Suspected Chinese snoops use 2017 unpatched Windows flaw to spy on EU politicians

Cyber spies linked to the Chinese government exploited a Windows shortcut vulnerability disclosed in March – but that Microsoft hasn’t fixed yet – to target European diplomats in an effort to steal defense and national security details.

Security firm Arctic Wolf attributed the espionage campaign to UNC6384 (aka Mustang Panda, Twill Typhoon), and in research published Thursday detailed how the suspected PRC spies used social engineering and the Windows flaw to deploy PlugX malware against personnel attending diplomatic conferences in September and October.

“This campaign demonstrates UNC6384’s capability for rapid vulnerability adoption within six months of public disclosure, advanced social engineering leveraging detailed knowledge of diplomatic calendars and event themes, and operational expansion from traditional Southeast Asia targeting to European diplomatic entities,” the Arctic Wolf Labs threat research team said.

[…]

Zero Day Initiative threat hunter Peter Girnus discovered and reported this flaw to Microsoft in March, and said it had been abused as a zero-day as far back as 2017, with 11 state-sponsored groups from North Korea, Iran, Russia, and China abusing ZDI-CAN-25373 for cyber espionage and data theft purposes.

Blame ZDI-CAN-25373

The attacks begin with phishing emails using very specific themed lures around European defense and security cooperation and cross-border infrastructure development. Those emails delivered a weaponized LNK file which exploited ZDI-CAN-25373 (aka CVE-2025-9491), a Windows shortcut vulnerability, to let the attackers secretly execute commands by adding whitespace padding within the LNK file’s COMMAND_LINE_ARGUMENTS structure.

The malicious files, such as one named Agenda_Meeting 26 Sep Brussels.lnk, use diplomatic conference themes as lures along with a decoy PDF document, in this case displaying a real European Commission meeting agenda on facilitating the free movement of goods at border crossing points between the EU and Western Balkan countries.

The LNK file, when executed, invokes PowerShell to decode and extract a tar (tape archive) archive containing three files to enable the attack chain via DLL side-loading, a malware delivery technique favored by several Chinese government crews, including Salt Typhoon.

DLL sideloading exploits the Windows DLL search order by tricking an application into loading a malicious DLL instead of the legitimate one.

The three files include a legitimate, but expired, Canon printer assistant utility with a valid digital signature issued by Symantec. Although the certificate expired in April 2018, Windows trusts binaries whose signatures include a valid timestamp, so this allows the attackers to bypass security tools and deliver malware using DLL sideloading.

The malicious DLL functions as a loader to decrypt and execute the third file in the archive, cnmplog.dat, which contains the encrypted PlugX payload.

PlugX, which has been around since at least 2008, is a Remote Access Trojan (RAT) that gives attackers all the remote access capabilities including command execution, keylogging, file uploading and downloading, persistent access, and system reconnaissance.

“This three-stage execution flow completes the deployment of PlugX malware running stealthily within a legitimate signed process, significantly reducing the likelihood of detection by endpoint security solutions,” the researchers wrote.

[…]

Source: Suspected Chinese snoops weaponize unpatched Windows flaw • The Register

‘Keep Android Open’ Campaign Pushes Back on Google’s Restrictions on installing your own software

A “Keep Android Open” campaign is pushing back on new rules from Google that will reportedly block users from sideloading apps on Android phones.

It’s unclear who’s running the campaign, but a blog post on the free Android app store F-Droid is directing users to visit the campaign’s website, which urges the public to lobby government regulators to intervene and stop the upcoming restrictions.

“Developers should have the right to create and distribute software without submitting to unnecessary corporate surveillance,” reads an open letter posted to the site.

In August, Google announced upcoming rules aimed at reducing the risk of malware infecting Android devices. It will require that all app installs on certified Android devices only occur if the developer has been verified by Google. For individual developers, this includes submitting an official ID, address, phone number, and paying a $25 fee.

Source: ‘Keep Android Open’ Campaign Pushes Back on Google’s Sideloading Restrictions | PCMag

Austria’s Ministry of Economy goes toward digital sovereignty, embraces Nextcloud

How do you modernize digital collaboration in government without losing control over your data?
Austria’s Federal Ministry for Economy, Energy and Tourism (BMWET) faced this challenge in 2024 – and decided to take a clear, pragmatic step toward digital sovereignty. Within just four months between Proof of Concept and rollout, the ministry went live with a secure Nextcloud environment, operated on its own infrastructure in Austria and designed to meet strict transparency and compliance requirements.

Why the ministry decided to act

The BMWET faced the challenge of modernizing its collaboration tools at a time when questions around data protection, compliance, and technological dependency were becoming increasingly relevant.
A risk analysis showed that relying entirely on cloud services from non-European providers would introduce significant legal and security uncertainties. Especially regarding the protection of sensitive information under GDPR and the upcoming NIS2 directive.

[…]

At the time of the project, BMWET was already in the process of adopting Microsoft 365 and Teams. Reversing that path wasn’t realistic. Instead, the ministry implemented a hybrid architecture: Nextcloud handles internal collaboration and secure data management, while Teams remains available for external meetings.

In collaboration with Nextcloud partner Sendent, Outlook integration ensured seamless email and calendar workflows, enabling employees to continue working with familiar tools while ensuring sensitive information stays within Austrian infrastructure.

A fast and well-coordinated rollout

From proof of concept to full deployment, the process took only a few months – an unusually fast timeline for a public sector ICT project. Working closely with Nextcloud and implementation partner Atos Austria, BMWET rolled out the new platform across 1,200 employees.

[…]

Source: Austria’s Ministry of Economy takes decisive steps toward digital sovereignty – Nextcloud

Samsung Electronics Unveils Tri-Fold Phone: Two Hinges, Tablet Screen

[…] The unveiled Tri-Fold Phone differed from typical foldable phones in that it had two hinges, joints, in the folding section. Its screen ratio was significantly longer horizontally than vertically, resembling a ‘small tablet PC.’ Compared to conventional foldable phones, it featured three external screens aligned side by side. However, the device could not be touched or tested directly; it was only visible through a glass display case.

On the 28th, Samsung Electronics unveils the physical model of the 'Tri-Fold Phone, right' at the APEC K-Tech showcase. /Courtesy of Park Ji-min
On the 28th, Samsung Electronics unveils the physical model of the ‘Tri-Fold Phone, right’ at the APEC K-Tech showcase. /Courtesy of Park Ji-min

The Tri-Fold Phone, which Samsung Electronics announced would be released within the year, was designed with a structure allowing the screen to be folded twice. The external screen features a display measuring approximately 6.5 inches, similar to the Galaxy Fold series, while the fully unfolded screen measures approximately 10 inches, comparable to a tablet. Considering the complex hinge structure and additional display components, the launch price is estimated to be over 3 million Korean won.

[…]

Source: Samsung Electronics Unveils Tri-Fold Phone: Two Hinges, Tablet Screen

Clearview AI faces criminal heat for ignoring EU data fines – wait: these creeps still exist?

Privacy advocates at Noyb filed a criminal complaint against Clearview AI for scraping social media users’ faces without consent to train its AI algorithms.

Austria-based Noyb (None of Your Business) is targeting the US company and its executives, arguing that if successful, individuals who authorized the data collection could face criminal penalties, including imprisonment.

The complaint focuses largely on Clearview’s apparent disregard for fines from France, Greece, Italy, the Netherlands, and the UK. Aside from the UK — where Clearview recently lost its appeal of a $10 million fine from the Information Commissioner’s Office — the company has yet to pay other fines totaling more than $100 million, Noyb claims.

“EU data protection authorities did not come up with a way to enforce its fines and bans against the US company, allowing Clearview AI to effectively dodge the law,” said Noyb in its announcement today.

Max Schrems, privacy lawyer and founder of Noyb, said: “Clearview AI seems to simply ignore EU fundamental rights and just spits in the face of EU authorities.”

The criminal complaint, filed with Austrian public prosecutors, hinges on Article 84 of the GDPR, which allows EU member states to seek proportionate punishments for data protection violations, including through criminal proceedings.

Clearview AI claims it has collected more than 60 billion images to help law enforcement agencies improve facial recognition tech.

Scraping data is not inherently illegal, however, Clearview’s sweeping collection of social media photos for commercial gain has repeatedly violated GDPR regulations across Europe.

Austria ruled the company’s practices illegal in 2023, though it imposed no fine.

Noyb is using a provision in Austria’s own implementation of the GDPR that allows criminal proceedings to be brought against managers of organizations that flout data protection laws.

“We even run cross-border criminal procedures for stolen bikes, so we hope that the public prosecutor also takes action when the personal data of billions of people was stolen – as has been confirmed by multiple authorities,” said Schrems.

Source: Clearview AI faces criminal heat for ignoring EU data fines • The Register

Physicists Built a Pixel So Small, Full HD Could Fit on a Bread Crumb

[…] In a recent Science Advances paper, researchers report the creation of the smallest pixel ever, using optical antennas that convert radiation into focused energy bits. The pixel measures just 300 by 300 nanometers—around 17 times smaller than a conventional OLED pixel, but with a similar brightness.

To put the size into context, a display with an area of just one square millimeter could fit a resolution of 1920 x 1080 pixels using the new technology. The tiny pixel also glows on its own, making it potentially revolutionary for the next generation of smart, portable devices.

[…]

the team identified a way to effectively block these unwanted structures, called filaments, from potentially destroying the pixel.  Specifically, they fabricated a thin, insulating layer with a tiny circular opening at its center and layered it over a gold optical antenna.

Smallest Pixel Diagram
A conceptual illustration of the nanoscale OLED. © Hecht et al., 2025.

The arrangement proved surprisingly effective in preventing filaments from forming. The optical antenna additionally helped focus electromagnetic energy and amplify the brightness, according to the paper. As a result, “even the first nanopixels were stable for two weeks under ambient conditions,” said Bert Hecht, study senior author and a physicist at the University of Würzburg, in the release.

That said, the system is still a prototype, with about 1% efficiency. However, the researchers noted that because the current paper eliminates one of the biggest challenges of scaling down pixels, the next steps should be slightly easier.

“With this technology, displays and projectors could become so small in the future that they can be integrated almost invisibly into devices worn on the body—from eyeglass frames to contact lenses,” the researchers added.

Source: Physicists Built a Pixel So Small, Full HD Could Fit on a Bread Crumb

‘Change course now’: humanity has missed 1.5C climate target, says UN head

[…]Humanity has failed to limit global heating to 1.5C and must change course immediately, the secretary general of the UN has warned.

In his only interview before next month’s Cop30 climate summit, António Guterres acknowledged it is now “inevitable” that humanity will overshoot the target in the Paris climate agreement, with “devastating consequences” for the world.

He urged the leaders who will gather in the Brazilian rainforest city of Belém to realise that the longer they delay cutting emissions, the greater the danger of passing catastrophic “tipping points” in the Amazon, the Arctic and the oceans.

“Let’s recognise our failure,” he told the Guardian and Amazon-based news organisation Sumaúma. “The truth is that we have failed to avoid an overshooting above 1.5C in the next few years. And that going above 1.5C has devastating consequences. Some of these devastating consequences are tipping points, be it in the Amazon, be it in Greenland, or western Antarctica or the coral reefs.

He said the priority at Cop30 was to shift direction: “It is absolutely indispensable to change course in order to make sure that the overshoot is as short as possible and as low in intensity as possible to avoid tipping points like the Amazon. We don’t want to see the Amazon as a savannah. But that is a real risk if we don’t change course and if we don’t make a dramatic decrease of emissions as soon as possible.”

The planet’s past 10 years have been the hottest in recorded history. Despite growing scientific alarm at the speed of global temperature increases caused by the burning of fossil fuels – oil, coal and gas – the secretary general said government commitments have come up short.

Fewer than a third of the world’s nations (62 out of 197) have sent in their climate action plans, known as nationally determined contributions (NDCs) under the Paris agreement. The US under Donald Trump has abandoned the process. Europe has promised but so far failed to deliver. China, the world’s biggest emitter, has been accused of undercommitting.

António Guterres speaking at a Cop29 podium
António Guterres giving his speech at Cop29 in Baku, Azerbaijan, in November 2024. Photograph: Anatoly Maltsev/EPA

Guterres said the lack of NDC ambition means the Paris goal of 1.5C will be breached, at least temporarily: “From those [NDCs] received until now, there is an expectation of a reduction of emissions of 10%. We would need 60% [to stay within 1.5C]. So overshooting is now inevitable.”

He did not give up on the target though, and said it may still be possible to temporarily overshoot and then bring temperatures down in time to return to 1.5C by the end of the century, but this would require a change of direction at and beyond Cop30.

[…]

Source: ‘Change course now’: humanity has missed 1.5C climate target, says UN head | Climate crisis | The Guardian

China’s New Influencer Law Says Only Degree-Holders Can Discuss Professional Topics. Idiots say this limits freedom of ideas

China’s new influencer law, which took effect on October 25, requires anyone creating content on sensitive topics, such as medicine, law, education, or finance, to hold official qualifications in those fields.

The Cyberspace Administration of China (CAC) says the goal is to fight misinformation and protect the public from false or harmful advice. But, the move has also raised concerns about censorship and freedom of expression.

Under the new rules, influencers who talk about regulated topics must show proof of their expertise, such as a degree, professional license, or certificate. Platforms like Douyin (China’s version of TikTok), Bilibili, and Weibo must verify creators’ credentials and make sure their content includes proper citations and disclaimers.

For example, influencers must clearly state when information comes from studies or when a video includes AI-generated material. Platforms are also required to educate users about their responsibilities when sharing content online.

The CAC has gone even further by banning advertising for medical products, supplements, and health foods to prevent hidden promotions disguised as “educational” videos.

However, critics warn that the law could harm creativity and limit freedom of speech. By controlling who can talk about certain topics, they argue, China might not only block misinformation but also restrict independent voices and critical debate.

Many worry that “expertise” will be defined too narrowly, giving authorities more power to silence people who question official narratives or offer alternative views.

Others, however, welcomed the move, saying that the new law would allow for well-informed content on important and sensitive topics. Many argued that only professionals in their field should be able to speak about and discuss said topic to prevent misinformation.

The rise of influencer culture has changed how people get information. Influencers are valued for being relatable and authentic, and being able to connect with audiences in ways traditional experts cannot. However, when these creators share misleading or inaccurate information, the effects can be serious, supporters of the new law argue.

Source: China’s New Influencer Law Says Only Degree-Holders Can Discuss Professional Topics China’s New Influencer Law Says Only Degree-Holders Can Discuss Professional Topics

Unfortunately, having people doing “research” by watching one Youtube video and then telling people that vaccines don’t work, or that 5G space bats cause covid and people want to inject chips into you has proven to be an absolute disaster, which prolonged a global pandemic and killed a lot of people.

These people should be jailed and it is a crying shame that a country like China is taking the lead in this, and not the EU.

Python Foundation rejects $1.5M grant with no-DEI strings

[…]The programming non-profit’s deputy executive director Loren Crary said in a blog post today that the National Science Foundation (NSF) had offered $1.5 million to address structural vulnerabilities in Python and the Python Package Index (PyPI), but the Foundation quickly became dispirited with the terms of the grant it would have to follow.

“These terms included affirming the statement that we ‘do not, and will not during the term of this financial assistance award, operate any programs that advance or promote DEI [diversity, equity, and inclusion], or discriminatory equity ideology in violation of Federal anti-discrimination laws,'” Crary noted. “This restriction would apply not only to the security work directly funded by the grant, but to any and all activity of the PSF as a whole.”

To make matters worse, the terms included a provision that if the PSF was found to have violated that anti-DEI diktat, the NSF reserved the right to claw back any previously disbursed funds, Crary explained.

“This would create a situation where money we’d already spent could be taken back, which would be an enormous, open-ended financial risk,” the PSF director added.

The PSF’s mission statement enshrines a commitment to supporting and growing “a diverse and international community of Python programmers,” and the Foundation ultimately decided it wasn’t willing to compromise on that position, even for what would have been a solid financial boost for the organization.

“The PSF is a relatively small organization, operating with an annual budget of around $5 million per year, with a staff of just 14,” Crary added, noting that the $1.5 million would have been the largest grant the Foundation had ever received – but it wasn’t worth it if the conditions were undermining the PSF’s mission.

The PSF board voted unanimously to withdraw its grant application.

The non-profit would’ve used the funding to help prevent supply chain attacks; create a new automated, proactive review process for new PyPI packages; and make the project’s work easily transferable to other open-source package managers. […]

Source: Python Foundation rejects $1.5M grant with no-DEI strings • The Register

AI generates surge in expense receipt fraud

[…] Software provider AppZen said fake AI receipts accounted for about 14 per cent of fraudulent documents submitted in September, compared with none last year. Fintech group Ramp said its new software flagged more than $1mn in fraudulent invoices within 90 days.
About 30 per cent of US and UK financial professionals surveyed by expense management platform Medius reported they had seen a rise in falsified receipts following the launch of OpenAI’s GPT-4o last year.
An AI-generated receipt
An AI-generated receipt © AppZen
“These receipts have become so good, we tell our customers, ‘do not trust your eyes’,” said Chris Juneau, senior vice-president and head of product marketing for SAP Concur, one of the world’s leading expense platforms, which processes more than 80mn compliance checks monthly using AI.
Several platforms attributed a significant jump in the number of AI-generated receipts after OpenAI launched GPT-4o’s improved image generation model in March.
[…]

Source: ‘Do not trust your eyes’: AI generates surge in expense fraud

Qualcomm announces AI chips to compete with AMD and Nvidia

[…]Qualcomm said that both the AI200, which will go on sale in 2026, and the AI250, planned for 2027, can come in a system that fills up a full, liquid-cooled server rack.

Qualcomm is matching Nvidia and AMD

, which offer their graphics processing units, or GPUs, in full-rack systems that allow as many as 72 chips to act as one computer. AI labs need that computing power to run the most advanced models.

Qualcomm’s data center chips are based on the AI parts in Qualcomm’s smartphone chips called Hexagon neural processing units, or NPUs.

[…]

Qualcomm said its chips are focusing on inference, or running AI models, instead of training, which is how labs such as OpenAI create new AI capabilities by processing terabytes of data.

The chipmaker said that its rack-scale systems would ultimately cost less to operate for customers such as cloud service providers, and that a rack uses 160 kilowatts, which is comparable to the high power draw from some Nvidia GPU racks.

Malladi said Qualcomm would also sell its AI chips and other parts separately, especially for clients such as hyperscalers that prefer to design their own racks.

[…]

The company declined to comment, the price of the chips, cards or rack, and how many NPUs could be installed in a single rack.

[…]

Qualcomm said its AI chips have advantages over other accelerators in terms of power consumption, cost of ownership, and a new approach to the way memory is handled. It said its AI cards support 768 gigabytes of memory, which is higher than offerings from Nvidia and AMD.

[…]

Source: Qualcomm announces AI chips to compete with AMD and Nvidia

In stunning display of (in)stability, PayPal / Paxos stablecoin accidentally mints $300 Trillion(!) and then revokes it

Paxos Trust Company, the blockchain infrastructure partner for PayPal’s stablecoin, has publicly admitted to a catastrophic “technical error” that led to the accidental creation of $300 trillion worth of PayPal USD (PYUSD) tokens. The mistake, which was identified and rectified within minutes, temporarily created a theoretical sum exceeding the entire global money supply. This incident immediately triggers intense scrutiny from financial regulators, including the New York Department of Financial Services (NYDFS), and casts a shadow over the operational integrity of the burgeoning stablecoin market. For PayPal, the error represents a significant reputational blow, challenging the perception of its carefully managed entry into digital assets.

This failure represents a stark vulnerability in the automated systems underpinning digital assets. While blockchain technology promises immutable and transparent transactions, Paxos is now confronting the reality that its risk management protocols were insufficient to prevent a near-infinite minting event. The company’s promise to be “much better than this” highlights the critical gap between theoretical blockchain security and the practical operational controls required for regulated financial services. This matters because it demonstrates that for institutional adoption to proceed, the infrastructure must be as foolproof as the legacy financial systems it seeks to augment or replace, not a source of existential, self-inflicted risk.

For fintech executives and digital asset custodians, this is a critical warning. The forward-looking insight is clear: the path to mainstream stablecoin adoption will be paved with relentless focus on operational controls and third-party audits. This event will force a sector-wide review of minting and burning mechanisms, likely leading to more conservative, multi-signature requirements and real-time monitoring mandates from regulators. The most trusted players will be those who can transparently demonstrate ironclad technical and procedural safeguards, turning this public failure into an industry-wide mandate for bulletproof operational excellence.

Source: Paxos Admits $300T PayPal Stablecoin Minting Error

Part of what is not mentioned, is that they revoked the value very quickly. The minting is one thing, but how trustworthy can any value store be when it the value can be revoked one-sidedly at any time by the press of a button?