Google has delisted over 749 million URLs from Anna's Archive, a shadow library and meta-search engine for pirated books, representing 5% of all copyright takedown requests ever filed with the company. TorrentFreak reports: Google's transparency report reveals that rightsholders asked Google to remove 784 million URLs, divided over the three main Anna's Archive domains. A small number were rejected, mainly because Google didn't index the reported links, resulting in 749 million confirmed removals. The comparison to sites such as The Pirate Bay isn't fair, as Anna's Archive has many more pages in its archive and uses multiple country-specific subdomains. This means that there's simply more content to take down. That said, in terms of takedown activity, the site's three domain names clearly dwarf all pirate competition.

Since Google published its first transparency report in May 2012, rightsholders have flagged 15.1 billion allegedly infringing URLs. That's a staggering number, but the fact that 5% of the total targeted Anna's Archive URLs is remarkable. Penguin Random House and John Wiley & Sons are the most active publishers targeting the site, but they are certainly not alone. According to Google data, more than 1,000 authors or publishers have sent DMCA notices targeting Anna's Archive domains. Yet, there appears to be no end in sight. Rightsholders are reporting roughly 10 million new URLs per week for the popular piracy library, so there is no shortage of content to report.

A massive cyberattack on Swedish IT supplier Miljodata exposed personal data from up to 1.5 million citizens, prompting a national privacy investigation and scrutiny into security failures across multiple municipalities. BleepingComputer reports: MiljÃdata is an IT systems supplier for roughly 80% of Sweden's municipalities. The company disclosed the incident on August 25, saying that the attackers stole data and demanded 1.5 Bitcoin to not leak it. The attack caused operational disruptions that affected citizens in multiple regions in the country, including Halland, Gotland, Skelleftea, Kalmar, Karlstad, and Monsteras.

Because of the large impact, the state monitored the situation from the time of disclosure, with CERT-SE and the police starting to investigate immediately. According to IMY, the attacker exposed on the dark web data that corresponds to 1.5 million people in the country, creating the basis for investigating potential General Data Protection Regulation (GDPR) violations. [...] Although no ransomware groups had claimed the attack when Miljodata disclosed the incident, BleepingComputer found that the threat group Datacarry posted the stolen data on its dark web portal on September 13. The leaked database has been added to Have I Been Pwned, which contains information such as names, email addresses, physical addresses, phone numbers, government IDs, and dates of birth.

Apple will officially discontinue support for its original Home architecture (formerly HomeKit) on February 10, 2026. As MacRumors points out, Apple has informed users that they need to "update now to avoid interruptions." AppleInsider reports: The underlying HomeKit architecture was revamped in March 2023 alongside iOS 16.4, so Apple has been supporting both the new and old architecture for the last two years. There were initial problems with stability that may have discouraged some users from upgrading, but those problems have now been addressed.

When Apple stops supporting the original HomeKit architecture, it will break support for the Home app on devices running older versions of iOS, iPadOS, and macOS. iOS 16.2, iPadOS 16.2, macOS 13.1, tvOS 16.2, and watchOS 9.2 are the minimum versions of Apple's platforms that work with the updated Apple Home app, and older devices will lose access.

The update includes support for features like guest access, support for robot vacuum cleaners, and Activity History. Apple says it also provides faster, more reliable performance, especially for smart homes with a lot of HomeKit and Matter accessories installed.

Google's new "Project Suncatcher" aims to launch Tensor Processing Units (TPUs) into space, creating a solar-powered, satellite-based AI network capable of scaling machine learning beyond Earth's limits. Google says a "solar panel can be up to 8 times more productive than on earth" for near-continuous power using a "dawn-dusk sun-synchronous low earth orbit" that reduces the need for batteries and other power generation. 9to5Google reports: These satellites would connect via free-space optical links, with large-scale ML workloads "distributing tasks across numerous accelerators with high-bandwidth, low-latency connections." To match data centers on Earth, the connection between satellites would have to be tens of terabits per second, and they'd have to fly in "very close formation (kilometers or less)."

Google has already conducted radiation testing on TPUs (Trillium, v6e), with "promising" results: "While the High Bandwidth Memory (HBM) subsystems were the most sensitive component, they only began showing irregularities after a cumulative dose of 2 krad(Si) -- nearly three times the expected (shielded) five year mission dose of 750 rad(Si). No hard failures were attributable to TID up to the maximum tested dose of 15 krad(Si) on a single chip, indicating that Trillium TPUs are surprisingly radiation-hard for space applications."

Finally, Google believes that launch costs will "fall to less than $200/kg by the mid-2030s." At that point, the "cost of launching and operating a space-based data center could become roughly comparable to the reported energy costs of an equivalent terrestrial data center on a per-kilowatt/year basis."

IBM will cut thousands of roles this quarter while it continues to shift focus to higher-growth software and services, Bloomberg News reported on Tuesday. From a report: "We routinely review our workforce through this lens and at times rebalance accordingly," Bloomberg quoted a company spokesperson saying. "In the fourth quarter we are executing an action that will impact a low single-digit percentage of our global workforce."

Longtime Slashdot reader sinij shares a report from Car and Driver: Sales of electric vehicles surged in September as shoppers rushed to take advantage of the $7500 federal EV tax credit before it disappeared at the end of the month. With the government subsidies now gone, EV sales were expected to take a hit in October. While only a few automakers still report sales on a monthly basis, the results we do have do not paint a rosy picture for EVs in a post-tax credit world.

The Korean automakers were hit particularly hard by the loss of the tax credit. The Hyundai Ioniq 5, which was the fifth-best-selling EV through the third quarter of this year, experienced a 63 percent drop, moving 1642 units in October 2025, down from 4498 in 2024. Its platform-mates saw similar declines. The Kia EV6 moved just 508 units, down 71 percent versus the same month the year before, while the luxurious Genesis GV60 only found 93 buyers, a 54 percent slide year over year. Things were even worse at Honda. While the Acura ZDX was recently discontinued after just a single model year, the related Honda Prologue remains on sale but registered just 806 units, down 81 percent from 4130 sales in October 2024. [...]

Obviously, this isn't the full picture, as several major players -- including General Motors, Toyota, Nissan, and Volkswagen -- only release sales reports on a quarterly basis, and others, such as Tesla and Rivian, don't break out individual sales at all. But with four of the top 10 bestselling EVs through Q3 all showing noteworthy declines in October, it spells trouble for the EV market at large. The end-of-year sales figures will provide a much clearer picture of whether October was just a blip or the start of a much more widespread problem for EV sales.

Australia's new "solar sharer" program will give households in NSW, south-east Queensland, and South Australia at least three hours of free solar power each day starting in 2026 -- even for those without rooftop panels. Other areas will potentially follow in 2027. The Guardian reports: The government said Australians could schedule appliances such as washing machines, dishwashers and air conditioners and charge electric vehicles and household batteries during this time. The solar sharer scheme would be implemented through a change to the default market offer that sets the maximum price retailers can charge customers for electricity in parts of the country. The climate change and energy minister, Chris Bowen, said the program would ensure "every last ray of sunshine was powering our homes" instead of some solar energy being wasted.

Australians have installed more than 4m solar systems and there is regularly cheap excess generation in the middle of the day. Part of the rationale for the program is that it could shift demand for electricity from peak times -- particularly early in the evening -- to when it is sunniest. This could help minimize peak electricity prices and reduce the need for network upgrades and intervention to ensure the power grid was stable.

In its largest rollout yet, Waymo said it will launch its driverless robotaxi service to Las Vegas, San Diego, and Detroit in 2026. The Alphabet unit will also debut new Zeekr-built vehicles developed with Geely to complement its existing Jaguar I-PACE fleet. Reuters reports: The new Zeekr model, developed with Chinese automaker Geely, are designed specifically for robotaxi use cases and will be rolled out gradually as the company expands its service. [...] Waymo plans to launch the service in Las Vegas next summer, while in San Diego, it is working with local officials and first responders to secure deployment permits. In Detroit, the company said its winter-weather testing in Michigan's Upper Peninsula has strengthened its ability to operate year-round, where it has long maintained engineering operations.

Coca-Cola has doubled down on AI-generated holiday ads despite widespread criticism of last year's uncanny results. This year the beverage company is replacing human actors with oddly animated animals in a visually inconsistent campaign. The Verge reports: There's no consistent style, switching between attempted realism and a bug-eyed toony look, and the polar bears, panda, and sloth move unnaturally, like flat images that have been sloppily animated rather than rigged 3D models in CG. Compared to the convincing deepfake videos being generated by tools like OpenAI's Sora 2 or Google's Veo 3, the videos produced for this Coke ad feel extremely dated.

The only notable improvement to my eyes is that the wheels on the iconic Coke trucks are actually consistently turning this year, rather than gliding statically over snow-covered roads. The Wall Street Journal reports that Coca-Cola teamed up with Silverside and Secret Level on its latest holiday campaign, two of the AI studios that previously worked on the 2024 Coke Christmas ads.

Coca-Cola declined to comment on the cost of the new holiday campaign, according to The Wall Street Journal, but said that around 100 people were involved in the project -- a figure comparable to the company's older AI-free productions. That includes five "AI specialists" from Silverside who contributed by prompting and refining more than 70,000 AI video clips.

"Rockstar Games fired dozens of employees," reports Bloomberg, "in a move that a British trade union said was designed to prevent the workers from unionizing. The company said they were fired for misconduct." TheGrand Theft Automaker terminatedbetween 30 and 40 staffersacross multipleoffices in the UK and Canada on Thursday, according to aspokesperson for the Independent Workers' Union of Great Britain (IWGB). All of the employees were part of a private trade union chat groupon Discord and were either members of the union or attempting to organize at the company, the union spokesperson said.

"Rockstar has just carried out one of the most blatant and ruthless acts of union busting in the history of the games industry," Alex Marshall, president of theIWGB, said in a statement. "This flagrant contempt for the law and for the lives of the workers who bring in their billions is an insult to their fans and the global industry."
On BlueSky the IWGB union posted "We won't back down, and we're not scared — we will fight for every member to be reinstated."

Bloomberg notes that Grand Theft Auto VIis slated for release on May 26, 2026, "and is expected to be one of the top-selling video games of all time."

TechCrunch reports: OpenAI CEO Sam Altman recently said that the company is doing "well more" than $13 billion in annual revenue — and he sounded a little testy when pressed on how it will pay for its massive spending commitments.

His comments came up during a joint interview on the Bg2 podcast between Altman and Microsoft CEO Satya Nadella about the partnership between their companies. Host Brad Gerstner (who's also founder and CEO of Altimeter Capital) brought upreports that OpenAI is currently bringing in around $13 billion in revenue — a sizable amount, but one that's dwarfed by more than $1 trillion in spending commitments for computing infrastructure that OpenAI has made for the next decade.

"First of all, we're doing well more revenue than that. Second of all, Brad, if you want to sell your shares, I'll find you a buyer," Altman said, prompting laughs from Nadella. "I just — enough. I think there are a lot of people who would love to buy OpenAI shares."
Altman's answer continued, making the case for OpenAI's business model. "We do plan for revenue to grow steeply. Revenue is growing steeply. We are taking a forward bet that it's going to continue to grow and that not only will ChatGPT keep growing, but we will be able to become one of the important AI clouds, that our consumer device business will be a significant and important thing. That AI that can automate science will create huge value...

"We carefully plan, we understand where the technology — where the capability — is going to go, and the products we can build around that and the revenue we can generate. We might screw it up — like, this is the bet that we're making, and we're taking a risk along with that." (That bet-with-risks seems to be the $1.4 trillion in spending commitments — but Altman suggests it's offset by another absolutely certain risk: "If we don't have the compute, we will not be able to generate the revenue or make the models at this kind of scale.")

Satya Nadella, Microsoft's CEO, added his own defense, "as both a partner and an investor. There has not been a single business plan that I've seen from OpenAI that they have put in and not beaten it. So in some sense, this is the one place where in terms of their growth — and just even the business — it's been unbelievable execution, quite frankly..."

"Amazon has deployed Rivian's electric delivery vans in Canada for the first time," reports CleanTechnica, with 50 now deployed in the Vancouver area.

Amazon's director of Global Fleet and Products says there's now over 35,000 electric vans deployed globally — and that they've delivered more than 1.5 billion packages.

More from the blog Teslarati: In December 2024, the companies announced they had successfully deployed 20,000 EDVs across the U.S. In the first half of this year, 10,000 additional vans were delivered, and Amazon's fleet had grown to 30,000 EDVs by mid-2025. Amazon's fleet of EDVs continues to grow rapidly and has expanded to over 100 cities in the United States... The EDV is a model that is exclusive to Amazon, but Rivian sells the RCV, or Rivian Commercial Van, openly. It detailed some of the pricing and trim options back in January when it confirmed it had secured orders from various companies, including AT&T.

"People are creating 'dumb homes,'" the VP of research at the Global Wellness Institute, tells the web site Axios.

Some are swapping NASA-style setups for old-fashioned buttons, switches and knobs. Others are designing digital detox corners — all part of a bigger "analog wellness" movement...

The return to analog hobbies and spacesis about more than nostalgia for pre-internet times, researchers say. A home where "technology is always in the background, working and listening, feels anxiety-producing" instead of restorative, architect Yan M. Wang tells Axios... Design media brand Dwell named the decline of smart homes a top trend for 2025 and beyond.

Wealthy Los Angeles house hunters have started shunning WiFi-enabled, voice-activated appliances "to escape the $100 billion home-automation industry," according to the Hollywood Reporter. Meanwhile, landlines have found new fans — many of them parents who want to keep their kids off screens, the Washington Post reports.

CBS News investigates what happened when police thought they'd tracked down a "porch pirate" who'd stolen a package — and accused an innocent woman.

"You know why I'm here," the police sergeant tells Chrisanna Elser. "You know we have cameras in that town..." "It went right into, 'we have video of you stealing a package,'" Elser said... "Can I see the video?" Elser asked. "If you go to court, you can," the officer replied. "If you're going to deny it, I'm not going to extend you any courtesy...." [You can watch a video of the entire confrontation.] On her doorstep, the officer issued a summons, without ever looking at the surveillance video Elser had. "We can show you exactly where we were," she told him. "I already know where you were," he replied.

Her Rivian — equipped with multiple cameras — had recorded her entire route that day... It took weeks of her collecting her own evidence, building timelines, and submitting videos before someone listened. Finally, she received an email from the Columbine Valley police chief acknowledging her efforts in an email saying, "nicely done btw (by the way)," and informing her the summons would not be filed.
Elser also found the theft video (which the police officer refused to show her) on Nextdoor, reports Electrek. "The woman has the same color hair, but different facial and nose shape and apparent age than Elser, which is all reasonably apparent when viewing the video..."

But Elser does drive a green Rivian truck, which police knew had entered the neighborhood 20 times over the course of a month. (Though in the video the officer is told that a male driver in the same household passes through that neighborhood driving to and from work.) The problem may be their certainty — derived from Flock's network of cameras that automatically read license plates, "tracking movements of vehicles wherever they go..." The system has provoked concern from privacy and freedom focused organizations like the Electronic Frontier Foundation and American Civil Liberties Union. Flock also recently announced a partnership with Ring, seeking to use a network of doorbell cameras to track Americans in even more places.... [The police] didn't even have video of the truck in the area — merely tags of it entering... (it also left the area minutes later, indicating a drive through, rather than crawling through neighborhoods looking for packages — but police neglected to check the exit timestamps)... Elser has asked for an apology for [officer] Milliman's aggressive behavior during the encounter, but has heard nothing back from the department despite a call, email, and physical appearance at the police station.
The article points out that Rivian's "Road Cam" feature can be set to record footage of everything happening around it using the car's built in cameras for driver-assist features. But if you want to record footage all the time, you'll need to plug in a USB-C external drive to store it. (It's ironic how different cameras recorded every part of this story — the theft, the police officer accusing the innocent woman, and that innocent woman's actual whereabouts.)

Electrek's take? "Citizens should not need to own a $70k+ truck, or even a $100 external hard drive, to keep track of everything they do in order to prove to power-tripping officers that they didn't commit a crime."

"A nearby galaxy once thought to be dominated by dark matter seems to have a surprise supermassive black hole at its centre," reports New Scientist.

Yet scientists "are convinced dark matter is out there," writes Space.com. "The quest to detect it arguably remains both one of the most frustrating and most exhilarating challenges in modern physics."

And now they report that the century-old mystery of dark matter — the invisible glue thought to hold galaxies together — "just got a modern clue." Scientists say they may be one step closer to confirming the existence of this elusive material, thanks to new simulations suggesting that a faint glow at the center of the Milky Way could be dark matter's long-sought signature. "It's very hard to actually prove, but it does seem likely," Moorits Muru of the Leibniz Institute for Astrophysics Potsdam in Germany, who led the new study, told Space.com...

The findings, show that dark matter near the Milky Way's center might not form a perfect sphere as scientists long thought. Instead, it appears flattened, almost egg-shaped, and that shape closely mirrors the pattern of mysterious gamma rays observed by NASA's Fermi Gamma-ray Space Telescope... Using powerful supercomputers, [the researchers] recreated how the Milky Way formed, including billions of years of violent collisions and mergers with smaller galaxies. Those violent events, the researchers found, left deep "fingerprints" on the way dark matter is distributed in the galactic core.... matching the pattern of gamma-ray emission Fermi has observed, the new study reports...

If the excess truly arises from dark matter collisions, it would mark the first indirect evidence that weakly interacting massive particles [WIMPs], a leading dark matter candidate, really exist...
"We have run dozens of direct detection experiments around the globe hunting for WIMPS," notes Phys.org, in an article titled "The Empty Search for Dark Matter." We have run dozens of direct detection experiments around the globe hunting for WIMPS — dark matter particles in this particular mass range. And they're not all the same kind of experiments. There are also the scintillators, which use a giant vat of liquefied noble gas, like several tons of xenon. They wait for a dark matter particle to strike the xenon and cause it to scintillate, which is a fancy science word for "sparkle." We see the sparkle; we detect dark matter...

They're just one example of a broader class of dark matter candidates, with delightful names like Q-balls, WIMPzillas, and sterile neutrinos. We've tuned our different experiments to capture different mass ranges or interaction strengths to cover as much of that wide dark matter spectrum as possible. We've even tried to manufacture various kinds of dark matter in our particle collider experiments.

And we've found nothing.

Ubuntu "is adopting the memory-safe Rust language," reports ZDNet, citing remarks at this year's Ubuntu Summit from Jon Seager, Canonical's VP of engineering for Ubuntu: . Seager said the engineering team is focused on replacing key system components with Rust-based alternatives to enhance safety and resilience, starting with Ubuntu 25.10. He stressed that resilience and memory safety, not just performance, are the principal drivers: "It's the enhanced resilience and safety that is more easily achieved with Rust ports that are most attractive to me". This move is echoed in Ubuntu's adoption of sudo-rs, the Rust implementation of sudo, with fallback and opt-out mechanisms for users who want to use the old-school sudo command.

In addition to sudo-rs, Ubuntu 26.04 will use the Rust-based uutils/coreutils for Linux's default core utilities. This setup includes ls, cp, mv, and dozens of other basic Unix command-line tools. This Rust reimplementation aims for functional parity with GNU coreutils, with improved safety and maintainability.

On the desktop front, Ubuntu 26.04 will also bring seamless TPM-backed full disk encryption. If this approach reminds you of Windows BitLocker or MacOS FileVault, it should. That's the idea.
In other news, Canonical CEO Mark Shuttleworth said "I'm a believer in the potential of Linux to deliver a desktop that could have wider and universal appeal." (Although he also thinks "the open-source community needs to understand that building desktops for people who aren't engineers is different. We need to understand that the 'simple and just works' is also really important.")

Shuttleworth answered questions from Slashdot's readers in 2005 and 2012.

The address bar/ChatGPT input window in OpenAI's browser ChatGPT Atlas "could be targeted for prompt injection using malicious instructions disguised as links," reports SC World, citing a report from AI/agent security platform NeuralTrust: NeuralTrust found that a malformed URL could be crafted to include a prompt that is treated as plain text by the browser, passing the prompt on to the LLM. A malformation, such as an extra space after the first slash following "https:" prevents the browser from recognizing the link as a website to visit. Rather than triggering a web search, as is common when plain text is submitted to a browser's address bar, ChatGPT Atlas treats plain text as ChatGPT prompts by default.

An unsuspecting user could potentially be tricked into copying and pasting a malformed link, believing they will be sent to a legitimate webpage. An attacker could plant the link behind a "copy link" button so that the user might not notice the suspicious text at the end of the link until after it is pasted and submitted. These prompt injections could potentially be used to instruct ChatGPT to open a new tab to a malicious website such as a phishing site, or to tell ChatGPT to take harmful actions in the user's integrated applications or logged-in sites like Google Drive, NeuralTrust said.
Last month browser security platform LayerX also described how malicious prompts could be hidden in URLs (as a parameter) for Perplexity's browser Comet. And last week SquareX Labs demonstrated that a malicious browser extension could spoof Comet's AI sidebar feature and have since replicated the proof-of-concept (PoC) attack on Atlas.

But another new vulnerability in ChatGPT Atlas "could allow malicious actors to inject nefarious instructions into the artificial intelligence (AI)-powered assistant's memory and run arbitrary code," reports The Hacker News, citing a report from browser security platform LayerX: "This exploit can allow attackers to infect systems with malicious code, grant themselves access privileges, or deploy malware," LayerX Security Co-Founder and CEO, Or Eshed, said in a report shared with The Hacker News. The attack, at its core, leverages a cross-site request forgery (CSRF) flaw that could be exploited to inject malicious instructions into ChatGPT's persistent memory. The corrupted memory can then persist across devices and sessions, permitting an attacker to conduct various actions, including seizing control of a user's account, browser, or connected systems, when a logged-in user attempts to use ChatGPT for legitimate purposes....

"What makes this exploit uniquely dangerous is that it targets the AI's persistent memory, not just the browser session," Michelle Levy, head of security research at LayerX Security, said. "By chaining a standard CSRF to a memory write, an attacker can invisibly plant instructions that survive across devices, sessions, and even different browsers. In our tests, once ChatGPT's memory was tainted, subsequent 'normal' prompts could trigger code fetches, privilege escalations, or data exfiltration without tripping meaningful safeguards...."

LayerX said the problem is exacerbated by ChatGPT Atlas' lack of robust anti-phishing controls, the browser security company said, adding it leaves users up to 90% more exposed than traditional browsers like Google Chrome or Microsoft Edge. In tests against over 100 in-the-wild web vulnerabilities and phishing attacks, Edge managed to stop 53% of them, followed by Google Chrome at 47% and Dia at 46%. In contrast, Perplexity's Comet and ChatGPT Atlas stopped only 7% and 5.8% of malicious web pages.
From The Conversation: Sandboxing is a security approach designed to keep websites isolated and prevent malicious code from accessing data from other tabs. The modern web depends on this separation. But in Atlas, the AI agent isn't malicious code — it's a trusted user with permission to see and act across all sites. This undermines the core principle of browser isolation.
Thanks to Slashdot reader spatwei for suggesting the topic.

According to the Wall Street Journal, SpaceX is reportedly poised to secure a $2 billion Pentagon contract to develop hundreds of missile-tracking satellites for President Trump's ambitious Golden Dome defense system. The Independent reports: The planned "air moving target indicator" system in question could ultimately feature as many as 600 satellites once it is fully operational, The Wall Street Journal reports. Musk's company has also been linked to two more satellite ventures, which are concerned with relaying sensitive communications and tracing vehicles, respectively.

Golden Dome, inspired by Israel's "Iron Dome," was announced by Trump and Secretary of War Pete Hegseth at the White House in May and will amount to a complex system of satellites and weaponry capable of destroying incoming missiles before they hit American targets. The president promised it would be "fully operational" before he leaves office in January 2029, capable of intercepting rockets, "even if they are launched from space," with an overall price tag of $175 billion.

OpenAI has introduced Aardvark, a GPT-5-powered autonomous agent that scans, reasons about, and patches code like a human security researcher. "By embedding itself directly into the development pipeline, Aardvark aims to turn security from a post-development concern into a continuous safeguard that evolves with the software itself," reports InfoWorld. From the report: What makes Aardvark unique, OpenAI noted, is its combination of reasoning, automation, and verification. Rather than simply highlighting potential vulnerabilities, the agent promises multi-stage analysis -- starting by mapping an entire repository and building a contextual threat model around it. From there, it continuously monitors new commits, checking whether each change introduces risk or violates existing security patterns.

Additionally, upon identifying a potential issue, Aardvark attempts to validate the exploitability of the finding in a sandboxed environment before flagging it. This validation step could prove transformative. Traditional static analysis tools often overwhelm developers with false alarms -- issues that may look risky but aren't truly exploitable. "The biggest advantage is that it will reduce false positives significantly," noted Jain. "It's helpful in open source codes and as part of the development pipeline."

Once a vulnerability is confirmed, Aardvark integrates with Codex to propose a patch, then re-analyzes the fix to ensure it doesn't introduce new problems. OpenAI claims that in benchmark tests, the system identified 92 percent of known and synthetically introduced vulnerabilities across test repositories, a promising indication that AI may soon shoulder part of the burden of modern code auditing.

The FCC plans to repeal a Biden-era ruling that required ISPs to secure their networks under the Communications Assistance for Law Enforcement Act, instead relying on voluntary cybersecurity commitments from telecom providers. FCC Chairman Brendan Carr said the ruling "exceeded the agency's authority and did not present an effective or agile response to the relevant cybersecurity threats." Carr said the vote scheduled for November 20 comes after "extensive FCC engagement with carriers" who have taken "substantial steps... to strengthen their cybersecurity defenses." Ars Technica reports: The FCC's January 2025 declaratory ruling came in response to attacks by China, including the Salt Typhoon infiltration of major telecom providers such as Verizon and AT&T. The Biden-era FCC found that the Communications Assistance for Law Enforcement Act (CALEA), a 1994 law, "affirmatively requires telecommunications carriers to secure their networks from unlawful access or interception of communications."

"The Commission has previously found that section 105 of CALEA creates an affirmative obligation for a telecommunications carrier to avoid the risk that suppliers of untrusted equipment will "illegally activate interceptions or other forms of surveillance within the carrier's switching premises without its knowledge,'" the January order said. "With this Declaratory Ruling, we clarify that telecommunications carriers' duties under section 105 of CALEA extend not only to the equipment they choose to use in their networks, but also to how they manage their networks." A draft of the order that will be voted on in November can be found here (PDF).