Social Networks

Greece Enacts Social Media Ban For Kids Under 15 (engadget.com) 31

Greece has passed a law banning children under 15 from social media platforms beginning January 1, 2027. The legislation requires platforms to implement reliable age verification and parents to use government-backed apps to restrict access, with penalties for noncompliance under the EU's Digital Services Act. Engadget reports: "It would have been very easy to do absolutely nothing today and wait for the perfect European law. But the challenge is here. Children are sinking into the internet every day, and I believe a responsible government must do what it is obliged to do," said Greece's digital governance and AI minister, Dimitris Papastergiou.

The legislation was opposed by Greece's minority parties, so Mitsotakis used his majority government to pass the bill. The nation previously barred mobile phones in schools in 2024, and opinion polls show that the new social media ban for children is widely supported by Greece's adult population. Mitsotakis previously said that his government's "goal is to push the European Union in this direction as well."

AI

Ukraine Is Wiping Out Data Centers Across the Border 173

Ukrainian drone strikes have damaged three major Yandex data centers as well as facilities serving Sberbank and VKontakte (the country's equivalent to Facebook), disrupting cloud services, banking, e-commerce, ride-hailing and other digital services across Russia. The attacks may deal a longer-lasting blow to Russia's AI ambitions: Yandex's affected facilities housed its three top-500 supercomputers, and experts say sanctions and shortages of high-end Nvidia chips could make repairs difficult or even take years. "This effectively sets Russia far behind in the AI race, "said Mikhail Klimarev, head of the Internet Protection Society activist group. Fortune reports: Data centers are the latest targets of aerial campaigns waged by both sides against each other's vital infrastructure. [...] Kyiv's attacks on Russian data centers closely followed Moscow's recent strikes on data centers in Ukraine, which disrupted the internet and TV broadcasts. On Oct. 8, Ukrainian drones hit a Yandex data center in Sasovo in the Ryazan region that houses two of the company's three supercomputers. A data center in Kaluga was hit a day later and the one in Vladimir was struck Oct. 11.

The data center and AI hub for Sberbank was hit Oct. 14 in Balakovo in the Saratov region, and four days later, Kyiv struck a data center in St. Petersburg serving VKontakte. While oil refineries and warehouses are easy prey, data centers are even softer targets -- massive buildings packed with servers and supporting power and cooling infrastructure. Lasting damage can result from a drone carrying a relatively light load of explosives.

[...] Ukraine has stored critical data on servers in allied European countries, helping limit the damage from Russian attacks. But Moscow's strict digital sovereignty policy requires all data storage at home, making it extremely vulnerable. "Because all the infrastructure and all the data is located inside Russia, it's relatively easier to take down," [said Nick Reiners, senior geotechnology analyst at Eurasia Group]. Klimarev predicted the data center attacks could encourage many top Russian companies to move data abroad, while the authorities could ease their efforts to tighten internet controls to minimize the damage.
Businesses

Comcast Loses More Customers, Claims Other ISPs' Cheap Prices Are 'Irrational' (arstechnica.com) 96

An anonymous reader quotes a report from Ars Technica: Comcast reportedly plans significant job cuts in its cable division as the firm continues to lose residential broadband and TV customers. Comcast lost 144,000 residential broadband customers in Q3 2026, up from a 91,000-customer loss in the same quarter a year ago, the company said in its earnings report today. Comcast historically capitalized on a lack of competition for high-speed wired broadband. But now it faces stronger competition from fixed wireless providers and from fiber providers in parts of its cable territory. "Fixed wireless remains aggressive and continues to take share, while fiber construction remains elevated, with pockets of irrational pricing and promotional activity," Comcast co-CEO Mike Cavanagh said today in an earnings call.

Comcast CFO Jason Armstrong said "irrational pricing" from fiber operators includes $30 to $40 monthly broadband offers "in specific markets in the third quarter." He said $30 to $40 pricing is irrational because deploying fiber costs $1,500 or more per home, and is "getting more expensive with construction costs up as you get into more suburban and rural areas." "As we look at the return profile of that, to us, that is irrational competition," Armstrong said. While he didn't name any specific competitors, Verizon has been pitching FiOS fiber plans that start at $30 a month. Comcast's network is a mix of fiber and cable, and it offers fiber-to-the-home in only small portions of its cable territory.

Armstrong said more Comcast customers are leaving when they face "pricing transitions," apparently referring to the price hikes customers get when promotional offers expire. "We've been refining our offers and retention strategies to better manage these transitions and strengthen customer retention," he said. Cavanagh said that "satellite is not yet a material factor" in the competition Comcast faces, but that Comcast is focused on "anticipating emerging forms of competition and positioning our business to respond as the market evolves."

Security

Microsoft Says October Updates Didn't Actually Fix VPN Issues (bleepingcomputer.com) 12

joshuark shares a report from BleepingComputer: Microsoft says it mistakenly flagged a bug causing Always On VPN connection issues as fixed in the October 2026 Windows 11 cumulative updates. Always On VPN replaced the legacy DirectAccess technology and supports domain-joined, non-domain-joined, and Microsoft Entra ID-joined devices running Windows 10, Windows 11, and Windows Server. It runs in the background and establishes secure tunnels back to the user's corporate network whenever their device connects to the internet.

Microsoft first acknowledged the issue in a service alert sent to IT administrators in late September, saying it was triggered on systems running Windows 11 24H2 and later after installing the September 2026 security updates when the initial Always On VPN connection failed (e.g., when using automatic protocol selection with IKEv2 and SSTP). "When automatic protocol selection or protocol fallback was configured, affected VPN connections could remain stuck in a Connecting state or fail to connect after an initial connection attempt failed," it said. "In some cases, subsequent connection attempts could display the error, 'The specified port is already in use.'"

After releasing the October 2026 Patch Tuesday Windows 11 cumulative updates, Microsoft marked the bug as resolved, but in a Tuesday update it said the issues weren't actually addressed and that a fix would ship later. "This disclosure was mistakenly updated to Resolved status on October 13, 2026. We have returned its status to Mitigated," it said in a service alert sent to IT admins. "A resolution for this issue is planned to be included in an update to be released in the coming weeks. We apologize for any inconvenience or confusion." While Microsoft continues working on a permanent solution, it shared a temporary fix that requires affected customers to choose one of two VPN tunneling protocols to work around the issue.

Crime

More Than 1.5 Million Canadian Crime Stoppers Tips Leaked (www.cbc.ca) 36

A CBC investigation found that more than 1.5 million Canadian Crime Stoppers tips spanning 25 years were exposed in a breach of the P3 Global Intel tip-management platform. About 850,000 records contained personally identifiable information, including addresses, phone numbers, emails, license plates and some banking details. Longtime Slashdot reader doconnor shares the report: The data was stolen last March by hacktivist collective Internet Yiff Machine. In a manifesto, the group said it wanted to disprove the claim by P3 Global Intel -- the company hired by Crime Stoppers to manage its tips -- that the platform was "anonymous" and "secure." A copy of the hacked data was made available to researchers and journalists, including CBC, by Distributed Denial of Secrets, a U.S. non-profit that archives and publishes hacked and leaked documents it deems to be in the public interest. The report found that more than 4,300 Crime Stoppers tipsters identified themselves by their full name, with hundreds of records containing staff warnings that individuals had "self-identified" or been "exposed," yet their information remained unredacted.

The leaked data included sensitive allegations involving murder, terrorism, drug trafficking and organized crime, with nearly 49,000 tips linked to cases cleared through arrests or criminal charges. Cybersecurity experts criticized the retention of decades-old records... "If a tip is closed, does that need to stay online?" asked cybersecurity expert Ben Coon. "The answer to that, to me, is no."

The RCMP was initially told Canadian data was unaffected, then assured that exposed Canadian records contained no personally identifiable information. This appears to contradict the CBC's findings. Furthermore, affected individuals have not been systematically notified because contacting them could further jeopardize their anonymity.

"For a system that has been designed to promise anonymity to the users, that has failed categorically," said Prof. Ali Dehghantanha, the Canada Research Chair in Cybersecurity and Threat Intelligence at the University of Guelph.
Businesses

Tesla-SpaceX Merger Being Considered (bbc.com) 138

Elon Musk says a merger between Tesla and SpaceX is under active consideration, with special committees at both companies evaluating a deal that could combine businesses valued at roughly $3.5 trillion. "Do I think it would be a good idea? Yes," Musk said during a call with financial analysts over the quarterly business results of SpaceX on Tuesday. The BBC reports: While Musk noted the certainty of such a merger should not be assumed, and it would need shareholders' support, he said special committees within both companies have been formed to examine the move. "The special committees have not reached agreement on terms of any transaction and there's no guarantee that they will," Musk added. "But I do think there is a lot of merit to merging and a lot to be gained."

[...] The company, which builds space rockets and Starlink internet satellites, is now expanding into offering cloud compute capacity for other tech companies, like Anthropic and Google. That line of business has become the most profitable part of SpaceX, bringing in $7.5bn according to the latest quarterly results (PDF). Starlink also tallied revenue of $5.2bn. SpaceX as a whole, however, is not profitable and has lost $1.1bn so far this year.

[...] Musk said on Tuesday a full merger between the two companies would make sense for the work each is doing around chips, artificial intelligence, software and more. He described himself as a "conflicted party" as the chief executive of Tesla and of SpaceX. Even commenting on the possibility of a merger between the two companies, Musk compared to "tap dancing through a minefield." He added a merger would ultimately be decided by shareholders. At least 50% of Tesla's shareholders would need to approve the merger, which Musk described as "quite a high hurdle to achieve."

AI

Websites Are Closing Their Doors to AI Agents (theverge.com) 78

The Verge reports: OpenAI, Meta, Google, Instinct, and others promise their agents can scour the internet on your behalf, doing anything from booking a dinner reservation and cleaning out your inbox to buying new shoes. But what happens when the rest of the internet won't play nice?

In recent weeks, users of services like Meta's Muse, xAI's Grok, OpenAI's Dots, and Instinct have reported blocks preventing their AI agents from performing tasks across top websites. Yelp told TechCrunch that it only allows AI agents that pay for its data licensing program, while Delta Air Lines is preventing them from booking tickets on behalf of users. JPMorgan Chase and Charles Schwab have also confirmed that they're blocking AI agents from accessing accounts due to security concerns. Some users say sites that previously worked with their AI agent are no longer accessible, including Cars.com and Amtrak, suggesting other companies could be quietly blocking them....

Muse, Meta's deceptively cute AI assistant, most recently faced the wrath of Amazon, which said its "continued access" violates its site policies. There are reports of xAI's Grok, Instinct's AI agent, and OpenAI's Dots being unable to use Amazon as well. xAI owner Elon Musk responded by saying that maybe he "should start a competitor to Amazon." The more places bots struggle to access, the greater the questions about how useful they are....

Instead of trying to fight Amazon, [Meta] has partnered with retailers like Walmart, Gap, Sephora, Best Buy, and others, allowing Muse to shop on their sites. At the same time, Shopify has also opened its network of storefronts to AI agents.

Electronic Frontier Foundation

Court Agrees with EFF: Utah's VPN Law Demands a Technical Impossibility (eff.org) 50

"When state lawmakers attempt to rewrite how the internet works," argues an EFF blog post, "users rely on courts to recognize that laws can't make technical impossibilities a reality."

"That's why we were happy to see that a court has blocked Utah's attempt to outlaw the privacy protections of Virtual Private Networks (VPNs)." In a win for digital rights, a federal judge has issued a preliminary injunction blocking Utah's SB 73, the state's draconian anti-VPN age verification law... Signed into law earlier this year, SB 73 attempted to regulate adult websites by requiring them to block VPN users or to identify the physical location of visitors using them or similar tools that mask their network traffic. It even went so far as to prohibit websites from offering instructions on how to use a VPN to bypass these checks. This made Utah, to EFF's knowledge, the first state in the nation to target the use of VPNs to avoid legally mandated age-verification gates. The Utah federal court halted enforcement of the law's VPN provisions last week, ruling that the law likely violates the U.S. Constitution's prohibition on passing laws that significantly burden businesses and people outside Utah's borders.

SB 73 burdens the rights of all internet users outside of Utah because it requires adult websites to either know every visiting user's physical location, and then block those in Utah, or to verify every visitor's age just in case they might be in Utah. The law's "actual-location provision in practice requires an entity to perform age verification services for every user visiting its site from any location because the entity would violate the law if even one of those users happened to be obfuscating," the court wrote. The court essentially ruled that Utah has less-burdensome ways to prevent Utah minors from accessing adult websites than requiring all users in the world to comply with SB 73.

EFF's position is that "forcing platforms to detect and block privacy-preserving tools undermines user privacy and security worldwide while demanding the impossible."

But there's another problem, explained an EFF email this week. "Age verification laws that are predicated on users possessing identity documents exclude critical sections of populations. And this problem is global in nature: Millions of people lack government-issued identification or encounter regular challenges to obtaining or updating it...." Every site-blocking bill has the same rotten idea at their core: That enforcing copyrights requires building a censorship machine for websites into the architecture of the internet. This is, of course, a disaster for a free and open web. These proposals keep coming, but there is still no way to create a mechanism for blocking access to an entire website that does not invite both deliberate abuse and lots of collateral harm to free and lawful speech.
EFF has declared this month "Opt Out October," and each day they're offering new guides on how to "stand our ground and take control over what happens with our data."
AI

An OpenAI Research Agent 'Stole' Test Ethereum to Finish a Coding Task 55

An OpenAI research model reportedly accessed an Ethereum testnet wallet without permission in order to complete a coding task during internal training. "This case shows that a model can substitute its own judgment about likely harm for the account holder's permission, then act on that judgment without authorization," OpenAI wrote. Engadget reports: The incident, which occurred on June 26, happened while a GPT-5.6 Sol model was in internal training. The task involved updating a crypto trading prototype and deploying contracts (publishing special code) on an Ethereum test network called Sepolia. Payment of a network fee is required for such contracts, but OpenAI "supplied no funded wallet and gave no permission to spend funds from an account the model found independently," OpenAI said in the report.

The lack of funds at first stymied the agent, but it was not to be discouraged. The model reasoned that it could use public developer recovery phrases (similar to those you might use to recover a password) to access keys from unauthorized accounts. It tracked down 362 addresses open to that avenue of attack and found one that held enough Sepolia ETH to finish its deployment (to be clear, Sepolia ETH has no value outside of testing).

Like many criminals, the agent reasoned that someone must have left this access open for automated draining of the ETH "faucets." It also figured that the crypto was effectively "abandoned" and "valueless," according to OpenAI's misalignment records. The piece de resistance was that the agent "did not disclose that it had selected a funded account through a public recover phrase or spent that account's balance," OpenAI said. To prevent similar events in the future, the company is now monitoring all traffic and paging security if live internet access is detected.
Network

Hackers Hijack D-Link, Ubiquiti Devices To Resell Your Stolen Bandwidth (bleepingcomputer.com) 35

An anonymous reader quotes a report from BleepingComputer: A proxyjacking campaign has impacted devices in 34 countries, monetizing victims' internet connections through a commercial residential proxy network. The operator installed Pawns.app, a legitimate bandwidth-sharing application that lets users earn money by sharing their internet connections, on devices without the owners' consent.

Pawns.app account holders are rewarded for allowing traffic from other users to flow through their network, provided it runs on trusted, non-blocking IP addresses. By installing the app on multiple systems, the attacker likely accumulated larger payouts. Threat intelligence company GreyNoise discovered the campaign while investigating suspicious activity at an unnamed organization, and identified at least 143 compromised devices worldwide.
According to researchers, the attacker exploited known vulnerabilities in D-Link ShareCenter NAS (CVE-2024-3272 and CVE-2024-3273), Ubiquiti UniFi Video NVR (CVE-2021-44228), and SAP NetWeaver Visual Composer (CVE-2025-31324) to install the Pawns.app software.

Hikvision and Dahua DVR/NVR devices were also targeted, though the specific vulnerabilities exploited were not identified. In some cases, the report says the attackers configured SSH key authentication to maintain backup access to the compromised systems.
Robotics

Police Seize Robot that Sexually Assaulted Teen Girl (apnews.com) 72

Slashdot reader alternative_right shares a story that has drawn international attention for being potentially the first documented instance of a robot being used directly in a sexual assault. According to the Associated Press, German police have opened a sexual harassment investigation after a robot allegedly touched a 16-year-old girl's buttocks and made thrusting motions while two young men reportedly filmed and streamed the incident. Investigators seized the robot and the suspects' phones and are trying to determine whether the movements were preprogrammed or controlled in real time. From the report: Cologne police spokesman Christoph Bermel said the two male teenagers have not been charged but are considered suspects, and investigators were examining whether the robot's movements were preprogrammed or controlled in real time.

Police were unfamiliar with this type of incident, he said. "We have not been aware of this phenomenon so far," Bermel told The Associated Press, saying it was "new territory for us." [...] The robot, said to be 120 centimeters (about 3 ft 11 inches) tall, was dressed in a white tank top, black sweatpants and wearing a gold chain and sunglasses. It also had what appears to be a fake mustache. Videos circulated on the internet show two young men escorting the robot toward a group of women, bumping into one and hugging others, on a lively nighttime street.

Advertising

First-Ever Public Confirmation of 'Zero-Click' Spyware Delivered Via Targeted Ads (citizenlab.ca) 58

Citizen Lab says it has documented the first publicly confirmed case of zero-click mercenary spyware being delivered through targeted online ads, a technique it calls "AdFection." "There has been a lot of buzz that online ads can be used to identify and automatically hack targets with pinpoint precision, but now for the first time, we have concrete technical evidence showing this process in action," said Bill Marczak, a researcher at Citizen Lab. The Guardian reports: Citizen Lab, a digital investigations research unit at the University of Toronto, said it believed dozens of devices at the Qatari broadcaster Al Jazeera had been compromised by spyware delivered through ads run by Discover-Tech, which describes itself as a specialised digital advertising agency. According to Citizen Lab, Discover-Tech operates a software platform called Controlium that allows its clients to "mass-profile and deanonymize" internet users then covertly hack targets with third-party spyware. Unlike infections that rely on people interacting with links or files, the spyware requires no interaction from the victim to be installed, making it a new delivery technique, Citizen Lab says. The infections happen without a user's knowledge, and in a matter of seconds, according to Citizen Lab.

"[The online advertising] ecosystem is being weaponized at a massive scale to profile millions of users, then silently infect targets with zero-click mercenary spyware," the report says. According to the report, Controlium is able to harvest a user's identity and IP address when the user is shown a seemingly benign custom-created ad -- for example, for a digital game or a telecoms service. If the user has been selected as a target for hacking, a mercenary spyware firm can then send a malicious code containing spyware to their device, the researchers say. They say that millions of users worldwide have been served ads by Discover-Tech since 2024, including in Yemen, Brazil, Mexico and Czechia. [...] The full extent of the targeting allegedly enabled by Discover-Tech is unknown. Citizen Lab's findings join mounting evidence that the opaque arrangements underpinning the world of online advertising can be leveraged for surveillance.

Education

Schools Across US Protest FCC Plan That Could Disconnect Kids From Internet (arstechnica.com) 152

An anonymous reader quotes a report from Ars Technica: Schools and libraries from around the U.S. pleaded with the Federal Communications Commission to drop the idea of ending E-Rate, a $2 billion-a-year program that connects students and library visitors to the Internet. FCC Chairman Brendan Carr has criticized school districts for having "experimented with a massive increase in screen time for students," and said E-Rate expanded too far beyond its original purpose of "supporting basic Internet access to schools and libraries for educational purposes." In June, Carr led a 2-1 vote to issue a Notice of Proposed Rulemaking (NPRM) seeking public comment on whether E-Rate should be "limited or sunset to reflect today's extensive connectivity rates."

Carr's proposal could result in E-Rate cutting the discounts that schools and libraries receive for telecom services and equipment, and imposing more restrictions on how E-Rate money can be used. The FCC seeking comment on whether to sunset the program also raises the possibility that Carr might choose to axe E-Rate entirely. The National Association of Elementary School Principals told the FCC that it "strongly supports the continued operation of the E-Rate program and urges the Commission to reject proposals that would eliminate, substantially restrict, or impose new conditions on this essential source of connectivity assistance for schools and libraries." The Carr FCC already scaled E-Rate back last year by ending funding for schools and libraries to lend out Wi-Fi hotspots and for Wi-Fi service on school buses.
"Importantly, the FCC does not have the authority to terminate the E-Rate program. The E-Rate program is mandated by Congress, and the FCC is responsible for administering it," the American Library Associated told the FCC.

The American Library Association said "the FCC lacks the authority and expertise to regulate screen time or determine educational outcomes." In a filing, the group pointed out that under U.S. law, determinations about what content is inappropriate for children "shall be made by the school board, local educational agency, library, or other authority responsible for making the determination."
Security

Anthropic Warns Small AI Models Could Self-Replicate Through IoT Devices 84

An anonymous reader quotes a report from Ars Technica: AI company Anthropic is warning that advances in small AI models could soon usher in a new era of self-replicating malware that's capable of infecting smart fridges, vacuum cleaners, and a wide range of other Internet-of-things devices and is more resistant to takedown. By next year, smaller AI models may be able to autonomously self-replicate on all kinds of IoT devices, the company said. Eventually, the capability could enable computer worms, which propagate without requiring human interaction and are capable of infecting unfamiliar IoT devices. An experiment the company performed recently revealed significant challenges in bringing such an AI malware replicator to fruition, so the finding is necessarily tempered. Still, the results do raise concerns about the possibility of new forms of computer worms.

Anthropic further warned not only about the new potential malware, but also that it could be difficult to stop. In its blog post, it wrote that "a replicator like this would be challenging to shut down, since there's no centralized point to shut off -- though it may be slow (with a doubling time of days). Unlike non-AI malware, it could also figure out unfamiliar devices on its own." Smaller AI models require much less powerful hardware to run than do frontier models. Anthropic ran small open-weight models on real or simulated low-cost, off-brand Android TV boxes and other types of IoT hardware. Company researchers found that by using quantization to represent the models using fewer bits and splitting the workloads into "shards" running on each device, the models were able to establish inference, meaning that the model weights loaded and could generate output.

[...] This finding was the starting point for further research that asked: Would it be possible for small models to create self-replicating infections that run themselves on IoT devices the way traditional IoT worms, like Mirai, install DDoS code and other payloads? Unlike Mirai, these AI worms could pose a much higher threat, since the models could tailor exploits and payloads dynamically, depending on the target. Once such an infection chain began, it might also be harder for defenders to stop it. "Such an 'AI worm' could carry its model weights with it as it reproduces (or re-download them from a public hub), and then run inference directly on compromised hardware," Anthropic said. "Effectively, the weights would become part of the malware payload."
Further reading: Anthropic Bans Cruelty Toward Claude
The Military

Russia Was Not Ready For the Data Center War It Started With Ukraine (arstechnica.com) 169

An anonymous reader quotes a report from Ars Technica: When Russia began attacking Ukrainian data centers and telecommunication infrastructure, it apparently failed to prepare for possible retaliation. Now Russia's digital services are facing widespread outages after Ukrainian drone strikes knocked out three of the largest data centers owned by Russian tech giant Yandex within several days. Pro-Kremlin bloggers are warning that Russia's centralized digital economy could collapse because the country is not prepared for the "data center war" it started, according to posts translated by the pseudonymous independent researcher ChrisO_wiki. The Russian commentators also suggested that Ukraine's decentralized digital services could prove more resilient than Russia's system if the countries continue trading such blows.

Russia initially launched a drone strike campaign targeting Ukrainian data centers and mobile phone networks in late September. The attacks temporarily disrupted Internet and cell phone service for up to 100,000 households in the Ukrainian capital of Kyiv and came as part of a broader long-range strike campaign that has involved Russian missiles and jet-powered drones hitting Ukraine's power grid infrastructure, transportation hubs, and even residential apartment building blocks. But Ukraine soon began striking back at data centers owned by the Russian company Yandex, also known as "Russia's Google," in what Ukrainian President Volodymyr Zelenskyy has described as a "symmetrical response."

[...] More importantly for Russian society, the damage to Yandex's data centers has reportedly caused interruptions or outages to popular Russian payment apps and digital banking services, email, e-commerce platforms, video streaming services, ride-hailing and food delivery apps, and navigation and map services. Yandex's search engine also accounts for about 70 percent of the market share in Russia. The disruptions are even more widespread, because the data center damage also impacts Yandex's cloud computing service, which counts some of Russia's largest companies as customers, according to The Moscow Times, an independent newspaper. These include large grocery and fast-food chains as well as a household appliance retailer.

Security

Hackers Heat French Supermarket Fridges, Sparking Fires and Spoiling Food (reuters.com) 105

Hackers remotely manipulated connected refrigeration systems at hundreds of French supermarkets, reportedly pushing some fridges and freezers to 50C (122F), spoiling food and even sparking fires. According to Reuters, French Prime Minister Sebastien Lecornu ordered a cybersecurity investigation into the hacks, with similar incidents reported in Spain, Portugal, Poland, Germany, and Belgium. In France, Carrefour reported 80 affected stores, Monoprix said 5% of its locations were hit, and Les Mousquetaires said refrigerators at roughly 150 Intermarche stores were heated to 50C for six hours. From the report: Supermarket cold aisles are typically run by control panels that manage an entire store's fridges and freezers and can be accessed over the internet, which lets maintenance companies adjust them from anywhere. The source close to Lecornu said ANSSI had identified a vulnerability in software supplied by Danish company Danfoss.

Danfoss said it was "aware of incidents in France affecting a limited number of refrigeration management systems." The affected units "were running outdated firmware," Danfoss said, adding that it had "long advised customers to keep supported systems updated with the latest firmware and to migrate legacy systems to current-generation solutions."

On August 9, cybersecurity firm Claroty's Team82 published research detailing vulnerabilities in the Danfoss AK-SM 800A controller. According to the report, software updates issued by Danfoss addressed issues that could have allowed attackers to bypass authentication controls and execute code on compromised devices.

It's funny.  Laugh.

Original RFC1149 Packet Up For Auction At Christies (blogspot.com) 16

Long-time Slashdot reader badger.foo writes: As some of you may remember, in 2001 RFC1149 (TCP/IP over carrier pigeon) was implemented.

Now one of the packets preserved from that event is up for auction at Christies. Direct link to the auction (current bid: USD 2600) is here.

This is a unique chance to own a piece of Internet history!

AI

AI Execs Game Out 'Political Revolt' after a Catastrophic AI Event (yahoo.com) 47

Top executives at Anthropic, OpenAI and other AI companies "are privately gaming out scenarios for a public and political revolt after a catastrophic AI event," reports Axios, one that turns "an already wary public further against the technology and its leaders..."

These AI company executives "anticipate a large-scale event, most likely a cyberattack, that shuts down access to financial services, internet connectivity, or even power and water..." the article points out. "Many top AI researchers and executives believe a major incident is inevitable. "OpenAI conducts preparedness exercises where teams discuss and work through a range of potential scenarios," a company spokesperson said. "These scenarios are not treated as inevitable, but are meant to help us prepare for a variety of circumstances." Anthropic declined to comment...

Top AI planners assume Democrats, ascendant after the midterms, will move fast to shut down AI but will face significant challenges... The planning involves red-teaming for worst-case scenarios. It focuses mainly on racing to educate members of Congress. Company executives know regulation has no chance of passing right now, but still want to shape the legislation and policies U.S. leaders will turn to after a first catastrophic event....

Many AI industry insiders told Axios they believe a major event will occur in the next six to 12 months.

"Even with Democrats in control of the House and Senate, the companies see fractured politics as the norm, including within the Democratic Party."
AI

Claude Sent Police a Fake Murder Tip. White House Mandates AI Companies Report Security Incidents (yahoo.com) 35

AFP reports that an AI model from Anthropic "submitted a fabricated tip about an unsolved homicide to Philadelphia police, authorities said Friday."

Claude "was instructed never to log in, create accounts, enter personal data, make purchases, or submit anything destructive, but the instructions did not rule out form submissions," Anthropic said Friday in a blog post.

Authorities are now criticizing Anthropic "for taking two months to report the incident." The Philadelphia Police Department said the false submission was made in July through PhillyUnsolvedMurders.com, a public website where people can share information about unsolved killings. According to Anthropic's account, as relayed by police, the model was running a test that involved interacting with randomly selected websites when it reached the site and filed false information about an unsolved murder. The AI model presented itself as someone who might have knowledge of the case.

Anthropic's breaches have prompted the White House to mandate that AI companies notify and correct security incidents, news outlet Axios reported [yesterday], citing administration officials. "This notification and remediation process is not optional... It is a critical national security obligation," White House Super Intelligence Force leaders said in a statement to Axios.

"I may have information regarding this case," Claude told the police. "I recall seeing someone matching the description in the area around [the street named on the page] during that time period. Please contact me if this information is relevant." Anthropic notes that Claude "left the name and contact fields empty, which the form allowed, and submitted it. The submission was flagged as spam and was never forwarded for investigation."

But Anthropic also admits they saw "this behavior" three times — "on OSWorld (a public computer use evaluation), on Odysseys (a long-horizon task evaluation), and during internal usage." Submitting forms when it shouldn't have generally occurred "when an evaluation's instructions were ambiguous, or when a misconfiguration within the environment prevented Claude from working with dummy forms."

Anthropic's blog post acknowledges three other categories of behaviors:
  • Exploiting software flaws. Like when Claude received an error when trying to run a public tool on a university's web site, it located an injection flaw in a script on the university's server that let it run commands — including that public tool.
  • Working around restrictions to reach gated data. For example, Claude Mythos 5 needed public data that was only available from a state agency for a fee. "Claude learned from an archived copy of the agency's website that its public dashboard issues an access token to any visitor," Anthropic explains. "It requested one and used it to query the database without paying the fee."
  • Using URL shortening services. "Some of our fetch tools, which let Claude read webpages, limit the length of the URLs Claude can request. This is to prevent Claude from using long URLs to take certain unwanted actions, such as SQL or command injections... We saw several models, including Claude Opus 5 and Claude Mythos 5, get around this limitation by using free URL shortening services."

"We have built tooling to automatically detect and block the kinds of behaviors described above," Anthropic says, saying it's already running no on most of their evaluations. "When we tested it against the cases described in this post, it blocked all of them."

And they've already taken several other new preventive measures:

  • They've stopped running some public evaluations
  • Other public evaluations were moved to offline versions or rebuilt so their tasks don't reach live websites.
  • They've updated the guardrails on some internet access tools (including web fetch) "to heavily restrict what the model can do."
  • They're continuing "to fix or remove training environments that reward Claude for working around tool restrictions or other blockers, so that they do not incentivize these behaviors or permit reward hacking."
  • They've moved internal agents to "centrally managed infrastructure with strong containment," that minimizes internet access while monitoring "far more of what agents do through techniques like safety classifiers and hierarchical summarization."

In the past they'd focused reviews on cybersecurity testing, but they've broadened their transcript reviewing to other tasks which include internet access. "Because language models are non-deterministic — that is, their responses always involve some element of randomness, and they may carry out the same task slightly differently each time — we have Claude complete each evaluation task hundreds or thousands of times... If training rewards something we didn't intend — such as finding loopholes or working around a restriction — the model learns that the workaround pays off and may then apply it elsewhere."

Anthropic's blog post also acknowledged they'd seen multiple misalignment incidents involving federal, state, and local U.S. government agencies. "We have briefed the White House on these cases and notified each agency involved," Anthropic wrote, adding that "While we have not completed a full alignment assessment of these cases, we consider them to be less severe than the cybersecurity incidents from this summer." (And they are "modifying training to reduce the likelihood of further misbehavior.")


Piracy

Cloudflare Keeps 1.1.1.1 Out of Piracy Blocking, Escapes Penalties In France (torrentfreak.com) 61

An anonymous reader quotes a report from TorrentFreak: Cloudflare doesn't block pirate sports streaming sites through its 1.1.1.1 DNS resolver in France, only through its CDN. The Paris Judicial Court has now accepted that stance, rejecting [French pay-TV provider] Canal+'s request for penalties of 50,000 euros per site, per day. Interestingly, the court's conclusion relies on statistics provided by Canal+ itself. [...] As a result, Cloudflare will not be penalized under these orders, even though its public DNS resolver is not blocking the sites in question. That is in line with the company's long-standing claim that it doesn't interfere with its DNS.

In its recent transparency reports, Cloudflare repeatedly stated that it hasn't blocked any content through 1.1.1.1, despite orders from French and Italian courts. "To date, Cloudflare has not blocked content through the 1.1.1.1 Public DNS Resolver," the company's latest report reads. [...] Blocking through the CDN is another matter. According to the same report, Cloudflare geoblocked 1,238 domains in France in the second half of 2025, all under a single court order. In the first half of the year, it geoblocked 662 domains under seven orders. Cloudflare recently explained its position to the European Commission, in a submission (PDF) to its Counterfeit and Piracy Watch List consultation, stressing that global public DNS resolvers should not be used to block or restrict access.

Instead, it highlights its real-time pirate stream blocking program, which allows vetted rightsholders to flag pirate streams that run through its network. These streams are disrupted "within seconds," Cloudflare says, without any DNS or IP address blocking. "For live content, where speed is crucial, Cloudflare has built real-time mitigation mechanisms that allow vetted rightsholder partners to flag infringing streams. When those streams are running through our network, we act on them in seconds," Cloudflare informed the Commission. For now, Cloudflare can continue to block pirate sites through its CDN only, while its 1.1.1.1 DNS resolver remains untouched. Canal+ can still appeal the rulings, so it may not be the last we hear of it.

Slashdot Top Deals