Slashdot

An anonymous reader quotes a report from Ars Technica: The Federal Communications Commission today voted to lower price caps on prison phone calls and closed a loophole that allowed prison telecoms to charge high rates for intrastate calls. Today's vote will cut the price of interstate calls in half and set price caps on intrastate calls for the first time. The FCC said it "voted to end exorbitant phone and video call rates that have burdened incarcerated people and their families for decades. Under the new rules, the cost of a 15-minute phone call will drop to $0.90 from as much as $11.35 in large jails and, in small jails, to $1.35 from $12.10." The new rules are expected to take effect in January 2025 for all prisons and for jails with at least 1,000 incarcerated people. The rate caps would take effect in smaller jails in April 2025. Worth Rises, a nonprofit group advocating for prison reform, said it "estimates that the new rules will impact 83 percent of incarcerated people (about 1.4 million) and save impacted families at least $500 million annually." The nonprofit Prison Policy Institute said that prison phone companies charge ancillary fees for things "like making a deposit to fund an account." The ban on those fees "also effectively blocks a practice that we have been campaigning against for years: companies charging fees to consumers who choose to make single calls rather than fund a calling account, and deliberately steering new consumers to this higher-cost option in order to increase fee revenue," the group said. The ancillary fee ban is a "technical-sounding change," but will help "eliminate some of the industry's dirtiest tricks that shortchange both the families and the facilities," the group said.

Read more of this story at Slashdot.

OpenAI on Thursday launched GPT-4o mini, a cost-efficient small AI model that will replace GPT-3.5 Turbo in ChatGPT. Reuters reports: Priced at 15 cents per million input tokens and 60 cents per million output tokens, the GPT-4o mini is more than 60% cheaper than GPT-3.5 Turbo, OpenAI said. It currently outperforms the GPT-4 model on chat preferences and scored 82% on Massive Multitask Language Understanding (MMLU), OpenAI said. MMLU is a textual intelligence and reasoning benchmark used to evaluate the capabilities of language models. A higher MMLU score signifies it can understand and use language better across a variety of domains, enhancing real-world usage. The GPT-4o mini model's score compared with 77.9% for Google's Gemini Flash and 73.8% for Anthropic's Claude Haiku, according to OpenAI. With the mini model currently supporting text and vision in the application programming interface, OpenAI said support for text, image, video and audio inputs and outputs would be made available in the future.

Read more of this story at Slashdot.

An anonymous reader quotes a report from TechCrunch: The U.S. Postal Service was sharing the postal addresses of its online customers with advertising and tech giants Meta, LinkedIn and Snap, TechCrunch has found. On Wednesday, the USPS said it addressed the issue and stopped the practice, claiming that it was "unaware" of it. TechCrunch found USPS was sharing customers' information by way of hidden data-collecting code (also known as tracking pixels) used across its website. Tech and advertising companies create this kind of code to collect information about the user -- such as which pages they visit -- every time a webpage containing the code loads in the customer's browser. In the case of USPS, some of that collected data included the postal addresses of logged-in USPS Informed Delivery customers, who use the service to see photos of their incoming mail before it arrives. It's not clear how many individuals had their information collected or for how long. Informed Delivery had more than 62 million users (PDF) as of March 2024. [...] The code also collected other data, such as information about the user's computer type and browser, which appeared as partly pseudonymized -- essentially scrambled in a way that makes it more difficult for humans to know where data came from, or who it relates to, by using randomized identifiers in place of real customer names. But researchers have long warned that pseudonymous data can still be used to re-identify seemingly anonymous individuals. TechCrunch also found that tracking numbers entered into the USPS website were also shared with advertisers and tech companies, including Bing, Google, LinkedIn, Pinterest and Snap. Some in-transit tracking data was also shared, such as the real-world location of the mail in the postal system, even if the customer was not logged in to USPS' website. USPS spokesperson Jim McKean said in a statement: "The Postal Service leverages an analytics platform for our own internal purposes, so that we understand the usage of our products and services and which we use on an aggregated basis to market our products. The Postal Service does not sell or provide any personal information that is collected from this analytics platform to any third party, and we were unaware of any configuration of the platform that collected personal information from the URL and that shared it without our knowledge with social media." "We have taken immediate action to remediate this issue," the spokesperson added, without saying what action was taken.

Read more of this story at Slashdot.

RealPage says it isn't doing anything wrong by suggesting to landlords how much rent they could charge. From a report: In a move to reclaim its own narrative, the property management software company published a microsite and a digital booklet it's calling "The Real Story," as it faces multiple lawsuits and a reported federal criminal probe related to allegations of rental price fixing. RealPage's six-page digital booklet, published on the site in mid-June, addresses what it calls "false and misleading claims about its software" -- the myriad of allegations it faces involving price-fixing and rising rents -- and contends that the software benefits renters and landlords and increases competition. It also said landlords accept RealPage's price recommendations for new leases less than 50 percent of the time and that the software recommends competitive prices to help fill units. [...] But landlords are left without concrete answers, as questions around the legality of this software are ongoing as they continue renting properties. "I don't think we're seeing this as a RealPage issue but rather as a revenue management software issue," says Alexandra Alvarado, the director of marketing and education at the American Apartment Owners Association, the largest association of landlords in the US. Alvarado says some landlords are taking pause and asking questions before using the tech.

Read more of this story at Slashdot.

The energy giant Shell has quietly backed away from a pledge to rapidly increase its use of "advanced recycling," a practice oil and petrochemical producers have promoted as a solution to the plastics pollution crisis. From a report: "Advanced" or "chemical" recycling involves breaking down plastic polymers into tiny molecules that can be made into synthetic fuels or new plastics. The most common form, pyrolysis, does so using heat. Shell has invested in pyrolysis since 2019, touting it as a way to slash waste. That year, the company used oil made via pyrolysis in one of its Louisiana chemical plants for the first time. And it began publicizing a new goal for the technology: "Our ambition is to use 1m tonnes of plastic waste a year in our global chemicals plants by 2025." But recently, the company rolled back that promise with little fanfare: "[I]n 2023 we concluded that the scale of our ambition to turn 1m tonnes of plastic waste a year into pyrolysis oil by 2025 is unfeasible," it said in its 2023 sustainability report, published in March. Reached for comment, a Shell spokesperson, Curtis Smith, said: "Our ambition, regardless of regulation, is to increase circularity and move away from a linear economy to one where products and materials are reused, repurposed and recycled."

Read more of this story at Slashdot.

Meta says it won't be launching its upcoming multimodal AI model -- capable of handling video, audio, images, and text -- in the European Union, citing regulatory concerns. From a report: The decision will prevent European companies from using the multimodal model, despite it being released under an open license. Just last week, the EU finalized compliance deadlines for AI companies under its strict new AI Act. Tech companies operating in the EU will generally have until August 2026 to comply with rules around copyright, transparency, and AI uses like predictive policing. Meta's decision follows a similar move by Apple, which recently said it would likely exclude the EU from its Apple Intelligence rollout due to concerns surrounding the Digital Markets Act.

Read more of this story at Slashdot.

Nvidia and French startup Mistral AI jointly announced today the release of a new language model designed to bring powerful AI capabilities directly to business desktops. From a report: The model, named Mistral-NeMo, boasts 12 billion parameters and an expansive 128,000 token context window, positioning it as a formidable tool for businesses seeking to implement AI solutions without the need for extensive cloud resources. Bryan Catanzaro, vice president of applied deep learning research at Nvidia, emphasized the model's accessibility and efficiency in a recent interview with VentureBeat. "We're launching a model that we jointly trained with Mistral. It's a 12 billion parameter model, and we're launching it under Apache 2.0," he said. "We're really excited about the accuracy of this model across a lot of tasks." The collaboration between Nvidia, a titan in GPU manufacturing and AI hardware, and Mistral AI, a rising star in the European AI scene, represents a significant shift in the AI industry's approach to enterprise solutions. By focusing on a more compact yet powerful model, the partnership aims to democratize access to advanced AI capabilities. Catanzaro elaborated on the advantages of smaller models. "The smaller models are just dramatically more accessible," he said. "They're easier to run, the business model can be different, because people can run them on their own systems at home. In fact, this model can run on RTX GPUs that many people have already."

Read more of this story at Slashdot.

Following a comprehensive internal review, NASA announced Wednesday its intent to discontinue development of its VIPER (Volatiles Investigating Polar Exploration Rover) project. NASA: NASA stated cost increases, delays to the launch date, and the risks of future cost growth as the reasons to stand down on the mission. The rover was originally planned to launch in late 2023, but in 2022, NASA requested a launch delay to late 2024 to provide more time for preflight testing of the Astrobotic lander. Since that time, additional schedule and supply chain delays pushed VIPER's readiness date to September 2025, and independently its CLPS (Commercial Lunar Payload Services) launch aboard Astrobotic's Griffin lander also has been delayed to a similar time. Continuation of VIPER would result in an increased cost that threatens cancellation or disruption to other CLPS missions. NASA has notified Congress of the agency's intent.

Read more of this story at Slashdot.

An anonymous reader shares a report: Indian crypto exchange WazirX on Thursday confirmed it had suffered a security breach after about $230 million in assets were "suspiciously transferred" out of the platform earlier in the day. The Mumbai-based firm said one of its multisig wallets had suffered a security breach, and it was temporarily pausing all withdrawals from the platform. Lookchain, a third-party blockchain explorer, reported that more than 200 cryptocurrencies, including 5.43 billion SHIB tokens, over 15,200 Ethereum tokens, 20.5 million Matic tokens, 640 billion Pepe tokens, 5.79 million USDT, and 135 million Gala tokens were "stolen" from the platform. WazirX reported holdings of about $500 million in its June proof-of-reserves disclosure.

Read more of this story at Slashdot.

An anonymous reader quotes a report from Reuters: Nearly a quarter of Japanese companies have adopted artificial intelligence (AI) in their businesses, while more than 40% have no plan to make use of the cutting-edge technology, a Reuters survey showed on Thursday. The survey, conducted for Reuters by Nikkei Research, pitched a range of questions to 506 companies over July 3-12 with roughly 250 firms responding, on condition of anonymity. About 24% of respondents said they have already introduced AI in their businesses and 35% are planning to do so, while the remaining 41% have no such plans, illustrating varying degrees of embracing the technological innovation in corporate Japan. Asked for objectives when adopting AI in a question allowing multiple answers, 60% of respondents said they were trying to cope with a shortage of workers, while 53% aimed to cut labour costs and 36% cited acceleration in research and development. As for hurdles to introduction, a manager at a transportation company cited "anxiety among employees over possible headcount reduction." Other obstacles include a lack of technological expertise, substantial capital expenditure and concern about reliability, the survey showed.

Read more of this story at Slashdot.

A drug has been shown to extend the lifespan of laboratory mice by nearly 25%, with treated mice displaying fewer cancers and improved health and strength. It earned them the nickname "supermodel grannies" due to their youthful appearance. "The drug is already being tested in people, but whether it would have the same anti-ageing effect is unknown," reports the BBC. From the report: The team at the MRC Laboratory of Medical Science, Imperial College London and Duke-NUS Medical School in Singapore were investigating a protein called interleukin-11. Levels of it increase in the human body as we get older, it contributes to higher levels of inflammation, and the researchers say it flips several biological switches that control the pace of ageing. The researchers performed two experiments. The first genetically engineered mice so they were unable to produce interleukin-11. The second waited until mice were 75 weeks old (roughly equivalent to a 55-year-old person) and then regularly gave them a drug to purge interleukin-11 from their bodies. The results, published in the journal Nature, showed lifespans were increased by 20-25% depending on the experiment and sex of the mice. Old laboratory mice often die from cancer, however, the mice lacking interleukin-11 had far lower levels of the disease. And they showed improved muscle function, were leaner, had healthier fur and scored better on many measures of frailty.

Read more of this story at Slashdot.

Astronomers say they've detected two gases that could indicate the presence of life forms lurking in the clouds of Venus. The Guardian reports: Findings presented at the national astronomy meeting in Hull on Wednesday bolster evidence for a pungent gas, phosphine, whose presence on Venus has been fiercely disputed. A separate team revealed the tentative detection of ammonia, which on Earth is primarily produced by biological activity and industrial processes, and whose presence on Venus scientists said could not readily be explained by known atmospheric or geological phenomena. "It could be that if Venus went through a warm, wet phase in the past then as runaway global warming took effect [life] would have evolved to survive in the only niche left to it -- the clouds," said Dr Dave Clements, a reader in astrophysics at Imperial College London, told the meeting. "Our findings suggest that when the atmosphere is bathed in sunlight the phosphine is destroyed," Clements said. "All that we can say is that phosphine is there. We don't know what's producing it. It may be chemistry that we don't understand. Or possibly life." In a second talk, Prof Jane Greaves, an astronomer at Cardiff University, presented preliminary observations from the Green Bank telescope indicating ammonia, which on Earth is made through either industrial processes or by nitrogen-converting bacteria. Greaves said: "Even if we confirmed both of these [findings], it is not evidence that we have found these magic microbes and they're living there today," adding that there were not yet "any ground truths." Dr Robert Massey, the deputy executive director at the Royal Astronomical Society, said in a statement: "These are very exciting findings but it must be stressed that the results are only preliminary and more work is needed to learn more about the presence of these two potential biomarkers in Venus's clouds. Nevertheless, it is fascinating to think that these detections could point to either possible signs of life or some unknown chemical processes. It will be interesting to see what further investigations unearth over the coming months and years."

Read more of this story at Slashdot.

An anonymous reader quotes a report from the New York Times: The image, as it happens, comes from dozens of brain scans produced by researchers at Washington University School of Medicine in St. Louis who gave psilocybin, the compound in "magic mushrooms," to participants in a study before sending them into a functional M.R.I. scanner. The kaleidoscopic whirl of colors they recorded is essentially a heat map of brain changes, with the red, orange and yellow hues reflecting a significant departure from normal activity patterns. The blues and greens reflect normal brain activity that occurs in the so-called functional networks, the neural communication pathways that connect different regions of the brain. The scans, published Wednesday in the journal Nature, offer a rare glimpse into the wild neural storm associated with mind-altering drugs. Researchers say they could provide a potential road map for understanding how psychedelic compounds like psilocybin, LSD and MDMA can lead to lasting relief from depression, anxiety and other mental health disorders. "Psilocybin, in contrast to any other drug we've tested, has this massive effect on the whole brain that was pretty unexpected," said Dr. Nico Dosenbach, a professor of neurology at Washington University and a senior author of the study. "It was quite shocking when we saw the effect size." Brian Mathur, a systems neuroscientist at the University of Maryland School of Medicine in Baltimore, says these findings cannot show exactly what causes the therapeutic benefit of psilocybin, but "it's possible psilocybin is directly causing" the brain-network changes. That, or it is creating a psychedelic experience that in turn causes parts of the brain to behave differently. The next step is to determine whether psilocybin's blood-flow changes in the brain or its direct effects on neurons, or both, are responsible for the brain-network disruptions. "The best part of this work is that it's going to provide a means forward for the field to develop further hypotheses that can and should be tested," Mathur says.

Read more of this story at Slashdot.

Longtime Slashdot reader Uncle_Meataxe shares a report from the Sacramento Bee: California's power grid handled a nearly three week long record-setting heat wave with few issues. The heat wave was the hottest 20-day period on record around Sacramento and set an all-time temperature record of 124 degrees in Palm Springs. Emergency alerts and calls for voluntary conservation were avoided this time around. Officials credit years of investment in renewable energy, especially battery storage that store solar power for use when the sun stops shining. CAISO last issued calls for voluntary conservation two years ago, during a 2022 bout of extreme heat. Since then, roughly 11,600 megawatts of new renewable energy sources have come onto California's electricity grid. That includes 10,000 megawatts of battery power, enough to power 10 million homes for a few hours. California is now home to the most grid batteries in the world outside of China, [said Elliot Mainzer, president and CEO of California Independent System Operator (CAISO)]. "Batteries performed very well in this event, they were charged and ready at the right times for optimization on the grid," he added. "That made a big, big difference." [...] Apart from battery storage, Mainzer also credited that success to less extreme temperatures in Southern California as well as noticeable slightly lower electricity consumption in the peak demand hours, from 4 p.m. to 9 p.m.

Read more of this story at Slashdot.

Dr. Fei-Fei Li, the so-called "godmother of AI," is working on a startup focused on developing technology capable of human-like visual data processing and advanced reasoning. According to the Financial Times (paywalled), the startup is called World Labs and is already worth $1 billion. Quartz reports: "Curiosity urges us to create machines to see just as intelligently as we can, if not better," Li said during a Ted talk in April. "And if we want to advance AI beyond its current capabilities, we want more than AI that can see and talk. We want AI that can do." Andreessen Horowitz and the AI fund Radical Ventures are funders of World Labs. Li is renowned for her contributions to AI. She invented ImageNet, a dataset used for advancing computer vision that many see as a catalyst for the AI boom. She consults with policymakers as they work to set up guardrails for the technology, and was named one of 12 national AI research resource task force members by the U.S. White House in 2021.

Read more of this story at Slashdot.

Valve had its employee and payroll data leaked through a poorly redacted document in an antitrust lawsuit in May, offering a rare glimpse into the company's small but impactful workforce over the years. As first noticed by SteamDB's Pavel Djundik, Valve's significant influence in PC gaming transactions has been maintained by just a few hundred employees. Kyle Orland reports via Ars Technica: It's striking to consider just how small Valve is compared to other major players in the game industry. In 2021, Microsoft estimated Valve's annual revenue at $6.5 billion, roughly on the same scale as EA's $7.5 billion in 2024 revenue. But Steam achieved those numbers with around 350 employees, compared to well over 13,000 people employed by EA. The disparity highlights just how much money Valve brings in with a relatively small workforce. And a lot of that is thanks to the chunk of revenue Valve takes from every sale on Steam. The dominant PC gaming marketplace has seen a massive increase in the number of annual game releases since 2012 or so, thanks to initiatives like Steam Greenlight and Steam Direct. Yet, surprisingly, the size of the "Steam" department inside Valve has shrunk in recent years, from a peak of 142 employees in 2015 down to just 79 in 2021. From the outside, having just 79 employees keeping track of more than 11,000 Steam releases in 2021 is a pretty incredible ratio. Some readers may also be surprised that Valve's "Games" department has represented a majority of the company's headcount since 2003. That has remained true (though to a lesser extent) even in more recent years, as Valve's output of new games has become much more occasional. It seems likely a large number of those Games department employees are devoted to ultra-popular Valve games like Dota 2 and Counter-Strike 2, which enjoy tens of millions of players and need significant support work. The leaked data also shows the slow rise of Valve's small Hardware department, which started with just three employees in 2011 as the company began work on its doomed Steam Machines initiative. Transitioning into the Valve Index era in the late 2010s, the hardware department still represented just a few dozen people and a paltry 3 to 4 percent of the company's annual payroll. By the time we hit 2021 and the run-up to the Steam Deck, the Hardware division still makes up just 12 percent of Valve's small total headcount. Looking back, it's impressive that such a small team was able to create a portable gaming device that quickly spawned a whole micro-industry of imitators. We can only hope the Hardware team got a little more employee support in the wake of the Steam Deck's market success.

Read more of this story at Slashdot.

An anonymous reader quotes a report from The Atlantic: Now, after years of contentious relationships with academic researchers, Meta is opening a small pilot program that would allow a handful of them to access Instagram data for up to about six months in order to study the app's effect on the well-being of teens and young adults. The company will announce today that it is seeking proposals that focus on certain research areas -- investigating whether social-media use is associated with different effects in different regions of the world, for example -- and that it plans to accept up to seven submissions. Once approved, researchers will be able to access relevant data from study participants -- how many accounts they follow, for example, or how much they use Instagram and when. Meta has said that certain types of data will be off-limits, such as user-demographic information and the content of media published by users; a full list of eligible data is forthcoming, and it is as yet unclear whether internal information related to ads that are served to users or Instagram's content-sorting algorithm, for example, might be provided. The program is being run in partnership with the Center for Open Science, or COS, a nonprofit. Researchers, not Meta, will be responsible for recruiting the teens, and will be required to get parental consent and take privacy precautions.

Read more of this story at Slashdot.

Matthew Connatser reports via The Register: US government is granting GlobalWafers up to $400 million in CHIPS Act cash to help fund its 300mm wafer manufacturing facilities in Texas and Missouri. The Commerce Department said GlobalWafers' Texas plant is a significant milestone for the US as it's the country's first facility for manufacturing 300mm wafers, the kind that are used for modern processes. The Missouri site will produce a silicon-on-insulator (SOI) variant of 300mm wafers, which are more geared towards defense and aerospace applications where chips need to be less prone to failure. Plans to build the Texas wafer plant were first revealed just over two years ago by the Taiwanese chip biz. It was an alternative use of a few billion dollars that were originally earmarked for acquiring German wafer maker Siltronic, an acquisition which didn't go as hoped due to resistance from German regulators. The Missouri plant meanwhile was announced in 2021 as a partnership between GlobalWafers and GlobalFoundries, the chip fab spun off from AMD that now focuses on older nodes rather than the cutting edge. This fab seems to be the smaller of the two, considering that its budget when first announced was just $800 million, and that seems to also cover an expansion of a 200mm SOI wafer plant. In total, GlobalWafers' Texas and Missouri factories will cost around four billion dollars, which means the maximum award funded by the CHIPS Act would cover up to ten percent of the budget. The Commerce Department claims that facilities will create 1,700 jobs in construction and 880 in manufacturing.

Read more of this story at Slashdot.

According to Axios, Meta will withhold future multimodel AI models from customers in the European Union "due to the unpredictable nature of the European regulatory environment." From the report: Meta plans to incorporate the new multimodal models, which are able to reason across video, audio, images and text, in a wide range of products, including smartphones and its Meta Ray-Ban smart glasses. Meta says its decision also means that European companies will not be able to use the multimodal models even though they are being released under an open license. It could also prevent companies outside of the EU from offering products and services in Europe that make use of the new multimodal models. The company is also planning to release a larger, text-only version of its Llama 3 model soon. That will be made available for customers and companies in the EU, Meta said. Meta's issue isn't with the still-being-finalized AI Act, but rather with how it can train models using data from European customers while complying with GDPR -- the EU's existing data protection law. Meta announced in May that it planned to use publicly available posts from Facebook and Instagram users to train future models. Meta said it sent more than 2 billion notifications to users in the EU, offering a means for opting out, with training set to begin in June. Meta says it briefed EU regulators months in advance of that public announcement and received only minimal feedback, which it says it addressed. In June -- after announcing its plans publicly -- Meta was ordered to pause the training on EU data. A couple weeks later it received dozens of questions from data privacy regulators from across the region. The United Kingdom has a nearly identical law to GDPR, but Meta says it isn't seeing the same level of regulatory uncertainty and plans to launch its new model for U.K. users. A Meta representative told Axios that European regulators are taking much longer to interpret existing law than their counterparts in other regions. A Meta representative told Axios that training on European data is key to ensuring its products properly reflect the terminology and culture of the region.

Read more of this story at Slashdot.

An anonymous reader quotes an excerpt from TechCrunch, written by Zack Whittaker: We're over halfway through 2024, and already this year we have seen some of the biggest, most damaging data breaches in recent history. And just when you think that some of these hacks can't get any worse, they do. From huge stores of customers' personal information getting scraped, stolen and posted online, to reams of medical data covering most people in the United States getting stolen, the worst data breaches of 2024 to date have already surpassed at least 1 billion stolen records and rising. These breaches not only affect the individuals whose data was irretrievably exposed, but also embolden the criminals who profit from their malicious cyberattacks. Travel with us to the not-so-distant past to look at how some of the biggest security incidents of 2024 went down, their impact and. in some cases, how they could have been stopped. These are some of the largest breaches highlighted in the report: AT&T's Data Breaches: AT&T experienced two data breaches in 2024, affecting nearly all its customers and many non-customers. The breaches exposed phone numbers, call records, and personal information, risking account hijacks for 7.6 million customers. Change Healthcare Hack: A ransomware attack on Change Healthcare resulted in the theft of sensitive medical data, affecting a substantial proportion of Americans. The breach caused widespread outages in healthcare services across the U.S. and compromised personal, medical, and billing information. Synnovis Ransomware Attack: The cyberattack on U.K. pathology lab Synnovis disrupted patient services in London hospitals for weeks, leading to thousands of postponed operations and the exposure of data related to 300 million patient interactions. Snowflake Data Theft (Including Ticketmaster): Cybercriminals stole hundreds of millions of records from Snowflake's corporate customers, including 560 million records from Ticketmaster. The breach affected data from multiple companies and institutions, exposing vast amounts of customer and employee information.

Read more of this story at Slashdot.