RSS 생중계

Taiwan deliberately throttled mobile internet across the Taichung area during its annual Han Kuang war games, "simulating communication disruption in the event of a Chinese attack or natural disaster," reports Reuters. The drill accompanied air-raid exercises and military rehearsals aimed at defending strategically important areas such as the Penghu Islands. From the report: Democratically governed Taiwan, which China claims as its own territory, has long lived with the threat of a Chinese invasion and holds air raid drills along with its war games every year, clearing people from the streets for 30 minutes. During the internet "throttling" drill, centered on Monday on the major central metropolis of Taichung, only basic mobile phone services worked, though other functions including ATMs, traffic signals, landlines and fixed internet services were not affected. Air raid sirens rang out in Taichung "and people cleared the streets from 2:30 p.m. (0630 GMT) as the government sent simultaneous text messages in Chinese and English telling mobile phone users of "simulated air attacks on communication infrastructure." The government had widely flagged the drill ahead of time, including posting details on social media, and most people Reuters spoke to at the Taichung train station were prepared. [...] The same exercise will be repeated on Thursday in northern Taiwan, including the capital Taipei.

Read more of this story at Slashdot.

An anonymous reader quotes a report from ABC News & Headlines: Andrew asked his personal assistant to book him a spot in one of his gym's coveted morning classes. It was a task he thought was well suited to this particular assistant because the booking form was online and because his assistant was not a person -- it was artificial intelligence (AI). But Andrew was shocked by what happened next. His AI assistant found a way to book the gym class months further in advance than the gym allowed, thanks to a vulnerability it discovered in the booking software. Then it went further, kicking someone out of the waiting list who was ahead of Andrew -- something it was not asked to do. The accidental hack is the first known Australian case of an emerging risk from a new generation of AI capable of behaving in unexpected ways.

Read more of this story at Slashdot.

Paramount is making commitments to major theater chains to release 30 films annually for three years if its Warner Bros. deal goes through. According to Deadline, the written commitment was sent to "at least one of the top two circuits" (specifically Regal but AMC may have also received a written commitment) and includes "an exclusive 45-day theatrical window and a 90-day window to streaming." From the report: When it comes to the big chains, it matters for them to be on the right side of history with Paramount, especially when it comes to future terms on a 30-film slate. However, the rest of Cinema United (formerly National Association of Theatre Owners) remains firmly against the Paramount-Warner Bros merger and its potential negative financial outcomes in regards to long-term box office, feature film output and loss of jobs. Cinema United has been trying to meet again with Paramount CEO David Ellison, but there's nothing on the books yet, we understand. [Regal boss Eduardo Acuna] told Deadline that in regards to Paramount's future promise to exhibition, "[Paramount CEO] David [Ellison] is sincere in making these commitments, and he has offered to execute a consent decree to the state AGs to that effect. These are tangible and important commitments that will benefit the industry and which can be measured." Paramount insiders tell us that no written commitments have been given to exhibitors yet, but Ellison is willing to make non-bespoke ones to any theater chain wanting them. And such deals will come with financial penalties to Paramount should it break such agreements with theaters. Again, this is all about Ellison standing behind his promises made at CinemaCon to maintain a hefty theatrical output and excellent windows to theaters.

Read more of this story at Slashdot.

BPF offers useful safety guarantees, but Kumar Kartikeya Dwivedi wants BPF programs to be even safer. At the 2026 Linux Storage, Filesystem, Memory-Management, and BPF Summit, he led a session (slides) discussing the possibility of adding domain-specific invariants to BPF programs. It was not a discussion intended to lead to the implementation of any particular kernel feature, but rather an overview of why additional formal verification might be needed, and how it could work with the existing BPF ecosystem.

The Django Python web-framework project has announced that it has accepted an annual release cycle proposal. This means that the project is moving from a somewhat complicated schedule that interspersed short-lived feature releases and long-term-support (LTS) releases to a simpler annual cycle where each release is supported for three years.

Every feature release gets three years of support: one year of mainstream bugfixes, then two years of security and data-loss fixes. The "LTS" label is retired — every feature release now carries that same, unique commitment.

No more LTS gap: no racing a deadline to jump two years of changes at once. Upgrade one year at a time, whenever suits you within the support window. Three versions are supported at any time, giving third-party packages a clear, rolling target.

This will take effect with the upcoming Django 2028 release, expected in January 2028.

Security updates have been issued by AlmaLinux (firefox, gpsd-minimal, kernel, libarchive, libgcrypt, and LibRaw), Debian (bind9, ca-certificates, chromium, dnsdist, icinga2, kitty, libheif, openjdk-21, pdns, pdns-recursor, thunderbird, and xen), Fedora (bird, erlang, kernel, mingw-glib2, nghttp2, p11-kit, perl, perl-Devel-Cover, perl-PAR-Packer, pgadmin4, polymake, python-nh3, python-wsgidav, python3.12, rabbitmq-server, rust-ammonia, seamonkey, and udisks2), Mageia (python-starlette), Oracle (gnutls, kernel, and LibRaw), Red Hat (container-tools:rhel8), Slackware (wpa_supplicant), and SUSE (azure-storage-azcopy, bouncycastle, ffmpeg-4, fuse-overlayfs, gleam, gstreamer-plugins-bad, libssh2-1, libssh2_org, libwireshark19, libXfont2-2, perl-Mojo-JWT, perl-Mojolicious, podman, python310, python313-Django4, and tekton-cli).

Three years ago environmental regulators discovered garbage burning deep inside California's Chiquita Canyon Landfill, reports the Los Angeles Times. [Alternate URL here.] "Few imagined the calamity that would follow. "Somehow the landfill's pollution control system introduced oxygen below, causing decades-old waste to rapidly decompose, generating intense heat and eventually widespread smoldering, federal and state environmental agencies say." The broiling temperatures released toxic gases and hazardous liquid waste that bubbled up to the surface. Residents living in the nearby communities of Val Verde and Castaic have phoned in thousands of complaints about stench and lived with nausea, dizziness and trouble breathing. The underground fire has tripled in size since first confirmed in 2023. Recently a towering mound of landfill garbage perched on the hillside has warped and threatens to trigger an avalanche of garbage and chemical-filled liquid, or leachate, according to the California water board... At first, local officials described the problem in Los Angeles County at Chiquita Canyon landfill as a rare, even unprecedented disaster. But since then, California regulators have learned that at least two other landfills are wrestling with high temperatures and similar conditions. Scorching temperatures have overtaken at least 30 acres of El Sobrante landfill near Corona in Riverside County. Excessive heat and emissions have also been detected at Avenal landfill in the San Joaquin Valley. The discoveries raise an unsettling question: How many landfills are dealing with these disasters? Are they a broader problem in California...? Chiquita Canyon closed at the end of the 2024 and stopped accepting new garbage. But the fire is expected to last for years as it continues to consume decades-old trash. "Underground landfill fires are complex, nearly impossible to extinguish once they break out," according to the article. "As regulators search for a solution, residents find themselves struggling with the wait." Some residents have worn gas masks — or sold their homes.

Read more of this story at Slashdot.

Futurism reports: 404 Media revealed that a July 10 audit by the Los Angeles Police Department Office of the Inspector General caught the department's ALPR cameras generating 161 false stolen-vehicle alerts in just two months — each one ending with officers pulling over an innocent driver. Factoring in 337 alerts which "resulted in the recovery of stolen vehicles," the LAPD's cameras carry an error rate of 32.3 percent, effectively giving officers a one-in-three chance at pulling an innocent person over. "The biggest issue remains this national database at the FBI called NCIC," Flock's CEO Garrett Langley recently told The Drive, complaining about "how archaic its structure is." Flock CEO: There's no feedback loop mechanisms; it's really just a static comma-separated file. We've tried to build around it. We have this concept called "suppression." A local agency — let's take Atlanta, where I live — might see that there's a stolen plate out of California, but it's already been resolved because it's a rental car or a dealer car. They can, in Flock, suppress that: "Never alert us on that for the next year." That's to get around the fact that they can't force another agency to remove it from NCIC. Ideally, the way it would work is if enough agencies — let's call it one, two, or three — flag a tag as no longer valid or a bad entry, it should just get removed. Or at least it should get pushed more aggressively by the FBI... I'm hopeful that they'll see there's an opportunity to make something like NCIC, which is an important tool not just for companies like Flock, but for police departments to work together, start to make some enhancements to modernize what's a central part of our policing system. Later in the interview he says "It's less than one in a million alerts that an officer says, 'I'm not sure if that's right.' Less than one in a million." A recent report from Business Insider shares a worst-case scenario. In the summer of 2024 a Sacramento police officer said Flock's cameras had sent six false alerts for the same vehicle, wrongly saying it had been stolen or used in a felony crime in the nearby suburb of Roseville: Roseville police could see that Flock's software kept confusing the "9" on the man's license plate for an "8." The car owner said he would take off his license plate cover. Soon after, it happened again. It's "easier at this point we have it memorized," a dispatch supervisor in Roseville wrote in an email to a colleague. Flock says that in optimal conditions, its cameras accurately read more than 96% of license plate characters. Hundreds of pages of records from the Roseville Police Department show a different picture. In 2023 and 2024, Flock sent 1,427 alerts to Roseville police, flagging vehicles as stolen or used in a felony after they passed one of the city's Flock cameras. An analysis by the police department found that in 71% of those alerts, Flock's machine-learning software incorrectly read the license plates... The cameras regularly missed vehicles, captured blurry images, misread license plate characters and states, and sent delayed alerts to police about vehicles possibly connected to crimes, the records show... A factor that contributed to the misread problems in Roseville was the "particularly unique deployment" that the city requested, a Flock spokeswoman said. Roseville said it has its cameras configured so that they capture only the backs of vehicles, a setup intended to avoid capturing personally identifiable information like faces. Roseville's setup included older hardware and placement of cameras higher and further from vehicles than the company typically recommends, Flock added... In Toledo, Ohio, driver Brandon Upchurch was mauled by a police dog after a Flock camera misread the "7" on his license plate as a "2." As a result of his injuries, he said, he lost his job and was evicted from his home. He settled a lawsuit against the city and police officer for $35,000. None of the incorrect Flock alerts in Roseville resulted in a traffic stop or arrest, a department spokesman said. Roseville requires police officers to verify that a license plate is stolen or have independent reasonable suspicion of a crime before making a traffic stop... Flock said Roseville's camera performance has significantly improved, which the police department disputed... Flock's cameras also missed vehicles altogether... At one point, Flock's product director for machine learning suggested that officers ask drivers to remove license plate frames that made it "very difficult for the machine vision to tell it is actually a 'E' and not an 'F.'" Roseville told Flock at the time that it wouldn't do so, according to the department spokesman. Flock in 2024 shared an analysis with Roseville's police department, outlining reasons for the misreads. Vehicles far from a camera were sometimes blurry. Plates were cut off by trees or license plate frames. And Flock's software confused similar characters, mistaking an "N" for a "V", or a "1" for a "4...." Roseville isn't the first organization to flag inaccuracies in Flock's technology. In 2021, the research firm IPVM independently tested Flock's license plate readers, concluding that Flock misidentified the state in about one of 10 reads, and that the system regularly misclassified vehicles' type and make. IPVM said that Flock subsequently blocked it from purchasing its cameras for testing. Thanks to long-time Slashdot reader Cognitive Dissident for sharing the article.

Read more of this story at Slashdot.

Microsoft's cloud storage app OneDrive got a new Photos app in the worst possible way, reports the blog Neowin . "The app is reportedly showing up even on Windows 11 Enterprise machines, despite apparently being a beta application aimed at consumer functionality." One admin questioned why a beta app was appearing on an Enterprise SKU in the first place, while another described the situation as yet another consumer-oriented feature being forced onto corporate PCs. Things get even more frustrating for IT departments because there does not appear to be a straightforward Microsoft-provided way to disable the app... Enterprise administrators generally need to know what is being installed on their managed devices, particularly when a software is labeled as beta. Quietly adding another application and leaving admins to clean it up themselves is therefore unlikely to win Microsoft many fans. But there's another problem, according to the blog Windows Latest. "OneDrive Photos automatically scans your system storage for photos," and apparently "doesn't need a Microsoft account to work, as it can also detect your local files." There's also a People section that groups similar faces in your photos. Microsoft asks for permission before turning it on and explicitly warns that facial data could be considered biometric data in some regions. The company says only you can see the grouped faces, that the data isn't shared with third parties, and that you can delete it by disabling the feature. In a statement to Neowin, Microsoft admitted this new photos "experience" they're "incubating" had gone "more broadly than it should have," and then promised that "We're fixing that." The spokesperson also said the Windows Photos app will "always give you the option of local and cloud photos" and, also a choice of whether or not to use it OneDrive." But there's another "awkward catch," notes the blog Digital Trends. "Users currently can't uninstall OneDrive Photos without removing the main OneDrive app too." Because OneDrive Photos is tied to the main OneDrive sync client, Windows 11 doesn't currently offer a separate uninstall option. The only straightforward way to get rid of OneDrive Photos right now is to uninstall OneDrive itself... Removing the main client can also affect its File Explorer integration and shortcuts... Microsoft says this will change. The company is working on controls that will let users remove OneDrive Photos separately from the main OneDrive app. On enterprise PCs managed through Intune, Microsoft says the app will automatically disappear where it isn't supported.

Read more of this story at Slashdot.

With nearly 70% of the market, "Meta wants its smart glasses to be a big hit," writes the Los Angeles Times. But after some users found ways to disable the light that warns people they're being filmed, "the high-tech specs have also turned into a liability, as some are calling the gadget 'pervert glasses...'" Some people are using the glasses — which look like a pair of regular spectacles — to record people without their knowledge and publish the videos on social media. The users secretly videotape and share what happens when they try to pick up women. Now, a growing number of people are worried they could be covertly recorded at the gym or even during sex. The backlash has prompted Meta to update its glasses to address privacy concerns, and some places have banned them, adding to the angst surrounding technology that's rapidly evolving... Concern about the glasses has exploded as more videos of interactions with people who don't know they're being recorded go viral on social media. On Instagram, some videos depict people getting approached in malls and grocery stores and on college campuses and sidewalks. While the videos are portrayed as jokes, the people filmed don't appear to know they're being recorded and sometimes seem uncomfortable, telling the strangers to leave them alone or stop harassing them. The outcry has spread beyond social media, with comedian Jimmy Kimmel calling the Meta devices "pervert glasses" on national television and singer Lorde telling concertgoers that smart glasses are "not sexy..." Instagram, which is owned by Meta, has been disabling accounts and taking down some of these pickup or prank videos for violating the platform's rules against harassment and bullying... Businesses are making their own calls about smart glasses. The 5 Point Cafe in Seattle, which banned Google Glass in the past, has banned Meta glasses from its diner and dive bar. "Leave your Meta-SpyBan Display at home, they are officially Ray-Ban-ned from all of our restaurants. We serve privacy, not side-eye surveillance," a 2025 Instagram post from the business states. An Android app that warns people if they're being recorded has roughly 110,000 downloads in the last six months, according to the article. The app's creator says it's a "social problem" that "we don't value privacy, that we feel entitled to use others for our entertainment or our private gain, and technology amplifies that." The American Civil Liberties Union and more than 70 organizations even sent a letter urging Meta to promise they'd leave facial recognition features out of their smart glasses, according to the article. But a Meta spokesperson said "no final decision has been made."

Read more of this story at Slashdot.

Linux Torvalds expects Linux 7.2 should be released next weekend "unless something really bad pops up," Torvalds said while announcing today's release candidate. But there's something interesting about Linux 7.2-rc7, writes Phoronix. "By the time the Linux kernel typically hits a -rc7 release things have usually settled quite well. But in today's world of AI/LLM coding/review agents, the kernel activity continues at an all-time high." Tons of bug fixes continued to trickle in across the kernel spectrum for all sorts of issues. The HWMON hardware monitoring subsystem saw several critical and high severity bug fixes, on the memory management side was a nasty race condition leading to a use-after-free in the kernel for the past eight years, Btrfs restored its fixup worker infrastructure to deal with silent data loss, lots of AI patches in the networking realm, and the kernel was patched for the Safe RET Interrupt Vulnerability. Linus Torvalds wrote in the 7.2-rc7 announcement: "Another week, another -rc. I can't say that I'm exactly thrilled about the size of this all, but it is what it is: the new normal with a lot of fixes, many of them due to review by various AI tools. And nothing looks particularly scary per se — it's just that there's a lot here. Most of it is fairly small, although we have a couple of larger diffs: s390/zcrypt fixes stand out in the diffstat, and so does btrfs bringing back the fixup worker infrastructure. And some netfilter ipset fixes. But aside from a few places like that, most of this is just lots of tiny fixes. It's pretty much spread all over — drivers (gpu, sound, networking, you name it), filesystems, core networking, arch code...

Read more of this story at Slashdot.

A Massachusetts newspaper reports that officials from the town of Littleton (population: 10,141) "claim that five of Littletonâ(TM)s Flock Safety cameras were switched back on by the company without informing the town..." Littleton had temporarily disabled the cameras earlier this year, according to the article, but upon discovering the cameras were still active, city officials say they directed the Town Counsel "to initiate cancellation of the contract with Flock Safety and directed the Town Administrator to coordinate the removal of the six cameras on public ways" with the Department of Public Works. Ironically, it's the same Massachusetts town that last year was breached by a Chinese-state-sponsored hacking group for over 300 days. Flock has also considered dashcams installed in hundreds of thousands Uber, Lyft, and delivery drivers' vehicles "to scan license plates those drivers travelled passed," reports 404 Media, "essentially turning Uber and Lyft drivers into roaming surveillance vehicles, according to a Flock presentation shared with 404 Media... 404 Media first revealed the intended partnership last August when multiple sources provided information on the plan."

Read more of this story at Slashdot.

The 7.2-rc7 kernel prepatch is out for testing. It is still bigger than Linus would like, but he said nonetheless: "I don't currently see any value in delaying the 7.2 release, so I would expect that to happen next weekend unless something really bad pops up."

NBC Bay Area brings news from the small idyllic California town of Suisun City (population: 29,518). The city council "declared a state of emergency Saturday after a cyberattack that shut down computer systems including 911 public safety operations." "Malicious software infected and compromised" the East Bay community's information technology systems beginning about 5:45 a.m. Friday, the city said on social media. "The cybersecurity incident hit critical public safety operations, including 911 routing, police and fire dispatch, records and city services," the city said. "There is no imminent threat to the public," the city said. "All public safety services remain active." The city shut down its computer network "to contain the threat and preserve evidence for a federal investigation," it said... Suisun City dispatchers were taking calls for service through the Solano County dispatch center. Suisun City police officers and firefighters continued responding to calls for service, the city said, but online city services and internal operations were unavailable. On Facebook the city said it was "working alongside federal, state and regional agencies, including the FBI, Department of Homeland Security and California Office of Emergency Services to maintain emergency services, investigate the incident and restore systems." But Bay Area PBS station KQED notes "They're only the latest local government in the region to be hit by a cyberattack." Earlier this year, Foster City and Pittsburg experienced other cyberattacks that impacted services in their own jurisdictions. Foster City declared a state of emergency after a March ransomware incident that shut down many services for more than a week. Pittsburg lost almost $1 million to a February phishing attack, though over half of it was recovered.

Read more of this story at Slashdot.

XDA Developers reports: On July 22, a Google DeepMind engineer, Vidy Thatte, shared a snippet of a browser he built that "treats every URL as a prompt and generates a site from scratch" on his X account. Just a few days later, he shared a TestFlight link to let iPhone users try it for themselves. The browser he launched is called Gem, and it's a browser that doesn't really...browse. Instead of fetching a webpage from a server the way Chrome or Safari would, Gem hands whatever URL you type over to Google's Gemini 3.5 Flash Lite model and asks it to generate a webpage on the spot. If you enter a real address, it builds its own interpretation of that site rather than loading the actual thing. If you enter an address that doesn't exist, Gemini simply invents a website to fill the gap. In other words, you're not visiting the internet so much as browsing one the model dreams up as you go... Thatte's reasoning was that with a model as fast and cheap as Gemini 3.5 Flash Lite, there's almost no practical difference anymore between loading a website and generating a brand-new one on the fly... Given that this tool is powered by Gemini 3.5 Flash Lite and it's just a side project rather than a full-fledged tool, it doesn't come with Gemini access baked in. Gem requires you to bring your own Gemini API key to get it working. Every URL you enter fires off a request to the model, so Gem needs a key linked to your Google account to actually generate anything, with the usage billed to you. You can grab a key for free from Google AI Studio, paste it into Gem's settings, and you're ready to start typing URLs. While you can get started for free, I ran into the limits within two minutes of playing around. So, I did need to enable billing on my API key and decided to load $10 into it. While the blogger's own site seemed to only get the Gemini logo, tapping it revealed nearly two dozen remixing options. (Dark mode, Reader, Neubrutalist, Broadsheet, Blueprint, Comic book, Punk zine, Notebook, Chalkboard, Receipt, Teletext, System 7, Wes Anderson, Cyber neon, Clay, Frosted, Geocities, Matrix, Museum, Windows 95, Terminal, Vaporwave, and Hide images...) "The frontend morphed before my eyes. Every new edit took practically no time to load, and within a second or two, the same XDA articles would reappear wearing a completely different skin..." For another site, it kept the article headlines, but then rewrote all the text! And after entering an address they knew didn't exist — their own name — in two seconds the browser whipped up a slick portfolio "genuinely been better than some of what those tools produced with far more time and context to work with." But its link to a LinkedIn profile led to a lookalike page, because Gem "had simply generated its own version of it, complete with a made-up follower number and details I never wrote...."

Read more of this story at Slashdot.

UI changes, lost productivity, untransferrable knowledge... Long-time Slashdot reader theodp notes the folly of switching to an unfamiliar new environment to be "more productive"... Champions of the switch often insist workers will be able to hit the ground running and get the same or even more work done from the get-go in the new unfamiliar environment because it's 'more productive', a conclusion often shaped by conventional wisdom or their 'gut' feeling. Decision makers may also mistake their own 'light-user experience' for universal experience and believe all platforms are fungible. And while it's one thing for a business to delay their delivery dates and increase costs of by introducing a switch in computing environments that's of unquantified and possibly dubious value, how about when a switch in computing environments is made that affects schoolchildren's ability to learn for a period of time? While a MacBook Neo is indeed cute, what might be the possible consequences of a decision to move 25,000 students off Windows and onto MacBook Neos...? Or even the impact on a single kid who moves from a MacBook school to a Windows or Chrome school (or whose home computing environment isn't aligned with their school). Or students whose teachers decide learning in their classrooms should now additionally require an understanding of the Minecraft computing platform/UI (used in over 40,000 school systems.). Learning Management Systems and now various AI Edtech tools used by schoolkids are also subject to change. "Even if students and schools eventually benefit from a new platform, one wonders how much learning is lost during the switch," theodp asks, "and what the financial and educational ROI and payback period look like." Are there bigger problems than we realize? Share your own thoughts and experiences in the comments. What are the costs of switching between Windows, MacOS, and Chrome?

Read more of this story at Slashdot.

The 7.1.8, 6.18.44, 6.12.103, and 6.6.151 stable kernels have been released; each contains a fair number of important fixes.

"I've been a Linux IT administrator for 30 years now," says crhylove (Slashdot reader #205,956), in a new video on his YouTube channel. "I used to be a data center architect, at one point I was the IT director for 13 school districts... I just haven't been happy overall with all the other operating systems out there... " I've been working really diligently for the last few weeks on a new distro called Spaced Linux that is exactly what I want to use. It's heavily focused on the [zoom-friendly 3D] Compiz desktop, very efficient using the rolling Devuan Ceres backend, and avoids dependency hell by promoting mostly Flatpak usage. It's still early, but it's already working great for me, it's been my daily driver for a few weeks now. There's also user-friendly themes — everything from Linux, Android, GeoWorks, and Mac OS X to Windows 11, Windows XP, and even Windows 3.11 (dark or light theme). And you can also run Windows apps under Wine. "And let me tell you, Nvidia runs perfectly on here now..." he adds in the video. "It just works really, really well out of the box." He's now working on voice-command features, bur argues that Spaced Linux "is already extremely usable, extremely efficient, extremely fast, and extremely reliable.... And it's only going to get better and better."

Read more of this story at Slashdot.

OpenAI announced Friday it's pausing work on its Astra AI model because of security concerns. The Guardian reports: The company had evaluated the agent, Astra, and found "significant advancements in agentic coding and cybersecurity", which had moved to a "critical" threshold... OpenAI stated that the model was not involved in an incident in which one of its AI agents went rogue during a test, accessed the open web and hacked a startup, Hugging Face... The reports have increased concerns about advancements in AI models and humans' ability to control them. Still, critics of the AI industry have warned that such disclosures from OpenAI and its competitors Anthropic and Meta could be designed to generate hype about the technology's power and thus spur additional interest from investors. To prevent potential rogue behavior from AI agents, OpenAI is "implementing stricter security controls for higher-capability models and associated activities, including isolated testing environments, restricted network and tool access", the company's blogpost stated. It will also install "enhanced model weight protections and encryption, additional monitoring and detection capabilities". The company will pause internal activities involving Astra that do not meet these new requirements. "We believe it's important to be transparent with the public and the safety and security communities about this potential shift in capabilities..." OpenAI wrote in a blog post titled "Responding to the next frontier of critical cyber capabilities." Under our Preparedness Framework, a model reaches the Critical cybersecurity threshold if it can identify and develop functional zero-day exploits of all severity levels in many hardened real-world critical systems without human intervention, or can devise and execute end-to-end novel strategies for cyberattacks against hardened targets given only a high level desired goal. While we continue to benchmark and assess this model, our preliminary evaluations indicate strong enough performance that we cannot rule out Critical capability level at this time... Accordingly, we have scaled up robustness testing of our safeguards and security controls so that they are appropriate for a deployment of these capabilities... - We are implementing stricter security controls for higher-capability models and associated activities, including isolated testing environments, restricted network and tool access, enhanced model weight protections and encryption, additional monitoring and detection capabilities, and sandboxed execution. - We are pausing internal activities involving Astra that do not yet meet these strengthened security control requirements. - We have implemented universal monitoring for risky actions and misalignment across all agentic applications of Astra, including training and evaluation. Monitors evaluate the model's Chain of Thought and trigger a security response to review and interrupt high risk activity. - We will work with relevant government agencies and select AI safety organizations to test the capabilities for this model... We believe advanced cyber-capable models should help defenders identify and address vulnerabilities before attackers do. We're committed to working alongside governments, safety institutes, and civil society to ensure that the frontier capabilities of models like Astra, and those that follow, are deployed responsibly and broadly for the benefit of all humanity.

Read more of this story at Slashdot.

Researchers believe they've found evidence that brain signals reveal the pitches of notes in imagined musical melodies, reports Medical Express: Study participants were 10 epilepsy patients with electrodes on the surfaces of their brains for clinical monitoring. Participants quietly listened to the first bars of children's tunes, then imagined the melodies of subsequent bars before humming their imagined melodies. Brain activity from the patients carried information about relative pitches within the imagined melodies. Combining note-level pitch predictions over time allowed the research team to recreate melodic changes that reflected the structure of the imagined songs... "The model we created decodes relative pitch classes — such as do, re, mi, fa, sol, and la — rather than exact, absolute pitches," [said tudy co-lead Jii Kwon of Seoul National University]. "This is useful because people often recognize melodies by the relationships between notes, even when the same melody is played in a different key."

Read more of this story at Slashdot.