Bitcoin

SBF Asks For 5-Year Prison Sentence, Calls 100-Year Recommendation 'Grotesque' (arstechnica.com) 189

An anonymous reader quotes a report from Ars Technica: Convicted FTX fraudster Sam Bankman-Fried pleaded for a lenient prison sentence in a court filing yesterday, saying that he isn't motivated by greed and "is already being punished." Bankman-Fried requested a sentence of 63 to 78 months, or 5.25 to 6.5 years. Because of "Sam's charitable works and demonstrated commitment to others, a sentence that returns Sam promptly to a productive role in society would be sufficient, but not greater than necessary, to comply with the purposes of sentencing," the court filing (PDF) said. Bankman-Fried's filing also said that he maintains his innocence and intends to appeal his convictions.

A presentence investigation report (PSR) prepared by a probation officer recommended that Bankman-Fried be sentenced to 100 years in prison, according to the filing. "That recommendation is grotesque," SBF's filing said, arguing that it is based on an erroneously calculated loss of $10 billion. The $10 billion loss asserted in the PSR is "illusory" because the "victims are poised to recover -- were always poised to recover -- a hundred cents on the dollar" in bankruptcy proceedings, SBF's filing said. The filing urged the court to "reject the PSR's barbaric proposal" of 100 years, saying that such sentences should only be for "heinous conduct" like terrorism and child sexual abuse.

The founder and ex-CEO of cryptocurrency exchange FTX, Bankman-Fried was convicted on seven charges with a combined maximum sentence of 110 years after a monthlong trial in US District Court for the Southern District of New York. The charges included wire fraud and conspiracy to commit wire fraud, securities fraud, commodities fraud, and money laundering. US government prosecutors are required to make a sentencing recommendation by March 15, and US District Judge Lewis Kaplan is scheduled to issue a sentence on March 28.

Intel

Intel Puts 1nm Process (10A) on the Roadmap For 2027 (tomshardware.com) 35

Intel's previously-unannounced Intel 10A (analogous to 1nm) will enter production/development in late 2027, marking the arrival of the company's first 1nm node, and its 14A (1.4nm) node will enter production in 2026. The company is also working to create fully autonomous AI-powered fabs in the future. Tom's Hardware: Intel's Keyvan Esfarjani, the company's EVP and GM and Foundry Manufacturing and Supply, held a very insightful session that covered the company's latest developments and showed how the roadmap unfolds over the coming years. Here, we can see two charts, with the first outlining the company's K-WSPW (thousands of wafer starts per week) capacity for Intel's various process nodes. Notably, capacity typically indicates how many wafers can be started, but not the total output -- output varies based on yields. You'll notice there isn't a label for the Y-axis, which would give us a direct read on Intel's production volumes. However, this does give us a solid idea of the proportionality of Intel's planned node production over the next several years.

Intel did not specify the arrival date of its coming 14A node in its previous announcements, but here, the company indicates it will begin production of the Intel 14A node in 2026. Even more importantly, Intel will begin production/development of its as-yet-unannounced 10A node in late 2027, filling out its roster of nodes produced with EUV technology. Intel's 'A' suffix in its node naming convention represents Angstroms, and 10 Angstroms converts to 1nm, meaning this is the company's first 1nm-class node. Intel hasn't shared any details about the 10A/1nm node but has told us that it classifies a new node as at least having a double-digit power/performance improvement. Intel CEO Pat Gelsinger has told us the cutoff for a new node is around a 14% to 15% improvement, so we can expect that 10A will have at least that level of improvement over the 14A node. (For example, the difference between Intel 7 and Intel 4 was a 15% improvement.)

EU

EU Lawmakers Back Draft Rules on Patents for Connected Cars, Telecom Equipment (reuters.com) 3

EU lawmakers on Wednesday approved draft rules governing patents key to technologies for telecom equipment and connected cars in the face of criticism from Nokia, Ericsson and other patent holders. From a report: The draft rules proposed by the European Commission in April last year seek to end costly and lengthy litigation over patents used in technologies for telecom equipment, mobile phones, computers, connected cars and smart devices. The European Parliament will now have to thrash out the details of the proposed rules with EU countries before it can become law. Nokia, Ericsson and Siemens in a letter to EU lawmakers in January, highlighted concerns from the European Patent Office, standard-setting body ETSI and other bodies on the draft rules. Lobbying group IP Europe, which counts Nokia, Ericsson and Qualcomm as its members, reiterated its opposition to the draft rules. "The beneficiaries would not be SMEs as claimed but big tech," IP Europe's managing director Patrick McCutcheon said ahead of the lawmakers' vote.
United States

AI, Drones, Security Cameras: San Francisco Mayor's Arsenal To Fight Crime (reuters.com) 65

San Francisco will vote next week on a divisive ballot measure that would authorize police to use surveillance cameras, drones and AI-powered facial recognition as the city struggles to restore a reputation tarnished by street crime and drugs. From a report: The Safer San Francisco initiative, formally called Proposition E, is championed by Mayor London Breed who believes disgruntled citizens will approve the proposal on Tuesday. Although technology fueled the Silicon Valley-adjacent city's decades-long boom, residents have a history of being deeply suspicious. In 2019, San Francisco, known for its progressive politics, became the first large U.S. city to ban government use of facial recognition due to concerns about privacy and misuse.

Breed, who is running for re-election in November, played down the potential for abuse under the ballot measure, saying safeguards are in place. "I get that people are concerned about privacy rights and other things, but technology is all around us," she said in an interview. "It's coming whether we want it to or not. And everyone is walking around with AI in their hands with their phones, recording, videotaping," Breed said. Critics of the proposal contend it could hurt disadvantaged communities and lead to false arrests, arguing surveillance technology requires greater oversight.

Government

White House Looks To Curb Foreign Powers' Ability To Buy Americans' Sensitive Personal Data With Executive Order (cnn.com) 117

President Joe Biden will issue an executive order on Wednesday aimed at curbing foreign governments' ability to buy Americans' sensitive personal information such as heath and geolocation data, according to senior US officials. From a report: The move marks a rare policy effort to address a longstanding US national security concern: the ease with which anyone, including a foreign intelligence services, can legally buy Americans' data and then use the information for espionage, hacking and blackmail. The issue, a senior Justice Department official told reporters this week, is a "growing threat to our national security."

The executive order will give the Justice Department the authority to regulate commercial transactions that "pose an unacceptable risk" to national security by, for example, giving a foreign power large-scale access to Americans' personal data, the Justice Department official said. The department will also issue regulations that require better protection of sensitive government information, including geolocation data on US military members, according to US officials. A lot of the online trade in personal information runs through so-called data brokers, which buy information on people's Social Security numbers, names, addresses, income, employment history and criminal background, as well as other items.

"Countries of concern, such as China and Russia, are buying Americans' sensitive personal data from data brokers," a separate senior administration official told reporters. In addition to health and location data, the executive order is expected to cover other sensitive information like genomic and financial data. Administration officials told reporters the new executive order would be applied narrowly so as not to hurt business transactions that do not pose a national security risk.
The White House's press release.
AI

Klarna Claims AI Is Doing Agents' Jobs 61

Buy-now-pay-later lender Klarna said its AI assistant, powered by OpenAI, is doing the equivalent work of 700 full-time agents and has had 2.3 million conversations, equal to two-thirds of the company's customer service chats, within the first month of being deployed. The AI tool resolved errands much faster and matched human levels on customer satisfaction, Klarna said.
Science

The Strange and Turbulent Global World of Ant Geopolitics (aeon.co) 10

Over the past four centuries quadrillions of ants have created a strange and turbulent global society that shadows our own. An excerpt from an Aeon article: In their native ranges, these multi-nest colonies can grow to a few hundred metres across, limited by physical barriers or other ant colonies. This turns the landscape to a patchwork of separate groups, with each chemically distinct society fighting or avoiding others at their borders. Species and colonies coexist, without any prevailing over the others. However, for the 'anonymous societies' of unicolonial ants, as they're known, transporting a small number of queens and workers to a new place can cause the relatively stable arrangement of groups to break down. As new nests are created, colonies bud and spread without ever drawing boundaries because workers treat all others of their own kind as allies. What was once a patchwork of complex relationships becomes a simplified, and unified, social system. The relative genetic homogeneity of the small founder population, replicated across a growing network of nests, ensures that members of unicolonial species tolerate each other. Spared the cost of fighting one another, these ants can live in denser populations, spreading across the land as a plant might, and turning their energies to capturing food and competing with other species. Chemical badges keep unicolonial ant societies together, but also allow those societies to rapidly expand.
Programming

White House Urges Devs To Switch To Memory-Safe Programming Languages (infoworld.com) 228

Tontoman shares a report: The White House Office of the National Cyber Director (ONCD) urged tech companies to switch to memory-safe programming languages, such as Rust, to improve software security by reducing the number of memory safety vulnerabilities. Such vulnerabilities are coding errors or weaknesses within software that can lead to memory management issues when memory can be accessed, written, allocated, or deallocated. They occur when software accesses memory in unintended or unsafe ways, resulting in various security risks and issues like buffer overflow, use after free, use of uninitialized memory, and double free that attackers can exploit.

Successful exploitation carries severe risks, potentially enabling threat actors to gain unauthorized access to data or execute malicious code with the privileges of the system owner. "For over 35 years, this same class of vulnerability has vexed the digital ecosystem. The challenge of eliminating entire classes of software vulnerabilities is an urgent and complex problem. Looking forward, new approaches must be taken to mitigate this risk," ONCD's report says. "The highest leverage method to reduce memory safety vulnerabilities is to secure one of the building blocks of cyberspace: the programming language. Using memory safe programming languages can eliminate most memory safety errors."

Facebook

Meta Wants Llama 3 To Handle Contentious Questions as Google Grapples With Gemini Backlash (theinformation.com) 22

An anonymous reader shares a report (paywalled): As Google grapples with the backlash over the historically inaccurate responses on its Gemini chatbot, Meta Platforms is dealing with a related issue. As part of its work on the forthcoming version of its large language model, Llama 3, Meta is trying to overcome a problem perceived in Llama 2: Its answers to anything at all contentious aren't helpful. Safeguards added to Llama 2, which Meta released last July and which powers the artificial intelligence assistant in its apps, prevent the LLM from answering a broad range of questions deemed controversial. These guardrails have made Llama 2 appear too "safe" in the eyes of Meta's senior leadership, as well as among some researchers who worked on the model itself, according to people who work at Meta.

[...] Meta's conservative approach with Llama 2 was designed to ward off any public relations disasters, said the people who work at Meta. But researchers are now trying to loosen up Llama 3 so it engages more with users when they ask about difficult topics, offering context rather than just shutting down tricky questions, said two of the people who work at Meta. The new version of the model will in theory be able to better distinguish when a word has multiple meanings. For example, Llama 3 might understand that a question about how to kill a vehicle's engine means asking how to shut it off rather than end its life. Meta also plans to appoint someone internally in the coming weeks to oversee tone and safety training as part of its efforts to make the model's responses more nuanced, said one of the people. The company plans to release Llama 3 in July, though the timeline could still change, they added.

AI

Google CEO Calls AI Tool's Controversial Responses 'Completely Unacceptable' (semafor.com) 151

Google CEO Sundar Pichai addressed the company's Gemini controversy Tuesday evening, calling the AI app's problematic responses around race unacceptable and vowing to make structural changes to fix the problem. The memo: I want to address the recent issues with problematic text and image responses in the Gemini app (formerly Bard). I know that some of its responses have offended our users and shown bias -- to be clear, that's completely unacceptable and we got it wrong.

Our teams have been working around the clock to address these issues. We're already seeing a substantial improvement on a wide range of prompts. No AI is perfect, especially at this emerging stage of the industry's development, but we know the bar is high for us and we will keep at it for however long it takes. And we'll review what happened and make sure we fix it at scale.

Our mission to organize the world's information and make it universally accessible and useful is sacrosanct. We've always sought to give users helpful, accurate, and unbiased information in our products. That's why people trust them. This has to be our approach for all our products, including our emerging AI products.

We'll be driving a clear set of actions, including structural changes, updated product guidelines, improved launch processes, robust evals and red-teaming, and technical recommendations. We are looking across all of this and will make the necessary changes.

Even as we learn from what went wrong here, we should also build on the product and technical announcements we've made in AI over the last several weeks. That includes some foundational advances in our underlying models e.g. our 1 million long-context window breakthrough and our open models, both of which have been well received.

We know what it takes to create great products that are used and beloved by billions of people and businesses, and with our infrastructure and research expertise we have an incredible springboard for the AI wave. Let's focus on what matters most: building helpful products that are deserving of our users' trust.

Businesses

Amazon Aggregator Thrasio Files For Bankruptcy (nasdaq.com) 18

Thrasio, which acquires third-party Amazon sellers, filed for Chapter 11 bankruptcy protection and received commitments for $90 million in new financing from existing shareholders, it said on Wednesday. From a report: Thrasio also entered into a restructuring agreement with some of its lenders to reduce debt of $495 million from its existing debt pile, it said in a statement. It listed its estimated assets in the range of $1 billion to $10 billion and estimated liabilities of $500 million to $1 billion, according to a document filed with the New Jersey bankruptcy court. Billions of dollars, primarily in loans and other debt instruments, were invested into newly-formed startups aggregating third-party merchants on Amazon's marketplace during the COVID-19 pandemic as investors bet heavily on a prolonged explosion in online shopping. But as lockdowns eased and consumers returned to brick-and-mortar stores, sales growth on Amazon slowed, leaving these heavily-indebted "roll-up" companies struggling to service interest payments.
The Almighty Buck

Zurich Paid 30,000 Workers Double In $200 Million Bank Glitch (fortune.com) 28

An anonymous reader quotes a report from Fortune: Zurich authorities have apologized to city employees after a technical glitch caused a double payment of monthly salaries that local officials are now trying to claw back. About 175 million francs ($200 million) was sent in error on Monday, which was the payday for February, according to a statement. Workers can't keep the money, and officials are trying to devise a streamlined process so that the 30,000 employees affected can easily return it.

A technical error at state-owned Zuercher Kantonalbank, which handles the city's salary transfers, is to blame. The bank itself said that faulty software from one of Swisscom AG's contractors caused the glitch. "Swisscom is aware of the seriousness of this incident and apologizes for the inconvenience caused," the telecommunications company said in a statement shared by the bank. The unexpected windfall prompted a flurry of employees calling up the city's offices to ask about the extra money, according to Swiss newspapers. Others mockingly described it as "inflation compensation" on the city's intranet, and demanded a repeat.

Earth

Antarctica's 'Doomsday Glacier' Began Melting In Mid-20th Century, Study Finds (thehill.com) 60

According to a new study published in the journal Proceedings of the National Academy of Sciences, West Antarctica's Thwaites Glacier began rapidly receding in the 1940s -- much earlier than scientists had previous thought. The Hill notes that it's often referred to as the "doomsday glacier" due to the potentially catastrophic consequences of its hypothetical collapse. From the report: While scientists had already observed the glacier's accelerated retreat by the 1970s, they did not know when it began. Coupled with earlier research about Thwaites's neighboring Pine Island Glacier, the study also provides new, potentially alarming, insight into the cause of the glacier's melting. Scientists tried to reconstruct the glacier's history using analysis of the marine sedimentary record, and they found the Thwaites and Pine Island glaciers both lost contact with the seafloor highs in the 1940s -- at around the same time. These significant changes happened against the backdrop of a massive El Nino weather phenomenon, the scientists found, showing the glaciers "were responding to the same driver(s)."

"The synchronous ice retreat of these two major ice streams suggests that, rather than being driven by internal dynamics unique to each glacier, retreat in the Amundsen Sea drainage sector results from external oceanographic and atmospheric drivers, which recent modeling studies show are modulated by climate variability," the study read. The scientists note that the glaciers' continued retreat shows how difficult it can be to reverse some of the consequences of naturally occurring weather events -- which they say is made even more difficult by human activity. "That ice streams such as Thwaites Glacier and Pine Island Glacier have continued to retreat since then indicates that they were unable to recover after the exceptionally large El Nino event of the 1940s," the scientists wrote. "This may reflect the increasing dominance of anthropogenic forcing since that time but implies that this involved large-scale, in additional to local, atmospheric and ocean circulation changes."

Nintendo

Nintendo Suing Makers of Open-Source Switch Emulator Yuzu (polygon.com) 107

Nintendo has filed a 41-page lawsuit against the makers of Yuzu, an open-source Nintendo Switch emulator, accusing them of "facilitating piracy at a colossal scale." Polygon reports: Yuzu is a free emulator that was released in 2018 months after the Nintendo Switch originally launched. The same folks who made Citra, a Nintendo 3DS emulator, made this one. Basically, it's a piece of software that lets people play Nintendo Switch games on Windows PC, Linux, and Android devices. (It also runs on Steam Deck, which Valve showed -- then wiped -- in a Steam Deck video clip.) Emulators aren't necessarily illegal, but pirating games to play on them is. But Nintendo said in its lawsuit that there's no way to legal way to use Yuzu.

Nintendo argued that Yuzu executes codes that "defeat" Nintendo's security measures, including decryption using "an illegally-obtained copy of prod.keys." "In other words, without Yuzu's decryption of Nintendo's encryption, unauthorized copies of games could not be played on PCs or Android devices," Nintendo wrote in the lawsuit. As to the alleged damages created by Yuzu, Nintendo pointed to the release of The Legend of Zelda: Tears of the Kingdom. Tears of the Kingdom leaked almost two weeks earlier than the game's May 12 release date. The pirated version of the game spread quickly; Nintendo said it was downloaded more than 1 million times before Tears of the Kingdom's release date. People used Yuzu to play the game; Nintendo said more than 20% of download links pointed people to Yuzu.

Though Yuzu doesn't give out pirated copies of games, Nintendo repeatedly said that most ROM sites point people toward Yuzu to play whatever games they've downloaded. Nintendo said its "expended significant resources to stop the illegal copying, marketing, sale, and distribution" of its Nintendo Switch games. It says that Yuzu earns the team $30,000 per month on its Patreon from more than 7,000 patrons. Nintendo said the company has earned at least $50,000 in paid Yuzu downloads. Nintendo said that Yuzu's Patreon doubled its paid members in the period between May 1 and May 12, when Tears of the Kingdom was released. Nintendo is asking the court to shut down the emulator, and for damages.

Medicine

Microplastics Found In Every Human Placenta Tested In Study (theguardian.com) 105

An anonymous reader quotes a report from The Guardian: Microplastics have been found in every human placenta tested in a study, leaving the researchers worried about the potential health impacts on developing fetuses. The scientists analyzed 62 placental tissue samples and found the most common plastic detected was polyethylene, which is used to make plastic bags and bottles. A second study revealed microplastics in all 17 human arteries tested and suggested the particles may be linked to clogging of the blood vessels. [...] Prof Matthew Campen, at the University of New Mexico, US, who led the research, said: "If we are seeing effects on placentas, then all mammalian life on this planet could be impacted. That's not good." He said the growing concentration of microplastics in human tissue could explain puzzling increases in some health problems, including inflammatory bowel disease (IBD), colon cancer in people under 50, and declining sperm counts. A 2021 study found people with IBD had 50% more microplastics in their feces. Campen said he was deeply concerned by the growing global production of plastics because it meant the problem of microplastics in the environment "is only getting worse."

The research, published in the Toxicological Sciences journal, found microplastics in all the placenta samples tested, with concentrations ranging from 6.5 to 790 micrograms per gram of tissue. PVC and nylon were the most common plastics detected, after polyethylene. The microplastics were analyzed by using chemicals and a centrifuge to separate them from the tissue, then heating them and analyzing the characteristic chemical signature of each plastic. The same technique was used by scientists at the Capital Medical University in Beijing, China, to detect microplastics in human artery samples. The concentration of microplastics in placentas was especially troubling, Campen said. The tissue grows for only eight months, as it starts to form about a month into pregnancy. "Other organs of your body are accumulating over much longer periods of time," he added.

Power

US Judge Halts Government Effort To Monitor Crypto Mining Energy Use (theguardian.com) 90

A federal judge in Texas has granted a temporary order blocking the U.S. government from monitoring the energy usage of cryptocurrency mining operations, stating that the industry had shown it would suffer "irreparable injury" if it was made to comply. The Guardian reports: The US Department of Energy had launched an "eemergency" initiative last month aimed at surveying the energy use of mining operations, which typically use vast amounts of computing power to solve various mathematical puzzles to add new tokens to an online network known as a blockchain, allowing the mining of currency such as bitcoin. The growth of cryptocurrency, and the associated mining of it, has been blamed for a surge in electricity use as data centers have sprung up across the US, even reviving, in some cases, ailing coal plants to help power the mining. [...]

"The massive energy consumption of cryptocurrency mining and its rapid growth in the United States threaten to undermine progress towards achieving climate goals, and threaten grids, communities and ratepayers," said Mandy DeRoche, deputy managing attorney of the clean energy program at Earthjustice. Until now, a lack of publicly available information has only benefited an "industry that has thrived in the shadows," DeRoche added.

The crypto mining industry, however, has claimed it is the victim of a "politically motivated campaign" by Joe Biden's administration and has, for now, succeeded in averting a survey that it contends is unfairly onerous. "This is an attack against legitimate American businesses with the administration feigning an emergency to score political points," said Lee Bratcher, president the Texas Blockchain Council, one of the groups that sued to stop the survey. "The White House has been clear that they desire to 'to limit or eliminate' bitcoin miners from operating in the United States. "Although bitcoin is resilient and cannot be banned, the administration is seeking to make the lives of bitcoin miners, their employees, and their communities too difficult to bear operating in the United States. This is deeply concerning."

Open Source

'Open Documentation Academy' Offers On-Ramp To OSS 4

"Documentation authors at Canonical have launched the Open Documentation Academy to offer an easy way to get started contributing to open-source projects," writes longtime Slashdot reader tykev. From the blog post: Open and inclusive collaboration, and the sharing of ideas, remains the best way to develop software (and to do many other things!), but we also recognise that this "getting involved" step can be difficult. Where do you start? Who do you ask? What needs to be done? We all very much want to help people become open source contributors by building an on-ramp process. It may take some time, and we will need to adapt, but this is exactly why we've started our Open Documentation Academy.

To help you get involved, the Open Documentation Academy provides a curated list of documentation tasks. Choose one, let us know, and get started. Tasks include testing and fixing tutorials, updating the outdated, restructuring large documents, and anything else you may want to suggest. Our list is growing, and a big part of the Documentation Academy will be ensuring there's always a wide range of tasks available, across as many projects and technologies as possible. And of course, we're here to help. We'll guide you through your first contributions, provide advice on approaches, and help you build your confidence.
The Almighty Buck

Uber-Like Surge Pricing Is Coming For Fast Food (sfgate.com) 198

Fast food chain Wendy's announced it's adopting a similar approach to Uber's Surge Pricing policy by dynamically adjusting the prices of its menu items during peak demand periods at certain locations. The controversial strategy seeks to leverage real-time data to align pricing and demand, enhancing efficiency and potentially improving customer satisfaction. From a report: During a conference call earlier this month, Wendy's CEO Kirk Tanner said the fast-food chain would experiment with dynamic pricing as early as next year. "Beginning as early as 2025, we will begin testing more enhanced features like dynamic pricing and daypart offerings, along with AI-enabled menu changes and suggestive selling," he said. "As we continue to show the benefit of this technology in our company-operated restaurants, franchisee interest in digital menu boards should increase, further supporting sales and profit growth across the system."

Prices seesaw all the time on the sites of online retailers like Amazon that use algorithms and artificial intelligence to monitor competitors and glean insights into individual shoppers, adjusting prices depending on interest in the product or in the brand, said Timothy Webb, an assistant professor at the University of Delaware's hospitality and sport business management program. Coupons and other offers are also routinely dangled in mobile apps to encourage people to make purchases. "A lot of this stuff is already happening even if you don't realize that it is happening. If you have the Starbucks app and I have the Starbucks app, we probably have different offers," Webb said. "We might not be in the drive-through and they just increased the prices, but we are already paying different prices for the same products."

But, he says, Wendy's fans will likely see moderate, not massive, price swings during periods of peak demand. "It's not like $200 or $300 on a flight. This is a hypercompetitive industry. If Wendy's goes up $2 to $3 on a burger at dinner time, I would be shocked. People have too many options. They will just walk down the street and eat at Burger King instead," Webb said. "There will just be little price changes here."

Security

Hackers Backed By Russia and China Are Infecting SOHO Routers Like Yours, FBI Warns (arstechnica.com) 36

An anonymous reader quotes a report from Ars Technica: The FBI and partners from 10 other countries are urging owners of Ubiquiti EdgeRouters to check their gear for signs they've been hacked and are being used to conceal ongoing malicious operations by Russian state hackers. The Ubiquiti EdgeRouters make an ideal hideout for hackers. The inexpensive gear, used in homes and small offices, runs a version of Linux that can host malware that surreptitiously runs behind the scenes. The hackers then use the routers to conduct their malicious activities. Rather than using infrastructure and IP addresses that are known to be hostile, the connections come from benign-appearing devices hosted by addresses with trustworthy reputations, allowing them to receive a green light from security defenses.

"In summary, with root access to compromised Ubiquiti EdgeRouters, APT28 actors have unfettered access to Linux-based operating systems to install tooling and to obfuscate their identity while conducting malicious campaigns," FBI officials wrote in an advisory Tuesday. APT28 -- one of the names used to track a group backed by the Russian General Staff Main Intelligence Directorate known as GRU -- has been doing just for at least the past four years, the FBI has alleged. Earlier this month, the FBI revealed that it had quietly removed Russian malware from routers in US homes and businesses. The operation, which received prior court authorization, went on to add firewall rules that would prevent APT28 -- also tracked under names including Sofacy Group, Forest Blizzard, Pawn Storm, Fancy Bear, and Sednit -- from being able to regain control of the devices.

On Tuesday, FBI officials noted that the operation only removed the malware used by APT28 and temporarily blocked the group using its infrastructure from reinfecting them. The move did nothing to patch any vulnerabilities in the routers or to remove weak or default credentials hackers could exploit to once again use the devices to surreptitiously host their malware. "The US Department of Justice, including the FBI, and international partners recently disrupted a GRU botnet consisting of such routers," they warned. "However, owners of relevant devices should take the remedial actions described below to ensure the long-term success of the disruption effort and to identify and remediate any similar compromises."

Those actions include:

- Perform a hardware factory reset to remove all malicious files
- Upgrade to the latest firmware version
- Change any default usernames and passwords
- Implement firewall rules to restrict outside access to remote management services

AI

Ghost Kitchens Are Advertising AI-Generated Food On DoorDash and Grubhub (404media.co) 48

Emanuel Maiberg reports via 404 Media: Dozens of Ghost kitchens, restaurants that serve food exclusively by delivery on apps like DoorDash and Grubhub, are selling food that they promote to customers with AI-generated images. It's common for advertisements to stage or edit pictures of food to make it look more enticing, but in these cases the ghost kitchens are showing people pictures of food that literally doesn't exist, and looks nothing like the actual items they're selling, sometimes because the faulty AI is producing physically impossible food items. [...] Some ghost kitchens exist as unmarked commercial kitchens with no actual restaurant you can visit that simply fulfill orders for a variety of brands that only exist on the food delivery services. Other ghost kitchens piggyback on existing, real restaurant kitchens to fulfill orders for those brands that exist only on food delivery apps.

[The food from a business on DoorDash called Pasta Lovers] actually comes from Tony's Pizzeria in North Brooklyn, which also fulfills orders for a cheesesteak brand called Philly Cheez, a hero sandwich brand called Hero Mania, and a wrap brand called That's A Wrap. All of these brands deliver food from different ghost kitchens across the country, and all of them feature the same type of AI-generated images to promote their food, some of which looks ridiculous. [...]

"We don't allow the use of AI-generated images and if we find a merchant is using any, we will remove those images from their menu," Grubhub, which also operates Seamless, told me in an email. However, at the time of writing the AI-generated images on Seamless I sent the company are still live on its site. "We know how important it is for diners to have realistic expectations of what they are ordering and should expect to receive, which is why we share image guidelines with our partners and our system reviews image submissions before they're allowed on our platform." "DoorDash is committed to showcasing realistic representations of meals that customers would receive when ordering online," DoorDash told me in an email. "Showcasing high-quality, accurate, and realistic menu images is crucial for maintaining customer trust and generating sales through DoorDash Marketplace."
"This is all incredibly depressing," concludes Maiberg. "A local pizzeria can't get by unless it makes sandwiches for ghost kitchen brands, the people who make a living taking photographs of food are being displaced by AI tools, and gigantic food delivery apps are still making money by taking a cut from restaurants and screwing over gig delivery drivers."

"AI-generated images of food that people can order and eat finally brings us to a shockingly literal manifestation of Jean Baudrillard's Simulacra. Baudrillard would say the Spicy Philly Cheese from Philly Cheez is "never that which conceals the truth -- it is the truth which conceals that there is none."

Slashdot Top Deals