RSS 생중계

A new benchmark assembled by a team of International Olympiad medalists suggests the hype about large language models beating elite human coders is premature. LiveCodeBench Pro, unveiled in a 584-problem study [PDF] drawn from Codeforces, ICPC and IOI contests, shows the best frontier model clears just 53% of medium-difficulty tasks on its first attempt and none of the hard ones, while grandmaster-level humans routinely solve at least some of those highest-tier problems. The researchers measured models and humans on the same Elo scale used by Codeforces and found that OpenAI's o4-mini-high, when stripped of terminal tools and limited to one try per task, lands at an Elo rating of 2,116 -- hundreds of points below the grandmaster cutoff and roughly the 1.5 percentile among human contestants. A granular tag-by-tag autopsy identified implementation-friendly, knowledge-heavy problems -- segment trees, graph templates, classic dynamic programming -- as the models' comfort zone; observation-driven puzzles such as game-theory endgames and trick-greedy constructs remain stubborn roadblocks. Because the dataset is harvested in real time as contests conclude, the authors argue it minimizes training-data leakage and offers a moving target for future systems. The broader takeaway is that impressive leaderboard jumps often reflect tool use, multiple retries or easier benchmarks rather than genuine algorithmic reasoning, leaving a conspicuous gap between today's models and top human problem-solvers.

Read more of this story at Slashdot.

Security updates have been issued by AlmaLinux (.NET 8.0, .NET 9.0, buildah, containernetworking-plugins, firefox, gstreamer1-plugins-bad-free, libsoup3, podman, skopeo, sqlite, thunderbird, unbound, valkey, varnish, and xz), Debian (webkit2gtk), Fedora (fido-device-onboard, python-django4.2, rust-git-interactive-rebase-tool, and thunderbird), Red Hat (libsoup), Slackware (libxml2), SUSE (java-11-openjdk, kernel, and wireshark), and Ubuntu (c3p0, dojo, python-django, python3.13, python3.12, python3.11, python3.10, python3.9, python3.8, python3.7, python3.6, and requests).

Longtime Slashdot reader UnknowingFool writes: A new documentary released last week on Netflix goes into detail about events leading up to the destruction of OceanGate's submersible, Titan that imploded on June 18, 2023 while attempting to visit the wreckage of the RMS Titanic off the coast of Newfoundland. The Titan used a carbon-fiber hull instead of more traditional materials like steel or titanium. "Through exclusive access to whistleblower testimony, pivotal audio recordings, and footage from the company's early days, the film provides an unprecedented look at the technical challenges, moral dilemmas, and shockingly poor decisions that culminated in the catastrophic expedition," explains Netflix in an article. Some highlights: - Titan's original carbon-fiber hull had been replaced with a second carbon-fiber one after the first one developed noticeable cracks. - Three scale models of the second hull failed tests. OceanGate decided to manufacture the second hull regardless of these failures. - Loud pops were heard in many dives; CEO Stockton Rush dismissed these as "seasoning". - Many employees raised numerous safety concerns. They were fired like lead pilot and head of marine operations, David Lochridge. Or they quit. - Some employees like Emily Hammermeister wanted to quit earlier, but external conditions like the COVID pandemic made it difficult. After the scale models failed, she refused to bolt anyone in the future submersible. She was given the two options of being fired or quit; she quit in the middle of the pandemic. - Rush's blindness to inconvenient facts: After the crack was discovered, Rush questioned Director of Engineering, Tony Nissen, about why Nissen did not anticipate the possibility of a crack. Nissen: "I wrote you a report that showed you it was there." Nissen had warned repeatedly that the hull's fibers were breaking (the pops) with each dive. Rush: "Well, one of us has to go." - Poor decisions by Rush extended beyond engineering decisions. After Rush fired Lochridge for raising safety concerns , Rush wanted Bonnie Carl, the company's accountant, to be his replacement pilot. While Carl was an experienced scuba diver, she quit as she was extremely uncomfortable being a pilot. Her explanation: "Are you nuts? I'm an accountant."

Read more of this story at Slashdot.

sciencehabit shares a report from Science.org: The newly discovered microbe provisionally known as Sukunaarchaeum isn't a virus. But like viruses, it seemingly has one purpose: to make more of itself. As far as scientists can tell from its genome -- the only evidence of its existence so far -- it's a parasite that provides nothing to the single-celled creature it calls home. Most of Sukunaarchaeum's mere 189 protein-coding genes are focused on replicating its own genome; it must steal everything else it needs from its host Citharistes regius, a dinoflagellate that lives in ocean waters all over the world. Adding to the mystery of the microbe, some of its sequences identify it as archaeon, a lineage of simple cellular organisms more closely related to complex organisms like us than to bacteria like Escherichia coli. The discovery of Sukunaarchaeum's bizarrely viruslike way of living, reported last month in a bioRxiv preprint, "challenges the boundaries between cellular life and viruses," says Kate Adamala, a synthetic biologist at the University of Minnesota Twin Cities who was not involved in the work. "This organism might be a fascinating living fossil -- an evolutionary waypoint that managed to hang on." Adamala adds that if Sukunaarchaeum really does represent a microbe on its way to becoming a virus, it could teach scientists about how viruses evolved in the first place. "Most of the greatest transitions in evolution didn't leave a fossil record, making it very difficult to figure out what were the exact steps," she says. "We can poke at existing biochemistry to try to reconstitute the ancestral forms -- or sometimes we get a gift from nature, in the form of a surviving evolutionary intermediate." What's already clear: Sukunaarchaeum is not alone. When team leader Takuro Nakayama, an evolutionary microbiologist at Tsukuba, and his colleagues sifted through publicly available DNA sequences extracted from seawater all over the world, they found many sequences similar to those of Sukunaarchaeum. "That's when we realized that we had not just found a single strange organism, but had uncovered the first complete genome of a large, previously unknown archaeal lineage," Nakayama says.

Read more of this story at Slashdot.

Denmark has deployed four uncrewed robotic sailboats (known as "Voyagers") for a three-month trial to boost maritime surveillance amid rising tensions in the Baltic region. The Associated Press reports: Built by Alameda, California-based company Saildrone, the vessels will patrol Danish and NATO waters in the Baltic and North Seas, where maritime tensions and suspected sabotage have escalated sharply since Russia's full-scale invasion of Ukraine on Feb. 24, 2022. Two of the Voyagers launched Monday from Koge Marina, about 40 kilometers (25 miles) south of the Danish capital, Copenhagen. Powered by wind and solar energy, these sea drones can operate autonomously for months at sea. Saildrone says the vessels carry advanced sensor suites -- radar, infrared and optical cameras, sonar and acoustic monitoring. Their launch comes after two others already joined a NATO patrol on June 6. Saildrone founder and CEO Richard Jenkins compared the vessels to a "truck" that carries sensors and uses machine learning and artificial intelligence to give a "full picture of what's above and below the surface" to about 20 to 30 miles (30 to 50 kilometers) in the open ocean. He said that maritime threats like damage to undersea cables, illegal fishing and the smuggling of people, weapons and drugs are going undetected simply because "no one's observing it." Saildrone, he said, is "going to places ... where we previously didn't have eyes and ears." The Danish Defense Ministry says the trial is aimed at boosting surveillance capacity in under-monitored waters, especially around critical undersea infrastructure such as fiber-optic cables and power lines.

Read more of this story at Slashdot.

An anonymous reader quotes a report from the BBC: Social media and video networks have become the main source of news in the US, overtaking traditional TV channels and news websites, research suggests. More than half (54%) of people get news from networks like Facebook, X and YouTube -- overtaking TV (50%) and news sites and apps (48%), according to the Reuters Institute. "The rise of social media and personality-based news is not unique to the United States, but changes seem to be happening faster -- and with more impact -- than in other countries," a report found. Podcaster Joe Rogan was the most widely-seen personality, with almost a quarter (22%) of the population saying they had come across news or commentary from him in the previous week. The report's author Nic Newman said the rise of social video and personality-driven news "represents another significant challenge for traditional publishers." Other key findings from the report include: - TikTok is the fastest-growing social and video platform, now used for news by 17% globally (up 4% from last year). - AI chatbot use for news is increasing, especially among under-25s, where it's twice as popular as in the general population. - Most people believe AI will reduce transparency, accuracy, and trust in news. - Across all age groups, trusted news brands with proven accuracy remain valued, even if used less frequently.

Read more of this story at Slashdot.

alternative_right shares a report from ScienceDaily: When the brain is under pressure, certain neural signals begin to move in sync -- much like a well-rehearsed orchestra. A new study from Johannes Gutenberg University Mainz (JGU) is the first to show how flexibly this neural synchrony adjusts to different situations and that this dynamic coordination is closely linked to cognitive abilities. "Specific signals in the midfrontal brain region are better synchronized in people with higher cognitive ability -- especially during demanding phases of reasoning," explained Professor Anna-Lena Schubert from JGU's Institute of Psychology, lead author of the study recently published in the Journal of Experimental Psychology: General. The researchers focused on the midfrontal area of the brain and the measurable coordination of the so-called theta waves. These brainwaves oscillate between four and eight hertz and belong to the group of slower neural frequencies. "They tend to appear when the brain is particularly challenged such as during focused thinking or when we need to consciously control our behavior," said Schubert, who heads the Analysis and Modeling of Complex Data Lab at JGU. The 148 participants in the study, aged between 18 and 60, first completed tests assessing memory and intelligence before their brain activity was recorded using electroencephalography (EEG). [...] As a result, individuals with higher cognitive abilities showed especially strong synchronization of theta waves during crucial moments, particularly when making decisions. Their brains were better at sustaining purposeful thought when it mattered most. "People with stronger midfrontal theta connectivity are often better at maintaining focus and tuning out distractions, be it that your phone buzzes while you're working or that you intend to read a book in a busy train station," explained Schubert. The findings have been published in the Journal of Experimental Psychology.

Read more of this story at Slashdot.

Google Cloud has attributed last week's widespread outage to a flawed code update in its Service Control system that triggered a global crash loop due to missing error handling and lack of feature flag protection. The Register reports: Google's explanation of the incident opens by informing readers that its APIs, and Google Cloud's, are served through our Google API management and control planes." Those two planes are distributed regionally and "are responsible for ensuring each API request that comes in is authorized, has the policy and appropriate checks (like quota) to meet their endpoints." The core binary that is part of this policy check system is known as "Service Control." On May 29, Google added a new feature to Service Control, to enable "additional quota policy checks." "This code change and binary release went through our region by region rollout, but the code path that failed was never exercised during this rollout due to needing a policy change that would trigger the code," Google's incident report explains. The search monopolist appears to have had concerns about this change as it "came with a red-button to turn off that particular policy serving path." But the change "did not have appropriate error handling nor was it feature flag protected. Without the appropriate error handling, the null pointer caused the binary to crash." Google uses feature flags to catch issues in its code. "If this had been flag protected, the issue would have been caught in staging." That unprotected code ran inside Google until June 12th, when the company changed a policy that contained "unintended blank fields." Here's what happened next: "Service Control, then regionally exercised quota checks on policies in each regional datastore. This pulled in blank fields for this respective policy change and exercised the code path that hit the null pointer causing the binaries to go into a crash loop. This occurred globally given each regional deployment." Google's post states that its Site Reliability Engineering team saw and started triaging the incident within two minutes, identified the root cause within 10 minutes, and was able to commence recovery within 40 minutes. But in some larger Google Cloud regions, "as Service Control tasks restarted, it created a herd effect on the underlying infrastructure it depends on ... overloading the infrastructure." Service Control wasn't built to handle this, which is why it took almost three hours to resolve the issue in its larger regions. The teams running Google products that went down due to this mess then had to perform their own recovery chores. Going forward, Google has promised a couple of operational changes to prevent this mistake from happening again: "We will improve our external communications, both automated and human, so our customers get the information they need asap to react to issues, manage their systems and help their customers. We'll ensure our monitoring and communication infrastructure remains operational to serve customers even when Google Cloud and our primary monitoring products are down, ensuring business continuity."

Read more of this story at Slashdot.

Intel will lay off 15% to 20% of its factory workforce starting in July, potentially cutting over 10,000 jobs as part of a broader effort to streamline operations amid declining sales and mounting competitive pressure. "These are difficult actions but essential to meet our affordability challenges and current financial position of the company. It drives pain to every individual," Intel manufacturing Vice President Naga Chandrasekaran wrote to employees Saturday. "Removing organizational complexity and empowering our engineers will enable us to better serve the needs of our customers and strengthen our execution. We are making these decisions based on careful consideration of what's needed to position our business for the future." The company reiterated that "we will treat people with care and respect as we complete this important work." Oregon Live reports: Intel announced the pending layoffs in April and notified factory workers last week that the cuts would begin in July. It hadn't previously said just how deep the layoffs will go. The company had 109,000 employees at the end of 2024, but it's not clear how many of those worked in its factory division -- called Intel Foundry. The Foundry business includes a broad array of jobs, from technicians on the factory floor to specialized researchers who work years in advance to develop future generations of microprocessors. Intel is planning major cuts in other parts of its business, too, but employees say the company hasn't specified how many jobs it will eliminate in each business unit. Workers say they believe the impacts will vary within departments. Overall, though, the layoffs will surely eliminate several thousand jobs -- and quite possibly more than 10,000.

Read more of this story at Slashdot.

An anonymous reader quotes a report from Ars Technica: Subscribers in Southern California of Spectrum's Internet service experienced outages over the weekend following what company officials said was an attempted theft of copper lines located in Van Nuys, a suburb located 20 miles from downtown Los Angeles. The people behind the incident thought they were targeting copper lines, the officials wrote in a statement Sunday. Instead, they cut into fiber optic cables. The cuts caused service disruptions for subscribers in Van Nuys and surrounding areas. Spectrum has since restored service and is offering a $25,000 reward for information leading to the apprehension of the people responsible. Spectrum will also credit affected customers one day of service on their next bill. "Criminal acts of network vandalism have become an issue affecting the entire telecommunications industry, not just Spectrum, largely due to the increase in the price of precious metals," the officials wrote in a statement issued Sunday. "These acts of vandalism are not only a crime, but also affect our customers, local businesses and potentially emergency services. Spectrum's fiber lines do not include any copper." Outage information service Downdetector showed that thousands of subscribers in and around Van Nuys reported outages starting a little before noon on Sunday. Within about 12 hours, the complaint levels returned to normal. Spectrum officials told the Los Angeles Times that personnel had to splice thousands of fiber lines to restore service to affected subscribers.

Read more of this story at Slashdot.

Threads is testing a new feature that lets users hide spoiler content by blurring images or text, which can then be revealed with a tap. The Verge reports: Meta spokesperson Alec Booker told The Verge that this is a "global test," though it's not clear how many people will gain access to it. Spoilers will also look a bit different depending on which device you're using. On desktop, spoilers are hidden by a gray block, but they appear behind a bunch of floating dots on mobile (which you can see in the GIF embedded [here]). "This feature is currently optimized for mobile, but we're working to improve the experience for desktop," Booker said.

Read more of this story at Slashdot.

A new Salesforce-led study found that LLM-based AI agents struggle with real-world CRM tasks, achieving only 58% success on simple tasks and dropping to 35% on multi-step ones. They also demonstrated poor confidentiality awareness. "Agents demonstrate low confidentiality awareness, which, while improvable through targeted prompting, often negatively impacts task performance," a paper published at the end of last month said. The Register reports: The Salesforce AI Research team argued that existing benchmarks failed to rigorously measure the capabilities or limitations of AI agents, and largely ignored an assessment of their ability to recognize sensitive information and adhere to appropriate data handling protocols. The research unit's CRMArena-Pro tool is fed a data pipeline of realistic synthetic data to populate a Salesforce organization, which serves as the sandbox environment. The agent takes user queries and decides between an API call or a response to the users to get more clarification or provide answers. "These findings suggest a significant gap between current LLM capabilities and the multifaceted demands of real-world enterprise scenarios," the paper said. [...] AI agents might well be useful, however, organizations should be wary of banking on any benefits before they are proven.

Read more of this story at Slashdot.

An anonymous reader quotes a report from TechCrunch: While Silicon Valley executives like those from Palantir, Meta, and OpenAI are grabbing headlines for trading their Brunello Cucinelli vests for Army Reserve uniforms, a quieter transformation has been underway in the U.S. Navy. How so? Well, the Navy's chief technology officer, Justin Fanelli, says he has spent the last two and a half years cutting through the red tape and shrinking the protracted procurement cycles that once made working with the military a nightmare for startups. The efforts represent a less visible but potentially more meaningful remaking that aims to see the government move faster and be smarter about where it's committing dollars. "We're more open for business and partnerships than we've ever been before," Fanelli told TechCrunch in a recent episode of StrictlyVC Download. "We're humble and listening more than before, and we recognize that if an organization shows us how we can do business differently, we want that to be a partnership." Right now, many of these partnerships are being facilitated through what Fanelli calls the Navy's innovation adoption kit, a series of frameworks and tools that aim to bridge the so-called Valley of Death, where promising tech dies on its path from prototype to production. "Your granddaddy's government had a spaghetti chart for how to get in," Fanelli said. "Now it's a funnel, and we are saying, if you can show that you have outsized outcomes, then we want to designate you as an enterprise service." In one recent case, the Navy went from a Request for Proposal (RFP) to pilot deployment in under six months with Via, an eight-year-old, Somerville, Massachusetts-based cybersecurity startup that helps big organizations protect sensitive data and digital identities through, in part, decentralization, meaning the data isn't stored in one central spot that can be hacked. (Another of Via's clients is the U.S. Air Force.) The Navy's new approach operates on what Fanelli calls a "horizon" model, borrowed and adapted from McKinsey's innovation framework. Companies move through three phases: evaluation, structured piloting, and scaling to enterprise services. The key difference from traditional government contracting, Fanelli says, is that the Navy now leads with problems rather than predetermined solutions. "Instead of specifying, 'Hey, we'd like this problem solved in a way that we've always had it,' we just say, 'We have a problem, who wants to solve this, and how will you solve it?'" Fanelli said.

Read more of this story at Slashdot.

Thousands of American investors have lost millions of dollars to sophisticated pump-and-dump schemes involving small Chinese companies listed on Nasdaq, prompting the Justice Department to declare the fraud a priority under the Trump administration's white-collar enforcement program. The scams recruit victims through social media ads and WhatsApp messages, directing them to purchase shares in obscure Chinese firms whose stock prices are artificially inflated before collapsing. Since 2020, nearly 60 China-based companies have conducted initial public offerings on Nasdaq raising $15 million or less each, with more than one-third experiencing sudden single-day price drops exceeding 50%. In one recent case, seven traders earned over $480 million by defrauding 600 victims who purchased shares in China Liberal Education Holdings.

Read more of this story at Slashdot.

The Kubernetes project has announced that it will be losing its "special status" with the Slack communication platform and will be downgraded to the free tier in a matter of days:

On Friday, June 20, we will be subject to the feature limitations of free Slack. The primary ones which will affect us will be only retaining 90 days of history, and having to disable several apps and workflows which we are currently using. The Slack Admin team will do their best to manage these limitations.

The project has a FAQ covering the change, its impacts, and more. The CNCF projects staff has proposed a move to the Discord service as the best option to handle the more than 200,000 users and thousands of posts per day from the Kubernetes community. The Kubernetes Steering Committee will be making its decision "in the next few weeks".

Tensions between OpenAI and Microsoft over the future of their famed AI partnership are flaring up. WSJ, minutes ago: OpenAI wants to loosen Microsoft's grip on its AI products and computing resources, and secure the tech giant's blessing for its conversion into a for-profit company. Microsoft's approval of the conversion is key to OpenAI's ability to raise more money and go public. But the negotiations have been so difficult that in recent weeks, OpenAI's executives have discussed what they view as a nuclear option: accusing Microsoft of anticompetitive behavior during their partnership, people familiar with the matter said. That effort could involve seeking federal regulatory review of the terms of the contract for potential violations of antitrust law, as well as a public campaign, the people said.

Read more of this story at Slashdot.

Researchers are cautioning users against clicking unsubscribe links embedded in email bodies, citing new data showing such actions can expose recipients to malicious websites and confirm active email addresses to attackers. DNSFilter found that one in every 644 clicks on unsubscribe links leads users to potentially malicious websites. "You've left the safe, structured environment of your email client and entered the open web," TK Keanini, DNSFilter's chief technology officer, told WSJ. The risks range from confirming to bad actors that an email address belongs to an active user to redirecting victims to fake websites designed to steal login credentials or install malware. Clicking such links "can make you a bigger target in the future," said Michael Bargury, CTO of security company Zenity.

Read more of this story at Slashdot.

A Dutch court on Monday confirmed a 2021 consumer watchdog's ruling saying that Apple had abused its dominant position by imposing unfair conditions on providers of dating apps in the App Store. From a report: The Rotterdam District Court ruled that the Dutch Authority for Consumers and Markets (ACM) was therefore right to impose an order subject to a penalty for non-compliance. The court ruled that ACM was right in finding that dating app providers had to use Apple's own payment system, were not allowed to refer to payment options outside the App Store, and had to pay a 30% commission (15% for small providers) to Apple.

Read more of this story at Slashdot.

Version 2.50.0 of the Git source-code management system has been released with a long list of new user features, performance improvements, and bug fixes. See the announcement and this GitHub blog post for details.

Microsoft has disabled Windows Hello's ability to authenticate users in low-light environments through a recent security update that now requires both infrared sensors and color cameras to verify faces. The change forces the system to see a visible face through the webcam before completing authentication with IR sensors. Windows Hello earlier relied solely on infrared sensors to create 3D facial scans, allowing the feature to work in complete darkness similar to iPhone's Face ID. Microsoft pushed the dual-camera requirement to address a spoofing vulnerability in the biometric system.

Read more of this story at Slashdot.