dark universe." Ordinary matter — the stuff that makes up stars, planets, and even people — accounts for only about 5 percent of the cosmos. The bulk of it is dark matter and dark energy, which do not emit light but leave clues where they've influenced space's expansion and the arrangement of galaxies.

"Current observations hint that our standard model of the universe is incorrect," said Julie McHenry, senior project scientist, referring to cosmologists' best recipe for the universe. "Roman will be able to confirm these and set us on the path to understanding what's right."

Roman will trace those clues in several ways at once. By mapping the positions and shapes of hundreds of millions of galaxies, it will show how structures have grown from the early universe to today. Subtle distortions in galaxy shapes will reveal how clumps of invisible space stuff bend their light on the way to us, exposing the hidden dark matter. At the same time, Roman will discover and track large numbers of a special kind of exploding star, known as Type Ia supernovas; their predictable brightness lets astronomers measure how quickly space has expanded over time.

Taken together, these measurements will allow scientists to test competing ideas about dark matter, dark energy, and even the laws of gravity themselves with far greater precision than ever before. Other observatories can make similar kinds of measurements, but none combines Roman's sharpness and sky coverage in the infrared, NASA mission leaders say, which lets it see more distant and dust-covered galaxies.

An Anonymous Coward writes:

https://www.phoronix.com/news/MS-Azure-Linux-Fedora-Based

Microsoft's in-house Azure Linux operating system used within Azure and for WSL and other purposes is reportedly pursuing an overhaul where it would be derived from Fedora Linux.

Azure Linux -- originally known as CBL-Mariner -- is already an RPM-based Linux distribution that is catering to the various Linux needs at Microsoft. It's scope and capabilities of Azure Linux have grown a lot the past few years and now it may evolve into being derived from Fedora.

With the recent proposal to build x86_64-v3 packages for Fedora 45, it turns out Microsoft is involved in backing this change proposal. Kyle Gospodnetich as one of the change proposal authors for x86_64-v3 packages with Fedora 45 happens to be a Microsoft Linux engineer.

The connection to Microsoft's stake in x86_64-v3 and Azure Linux looking at Fedora as a base was spelled out clearly this week during Fedora's Enterprise Linux Next (ELN) SIG meeting. It was noted that Microsoft as well as Fyra Labs are very interested in x86_64-v3 for Fedora.

In the meeting log it's explicitly laid out:

• "and since Microsoft is supporting that change, they probably would be able to donate compute resources"

• "Azure wants to rebase Azure Linux more or less on Fedora and they need x86_64-v3 for performance"

• "there was some nebulous plans of forking the whole distribution for this, they were guided in this direction...so I'd rather it not fail for that reason"

Very interesting and will be fascinating to see what other changes could happen with a Fedora-based Microsoft Azure Linux distribution. In any case, also great to see Microsoft pushing the x86_64-v3 micro-architecture feature level.

Original Submission

Arthur T Knackerbracket writes:

https://gizmodo.com/chinas-biggest-streaming-platform-wants-most-of-its-new-films-to-be-ai-generated-2000748454

A company dubbed China’s Netflix expects a near-complete AI takeover of film and TV within the next five years.

The streaming platform IQiyi plans to have AI create most of its new films and TV shows, per Bloomberg. CEO Gong Yu reportedly shared this at an annual content showcase, alongside an AI toolkit called Nadou Pro that can supposedly automate every step of filmmaking from scriptwriting to final rendering, with the help of AI models from Alibaba and ByteDance for its domestic version and Google Veo 3.1 for an international version.

The company’s goal is to use Nadou Pro to release a fully AI-generated movie that they hope will reach commercial success as early as this summer. IQiyi’s debut slate currently includes 16 AI-generated sci-fi and anime movies, Bloomberg reported.

Over the past year, AI-generated video content has seeped into every corner of the internet. From eerily realistic animal videos that have viewers question their sanity to viral TikToks on the messy love lives of talking fruits, short-form AI video slop is undeniably popular on the internet. But that popularity has yet to translate into any fully AI-generated, commercially successful, and engaging long-form content like movies and TV shows.

Nevertheless, the corporate world is taking notice. Earlier this year, Roku founder and CEO Anthony Wood predicted that “the first 100% AI-generated hit movie” would be released sometime within the next three years.

On the road to achieving that objective, Hollywood started spending big bucks on AI. YouTube introduced AI tools for content creation last September. Last summer, Netflix announced that it had officially begun using AI-generated final footage in shows, the first example that we know of being in the Argentine sci-fi show “El Eternauta.” Around the same time, Amazon MGM Studios launched an in-house team dedicated to building AI tools for film and TV production, and those tools have now reportedly launched in a closed beta program.