Spanish startup FOSSA Systems "has raised about $10.5 million to expand its connectivity constellation," reports Space News, noting some funding is backed by Spain's government: The support from the Spanish Society for Technological Transformation (SETT) comes a year after the fund injected 14 million euros into Spain's Sateliot , which is also developing a satellite connectivity network with security and defense applications. Spanish private investment firm Kibo Ventures led FOSSA's funding round, the six-year-old venture announced June 24, bringing its total raised to date to nearly 20 million euros. The proceeds will help fuel FOSSA's push beyond the tiny picosatellites it once used to connect low-power monitoring devices toward larger cubesats in low Earth orbit, enabling additional sovereign communications and space-based intelligence capabilities... The company's funding round follows a wave of investments this year in European ventures planning to develop sovereign space capabilities, including Austrian propulsion startup Gate Space, which secured 6.3 million euros earlier this month from a European Commission-backed accelerator program. "Our goal is to establish FOSSA as a European benchmark in sovereign space infrastructure," said Julián Fernández, FOSSA's CEO and cofounder.

Read more of this story at Slashdot.

Chinese AI systems "have matched the performance of Anthropic's powerful model Mythos in some cybersecurity scenarios," reports the Wall Street Journal. They call it "a development poised to reset the global tech race and pressure the White House in its overhaul of U.S. AI policy." Security researchers said that a new AI model, released this month by China's Zhipu AI, also known as Z.ai, can match the latest U.S. models when it comes to finding security bugs, although it still lags behind Anthropic's and OpenAI's products in other tasks. Overall, the capability gap between top U.S. models and those built by Chinese companies has narrowed significantly, and use of Chinese AI systems has surged as businesses seek to rein in runaway costs. A host of companies, including Microsoft, are weighing how they can offer Chinese models on their platforms, a development that is set to alter the balance of power among tech companies... Unlike models from Anthropic or OpenAI, Zhipu's GLM-5.2 is open-weight. That means it can be downloaded and run on hardware operated by anybody and can be modified and used without supervision. Open-weight models are ideal for users who want unfettered access to systems they control, but they are also ideal for hackers, who can run them in the shadows. GLM-5.2 has ranked as one of the 10 most-used AI models, according to data from OpenRouter, a company that provides access to more than 400 AI models. In some benchmarking tests, according to the cybersecurity company Semgrep, GLM-5.2 bested Anthropic's Claude Opus 4.8 model, which was released in May. When given further instructions, Opus 4.8 and GLM-5.2 can match Mythos in bug-finding ability, according to researchers... "Banning Fable while selling chips China needs to develop its own version is a gift to China," said Saif Khan, a distinguished technology fellow at the Institute for Progress think tank who worked on export restrictions in the Biden administration. The U.S. needs to maximize the use of Mythos and comparable models to harden its cyber defenses while it can, he added. Among the Mythos 5 and Fable 5 users that had lost access before Friday's decision to restore Mythos 5 access for some trusted entities: the National Security Agency, which had been testing the tools and found them impressive in trials, according to people familiar with the matter... "It is incentivizing companies across the globe to use cheaper but very capable Chinese open-weight models, while at the same time undermining the U.S. AI industry," said Niels Provos, a researcher who led security teams at Google and Stripe. "I don't understand it." Thanks to long-time Slashdot reader schwit1 for sharing the article.

Read more of this story at Slashdot.

As expected, Linux 7.2-rc1 was released a brief time ago to cap off the Linux 7.2 merge window. Now it's off for eight weeks or so of testing before Linux 7.2 stable is released that will in turn go on to power the likes of Fedora 45 and Ubuntu 26.10...

The New York Times tells the story of a 63-year-old retiree who wrote a check for several thousand dollaras to pay her taxes. But she discovered much later that her taxes were never paid because that check had been intercepted and then altered to be payable to someone else: In some cases, thieves may pilfer one or more checks from local mailboxes. Adam Rust, director of financial services for the Consumer Federation of America, said thieves sometimes "fish" for checks at free-standing drop boxes, using long tools with sticky pads on the ends to grab letters. In other cases, more sophisticated criminals may steal large batches of checks, copy them and then sell them on the internet. Often, the purloined checks are chemically altered in what's known as "check washing" to remove the name of the recipient. The thief replaces it with a fraudulent name, and often increases the amount of the check, before cashing or depositing it. The 63-year-old retiree's bank told her she'd waited too long to recover the funds: Schwab's "security guarantee," outlined on its website , says that "Schwab will cover losses in any of your Schwab accounts due to unauthorized activity." But fine print at the bottom of the page notes that reimbursement "requires your timely reporting of unauthorized activity to Schwab," and that Schwab "will not be liable for additional or increased losses resulting from a failure to report unauthorized activity in a timely manner." It notes that more details are available in account agreements... Notify your bank as soon as possible, said Scott Anchin, senior vice president of strategic initiatives and policy at the independent bankers association. Banks generally allow at least 30 days and sometimes up to 90 days from the time your statement is made available to you to report suspected check fraud, he said. So how can you avoid check fraud? Adam Rust, director of financial services for the Consumer Federation of America, just suggests that "No one should ever mail a check." If you must write a check, he said, try to deliver it in person or take it inside a post office to mail rather than relying on your own mailbox or public drop boxes. The American Bankers Association recommends using permanent "gel" ink pens when you do write checks to reduce the risk of tampering... And if you don't already, consider using your bank's online bill payment service. The article notes that even the U.S. federal government "has been moving away from paper checks for things like benefit payments and income tax refunds, saying digital payment methods are more secure."

Read more of this story at Slashdot.

With Q2'2026 working toward a close, here is a look back at the most popular Intel Linux and open-source news for the quarter...

America's Bureau of Alcohol, Tobacco, Firearms and Explosives has "canceled its contract for a surveillance tool that enables warrantless tracking of mobile devices," reports the Associated Press. They note the move comes "after lawmakers, a prosecutor and a judge raised concerns about the legality of the tool in criminal investigations." ATF, the federal agency responsible for enforcing the nation's gun laws, told The Associated Press that it discontinued what it called a "pilot" program using a tool called Webloc after Rep. Michael Cloud, a Republican from Texas, and Sen. Ron Wyden, a Democrat from Oregon, expressed reservations about the agency's use of bulk commercial location data. Webloc, which is made by a vendor called Penlink, sources data from consumer apps and advertising networks, which collect the location of mobile devices from consumers who download apps or browse the web... The U.S. Supreme Court ruled in 2018 that police needed a warrant to obtain historic movement data from cellphone companies on a criminal suspect. But it has never addressed the growing practice of commercially acquired data. Other users of Webloc include the U.S. military and U.S. Immigration and Customs Enforcement but also local law enforcement agencies such as police in places like Elk Grove, Calif. and Durham, N.C. The technology has also expanded around the world, with the national police in El Salvador and Hungarian intelligence agencies as customers, according to a report from earlier this year from Citizen Lab, a group of researchers at the University of Toronto who investigate digital threats to civil society. The article notes that other U.S. law enforcement agencies continue to buy commercial geolocation data, "including the FBI and the Department of Homeland Security."

Read more of this story at Slashdot.

Students are using AI-powered smart glasses to cheat on tests, reports CNN. "And in East Asia's test-obsessed societies, where a single exam could impact the trajectory of a student's future career and social status, educators are scrambling to get ahead of the problem." Already, countries are stepping up inspections for test-takers. For China's grueling annual college entrance exam earlier this month — which more than 10 million hopefuls take each year — authorities required screening of all glasses. In the United Kingdom, the head of England's exam watchdog warned earlier this month that AI glasses and smart devices like earpieces could worsen cheating in exams... [T]wo incidents in South Korea were the country's first reported cases of cheating with AI glasses... In Taiwan, the university where a prospective student was caught cheating is now reviewing rules and standard operating procedures for AI eyewears during examinations. But experts worry these individual cases point to a more widespread issue. "If we're seeing a few cases being reported, we're seeing a lot more cases not being reported," said Thomas Corbin, lecturer at Deakin University in Australia, who has conducted research around the usage of AI-powered glasses and other smart devices in academic assessment. With the rapid development of AI technology, however, smart glasses are becoming slimmer, less noticeable, while integrating AI models that can operate independently with connectivity, raising concerns not only about exam integrity, but also about broader privacy risks... "Wearable AI is as much of a challenge to exams as ChatGPT was to essays in 2022 and I just don't think there is any real way that we can reliably have exam practices moving forward," Corbin said.

Read more of this story at Slashdot.

The Onion joked the new movie Supergirl is about a hero who must single-handedly save the world "after the catastrophic collapse of interest in the genre." Unfortunately, The Hollywood Reporter says the film's reviews "range from negative to tepid praise (averaging a 58 percent Rotten Tomatoes score)." Many point fingers at the film's script, with Variety's line — "a comic-book movie with the worst script I can remember" — going viral... Not to pile on, but there's another recurring gripe from the reviews that stood out: Critics bashed the film as being murky, dark and gray, with poor VFX: "Muddy CG sludge" wrote one. Another said the film was full of "sludgy browns and grays" and "the visual murkiness of the settings makes it hard to follow the already unintelligible action sequences." A third wrote the "VFX is so rough it makes The Flash look like Avatar." Moviegoers increasingly despise murky, dark visuals (often used to hide weak effects), along with obvious CGI and incoherent action. They've seen it so many times they've become allergic. The Bulwark agrees that the action sequences are "terribly lit, incoherently staged, and just generally weightless and ugly... [I]t's reminiscent of the disaster that was The Flash: It's just very obvious during certain sequences that everyone was in a big green-screen warehouse and the camera was whipping around with the knowledge that everything would be painted in later, so who really gives a crap how anything looks on the day of." But they also call the movie "a tremendous slog of a film, a real step backwards for the James Gunn-overseen DC Universe of movies and TV shows" that's "neither fun nor exciting" and "feels empty." The film does have one bright spot: Lobo, who is played by Jason Momoa as something like Michael Keaton's Beetlejuice by way of Jason Momoa's Aquaman. He's blustery and cantankerous and saucy and just a little menacing; it's a perfect piece of casting and a really nice performance. Unfortunately, it's the only spark of life in what is otherwise a deeply dour, deeply boring piece of filmmaking... Supergirl is just a misfire on nearly every level, one that lacks the sincerity and fun of last year's reboot of this universe or the comic pathos present in Gunn's Peacemaker series on HBO Max. Reason calls it "dark, depressive, and dull" and "a downer of a movie in nearly every way." It's not fun. It's barely even righteous. It's just miserable. At one point, Supergirl flat-out murders a guy by pushing a giant sword through his neck. Somehow, I suspect even Zack Snyder would be appalled. Time argued fans of last decade's superhero movies "should be demanding more, not less." Though "Will there be rioting in the streets once audiences get some idea of how lousy Supergirl is? Probably not."

Read more of this story at Slashdot.

CachyOS is out today with a new feature release for this Arch Linux powered distribution that delivers stellar out-of-the-box performance...

While the Linux 7.2 kernel merge window is only ending later today to cap off the feature work on this next version of the Linux kernel, already for the Linux 7.3 kernel cycle later in the year there is one notable feature on the way: the DRM color format property is being introduced and being first supported by the AMDGPU kernel graphics driver...

With Q2'2026 quickly coming to an end, here's a recap of the most popular Linux kernel news over the past three months on Phoronix...

Today marks the last day of the Linux 7.2 merge window with Linux 7.2-rc1 due out later today. With the many new features and improvements merged over the past week since the Linux 7.1 stable debut, the Linux kernel source tree now exceeds 43 million lines...

For those using Vim with its GTK3 toolkit interface on Wayland, it soon should be delivering much better performance with pending patches...

Comme chaque semaine, nous faisons le point sur l'évolution des prix des processeurs Intel et AMD chez nos revendeurs partenaires. Si le marché des cartes graphiques reste relativement calme, celui des processeurs continue de connaître des mouvements parfois très importants, toujours dans un contexte où la flambée des prix de la mémoire pèse sur l'ensemble de l'industrie. Certaines références profitent encore de promotions intéressantes, tandis que d'autres enregistrent des hausses particulièrement marquées. Intel baisse presque partout, AMD en ordre dispersé Chez Intel, la tendance est globalement positive cette semaine. Le Core i5-14600K reste parfaitement stable à 287,90 U+20AC, tandis que le Core i7-14700K retrouve des couleurs en repassant de 425,90 U+20AC à 406,90 U+20AC, soit une baisse de 19 U+20AC. Le Core i9-14900K, de son côté, ne bouge absolument pas et reste affiché à 512,90 U+20AC. La famille Core Ultra est également orientée à la baisse. Le Core Ultra 5 245K perd 19 U+20AC pour revenir à 180,90 U+20AC, le Core Ultra 5 250K+ abandonne 6 U+20AC et passe à 220,90 U+20AC, tandis que le Core Ultra 7 265K demeure stable à 311,90 U+20AC. La plus forte baisse est à mettre au crédit du Core Ultra 7 270K+, qui chute de 331,90 U+20AC à seulement 300,90 U+20AC, soit 31 U+20AC de moins en une semaine. Enfin, le Core Ultra 9 285K recule légèrement de 11 U+20AC et s'affiche désormais à 542,90 U+20AC. […]

Lire la suite

"Enterprises may soon be paying as much for their developers' AI token usage as they do for their salaries," writes InfoWorld: According to Gartner, these costs will meet, or even exceed, the typical software engineer's monthly salary within the next two years. This is not only because developers are increasingly adopting generative AI and agentic tools, it reflects a trend toward consumption-based licensing models as vendors balance infrastructure investments with profitability... Gartner senior principal analyst Nitish Tyagi explained that it's important to note that Gartner's prediction is based on a global average salary of $2,000 per month; it doesn't mean AI token usage will exceed all salaries. For instance, in the US, yearly pay rates can be six digits or more. However, that kind of spend is not out of the realm of possibility, Tyagi emphasized. "I have heard scary numbers like 'My developer consumed $20K last month,' or 'A business user consumed $32K'." If these amounts sound shocking, that's the point. "The goal is to alarm the industry about the impact of token cost if it is not governed and controlled," he said... AI coding vendors have yet to deliver "mature, built-in cost optimization capabilities," Tyagi said, and prices will likely only continue to rise as vendors further build out their models while at the same time trying to remain profitable. Thus, enterprises struggle to forecast and control costs, and, because AI is moving so fast, many organizations lack the "maturity and frameworks" to determine ROI, he noted. Agent-driven workflows are difficult to govern, context windows become bloated, budgets are wiped out earlier than anticipated, and token spend becomes hard to justify.... "Without a governed engineering operating model, costs can escalate faster than the productivity gains these tools are designed to deliver," Tyagi said.

Read more of this story at Slashdot.

Avec plusieurs dizaines de millions de titres disponibles, la différence entre les grands services de streaming musical ne se joue plus sur la taille du catalogue. En 2026, ce qui distingue vraiment Spotify, Apple Music, Deezer et YouTube Music, c’est d’abord la qualité audio et le catalogue francophone. Ensuite, les options famille, la compatibilité et […]

L'article Comparatif streaming musical 2026 : Spotify, Apple Music ou Deezer ? a été publié en premier sur Bbox-Mag

Quelle semaine ! Avec les cerveaux qui chauffent autant que nos composants, les jours furent longs. Les températures étant sensiblement plus clémentes, il est temps de faire un point sur les tests et autres articles publiés dernièrement. Et c'est plutôt calme, mais avec des composants qui sont particulièrement de saison. […]

Lire la suite

Jack Nekhala had a business selling on Amazon — and in December he received an unusual offer, reports Bloomberg. A woman said she could bribe an Amazon employee "to help him retrieve $90,000 in funds that the e-commerce giant had frozen after suspending him over an alleged violation of review policy." Hoping to ingratiate himself with the company and restart his business, Nekhala offered to provide evidence, including recorded conversations and screen shots, that he said proved Amazon personnel were peddling inside information and influence. The smoking gun, Nekhala told the representative: information about his seller account. Only certain Amazon employees are supposed to have access to such details, but Nekhala had received them from the woman on WeChat, the Chinese messaging app. Nekhala's experience, which he documented and shared with Bloomberg, provides a rare glimpse into an international black market that has been a persistent scourge of Amazon's online store. On one side are sellers looking for a variety of favors: a competitive edge over their rivals, information on how to boost sales, a way to get themselves unsuspended. On the other are middlemen who lurk on message apps like Telegram, WeChat and WhatsApp offering access to people inside Amazon who can get things done for a price... It's impossible to determine the scope of the illicit activity, but it's an open secret among Amazon sellers and consultants, who are frequently approached on social-media platforms and messaging apps. "The message is always the same: 'I'm going to show you screenshots to prove I have inside access,'" said Chris McCabe, a former Amazon employee who runs a seller consulting firm... In 2020, federal prosecutors exposed an international bribery scheme involving Amazon sellers and employees. The ring allegedly extracted about $100 million in unfair advantages by bribing Amazon employees in Asia to help them sell more products and sabotage their competitors. Five people in the US were convicted and received jail terms or probation. Last year, law enforcement officials in India began investigating more than 20 former Amazon employees suspected of accepting bribes from trucking companies in exchange for routes, according to The Times of India. After Nekhala reported his own experience to Amazon, the representative committed to "do some digging" and to email him instructions on how his evidence could be shared, according to a recording of the conversation. But Nekhala said he never heard back. The employee who leaked his personal information had already been fired for unrelated misconduct, according to Amazon. Amazon told Bloomberg employee involvement was "very rare," and that "We invest heavily in this area and have dedicated teams and systems in place to prevent all types of fraud, including by our own employees."

Read more of this story at Slashdot.

Comme chaque dimanche, nous prenons quelques minutes pour admirer le travail des moddeurs du monde entier. Et cette semaine, nous retrouvons un nom que les habitués de la rubrique connaissent déjà bien : AK MOD. Le moddeur nous propose cette fois-ci un projet baptisé PROJECT 2026 ROG TIMEFUSION, une machine présentée lors du Computex 2026 et construite autour d'un composant particulièrement symbolique : la nouvelle carte mère ROG Crosshair 2006. Car oui, tout le concept du mod repose sur cette référence un peu particulière. ASUS a décidé de célébrer vingt ans de la gamme Crosshair avec cette édition spéciale, et AK MOD s'est amusé à imaginer ce que pourrait donner une rencontre entre un PC haut de gamme de 2006 et une configuration de 2026. Un exercice de style qui mélange nostalgie informatique et matériel ultra moderne. […]

Lire la suite

IBM spent a decade "building, testing and improving" quantum computing, reports the Wall Street Journal. "This year, the company is laying the groundwork to turn that technology into a fully-fledged, scalable business from an expensive science project." IBM said last month it plans to form a new independent subsidiary called Anderon, a foundry to produce the silicon wafers needed to make quantum-computing processors. The venture is seeded by a $1 billion investment from the Trump administration and another $1 billion of IBM's own cash. Anderon will give the company a new line of business in selling wafers to other quantum-computing companies. It will also provide a steady stream of wafers to continue developing its own quantum technology, positioning IBM to capture part of what the Boston Consulting Group projects will be a $90 billion to $170 billion market for quantum-computing providers by 2040... The company also plans to spend an additional $9 billion over five years to advance the final stages of its quest to build a quantum-mechanics-powered computer capable and reliable enough for widespread use, a goal known as fault tolerance. That computer, named Starling, is being targeted for 2029. With Anderon, IBM is thinking beyond Starling, or even a more powerful quantum computer planned for 2033.

Read more of this story at Slashdot.