Just another news site
Der Elektrokleinbus Kia PV5 ist mit zwei Akkugrößen lieferbar. Die maximale Reichweite liegt bei 400 km. (Elektroauto, Akku)
Der Elektrokleinbus Kia PV5 ist mit zwei Akkugrößen lieferbar. Die maximale Reichweite liegt bei 400 km. (Elektroauto, Akku) Im Handelsstreit zwischen China und den USA gibt es eine vorläufige Einigung. Doch Peking behält ein wichtiges Druckmittel in der Hand. (Seltene Erden, Donald Trump)
Im Handelsstreit zwischen China und den USA gibt es eine vorläufige Einigung. Doch Peking behält ein wichtiges Druckmittel in der Hand. (Seltene Erden, Donald Trump) Ansible ist eines der flexibelsten Tools zur Automatisierung von IT-Infrastrukturen. Dieser Workshop zeigt, wie sich Systemkonfiguration, Deployment und Administration durch Ansible effizienter gestalten lassen. (Golem Karrierewelt, Server-Applikationen)
Ansible ist eines der flexibelsten Tools zur Automatisierung von IT-Infrastrukturen. Dieser Workshop zeigt, wie sich Systemkonfiguration, Deployment und Administration durch Ansible effizienter gestalten lassen. (Golem Karrierewelt, Server-Applikationen) Robert Malone and Martin Kulldorff are two of the most concerning picks.
Anti-vaccine advocate and current health secretary Robert F. Kennedy Jr. took to social media Wednesday to announce the names of eight people he is appointing to a critical federal vaccine advisory committee—which is currently empty after Kennedy abruptly fired all 17 previous members Monday.
In the past, the vetting process for appointing new members to the Centers for Disease Control and Prevention's Advisory Committee on Immunization Practices (ACIP) could take years. But Kennedy has taken just two days.
The panel, typically stocked with vaccine, infectious disease, and public health experts, carefully and publicly reviews, analyzes, and debates vaccine data and offers recommendations to the CDC via votes. The CDC typically adopts the recommendations, which set clinical practices nationwide and determine insurance coverage for vaccinations.
Federighi talks to Ars about why the iPad’s Mac-style multitasking took so long.
CUPERTINO, Calif.—When Apple Senior Vice President of Software Engineering Craig Federighi introduced the new multitasking UI in iPadOS 26 at the company's Worldwide Developers Conference this week, he did it the same way he introduced the Calculator app for the iPad last year, or timers in the iPad's Clock app the year before—with a hint of sarcasm.
"Wow," Federighi enthuses in a lightly exaggerated tone about an hour and 19 minutes into a 90-minute presentation. "More windows, a pointier pointer, and a menu bar? Who would've thought? We've truly pulled off a mind-blowing release!"
This elicits a sensible chuckle from the gathered audience of developers, media, and Apple employees watching the keynote on the Apple Park campus, where I have grabbed myself a good-not-great seat to watch the largely pre-recorded keynote on a gigantic outdoor screen.
Jensen Huang bringt die AI Autobahn, ein KI-Rechenzentrum mit 10.000 Nvidia Blackwell-GPUs nach Deutschland. Beim Treffen mit Kanzler Merz am Freitag dürfte er mehr verraten. (Nvidia, KI)
Puzzle-based experiments reveal limitations of simulated reasoning, but others dispute findings.
In early June, Apple researchers released a study suggesting that simulated reasoning (SR) models, such as OpenAI's o1 and o3, DeepSeek-R1, and Claude 3.7 Sonnet Thinking, produce outputs consistent with pattern-matching from training data when faced with novel problems requiring systematic thinking. The researchers found similar results to a recent study by the United States of America Mathematical Olympiad (USAMO) in April, showing that these same models achieved low scores on novel mathematical proofs.
The new study, titled "The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity," comes from a team at Apple led by Parshin Shojaee and Iman Mirzadeh, and it includes contributions from Keivan Alizadeh, Maxwell Horton, Samy Bengio, and Mehrdad Farajtabar.
The researchers examined what they call "large reasoning models" (LRMs), which attempt to simulate a logical reasoning process by producing a deliberative text output sometimes called "chain-of-thought reasoning" that ostensibly assists with solving problems in a step-by-step fashion.
Last year’s data plus plenty of simulation are meant to create a level playing field.
This coming weekend will see the annual 24 Hours of Le Mans take place in France. In total, 62 cars will compete, split into three different classes. At the front of the field are the very fastest hypercars—wickedly fast prototypes that are also all hybrids, with the exception of the V12 Aston Martin Valkyries. In the middle are the pro-am LMP2s, followed by 24 GT3 cars—modified versions of performance cars that include everything from Ford Mustangs to McLaren 720s. It is racing nirvana. But with so many different makes and models of cars in the Hypercar class, some two-wheel drive, others with all-wheel drive, how do they ensure it's a fair race?
Sports car racing can be (needlessly) complicated at times. Take the Hypercar class at Le Mans. The 21 cars that will contest it are actually built to two separate rulebooks.
One, called LMH (for Le Mans Hypercar), was written by the organizers of Le Mans and the World Endurance Championship. These prototypes can be hybrids, with the electric motor on the front axle: Ferrari, Peugeot, and Toyota have all taken this route. But they don't have to be; the Aston Martin Valkyrie already had to lose a lot of power to meet the rules, so it just relies on its big V12 to do all the work. Most of the cars are purpose-built for the race, but Aston Martin went the other route and converted a road car for racing.
The test grew out of a discussion at Wikimedia’s 2024 conference.
Generative AI is permeating the Internet, with chatbots and AI summaries popping up faster than we can keep track. Even Wikipedia, the vast repository of knowledge famously maintained by an army of volunteer human editors, is looking to add robots to the mix. The site began testing AI summaries in some articles over the past week, but the project has been frozen after editors voiced their opinions. And that opinion is: "yuck."
The seeds of this project were planted at Wikimedia's 2024 conference, where foundation representatives and editors discussed how AI could advance Wikipedia's mission. The wiki on the so-called "Simple Article Summaries" notes that the editors who participated in the discussion believed the summaries could improve learning on Wikipedia.
According to 404 Media, Wikipedia announced the opt-in AI pilot on June 2, which was set to run for two weeks on the mobile version of the site. The summaries appeared at the top of select articles in a collapsed form. Users had to tap to expand and read the full summary. The AI text also included a highlighted "Unverified" badge.
The Lenovo ThinkStation P3 Tiny Gen 2 is a desktop computer that measures just 179 x 183 x 37mm (7″ x 7.2″ x 1.5″) and weighs just 1.4 kg (3 pounds). But while most mini PCs feature mobile chips designed for laptops, Lenovo’s se…
The Lenovo ThinkStation P3 Tiny Gen 2 is a desktop computer that measures just 179 x 183 x 37mm (7″ x 7.2″ x 1.5″) and weighs just 1.4 kg (3 pounds). But while most mini PCs feature mobile chips designed for laptops, Lenovo’s second-gen P3 Tiny supports up to a 65 watt Intel Core Ultra 9 […]
The post Lenovo ThinkStation P3 Tiny Gen 2 is a mini PC with up to Core Ultra 9 285 and NVIDIA RTX A1000 appeared first on Liliputing.
Jensen Huang bringt die AI Autobahn, ein KI-Rechenzentrum mit 10.000 Nvidia Blackwell-GPUs nach Deutschland. Beim Treffen mit Kanzler Merz am Freitag dürfte er mehr verraten. (