Arch Linux AUR Malware Breach Over 1,500 packages in the Arch Linux User Repository (AUR) were compromised with malware. While developers have deleted known malicious commits, the total number of affected packages may exceed the officially listed count. https://www.phoronix.com/news/Arch-Linux-AUR-More-Than-1500

Arch Linux AUR Malware Attack Over 400 community-maintained packages in the Arch Linux User Repository (AUR) were compromised with malware. While official Arch Linux packages remain secure, the AUR’s trust model is now under scrutiny as maintainers work to ban affected accounts and purge malicious content. https://www.phoronix.com/news/Arch-Linux-AUR-400-Compromised Ladybird Browser Ends Public PRs The Ladybird browser project has ceased accepting public pull requests, citing the ease with which AI tools can now produce convincing but malicious or low-quality code. To ensure security and architectural integrity as they approach an alpha release, only project maintainers are now permitted to introduce code changes. ...

AI Killers: First Confirmed Autonomous Drone Deaths Fully autonomous ‘Terminator’ drones have killed Russian soldiers in Ukraine, marking a watershed moment in warfare. In a one-off test, AI-controlled quadcopters were programmed to independently search for and destroy targets in a specific area without any human connection or video feed, removing human judgment from the lethal decision process. https://www.newscientist.com/article/2529849-fully-autonomous-drones-have-killed-human-soldiers-for-the-first-time Google Held Liable for AI Hallucinations A German court ruled that Google is directly liable for false claims made by its AI Overviews, rejecting the argument that users are responsible for fact-checking. The ruling stems from the AI falsely linking publishers to scams, concluding that because Google’s AI rewrites and structures information in its own words, it owns the resulting statements. ...

German Court Holds Google Liable for AI Hallucinations A German regional court ruled that Google is directly liable for defamatory false claims made by its AI Overviews. The court rejected Google’s defense that users should verify AI output, stating that because the AI rewrites content in its own words, it is the owner of that content and thus responsible for its accuracy. https://the-decoder.com/landmark-german-ruling-declares-googles-ai-overviews-are-googles-own-words-and-makes-it-liable-for-false-answers/ https://arstechnica.com/tech-policy/2026/06/nobody-needs-ai-to-search-the-internet-court-says-in-ruling-against-google/ Disgruntled Researcher Drops Windows Defender Zero-Day A bug hunter known as ‘Nightmare Eclipse,’ claiming to be a former Microsoft employee, has released a proof-of-concept exploit for a zero-day vulnerability called ‘RoguePlanet.’ The researcher is publicly disclosing flaws as revenge for alleged mistreatment and ignored reports by Microsoft. ...

OpenCV 5: A Massive Leap for Computer Vision OpenCV 5 modernizes the library for the AI era, enabling developers to run large vision models and LLMs directly within the library without external runtimes. It significantly boosts performance and model compatibility, making high-end AI capabilities more accessible for edge and production systems. https://opencv.org/opencv-5/ Miasma: The All-in-One Supply Chain Attack Toolkit A powerful new attack toolkit called Miasma has been open-sourced, revealing how hackers are bypassing GitHub protections to steal credentials from AI developers. The toolkit turns compromised accounts into a self-perpetuating ‘flywheel’ of infections, highlighting a critical vulnerability in the modern AI-assisted development pipeline. ...

Meta Purges Secret Face-Recognition Code from Smart Glasses Meta stripped unreleased face-recognition software from its smart glasses companion app after reports revealed the system could create biometric ‘faceprints’ of strangers. Despite executive claims that the feature was merely exploratory, the code included machinery to crop and index unrecognized faces for future processing. https://www.wired.com/story/meta-removes-face-recognition-code-meta-ai-app-smart-glasses/ https://tech.yahoo.com/ai/meta-ai/articles/meta-furious-over-bombshell-smart-151912446.html AI Data Centers Spark Nationwide US Backlash A growing ‘data center disdain’ movement is seeing working-class neighborhoods resist AI infrastructure at five times the rate of wealthy ones. In Utah, intense public outcry over water depletion forced developer Kevin O’Leary to slash a massive project’s footprint by 50%. ...

AI’s Massive Water Footprint Collides with US Drought AI data centers consumed roughly 264 billion gallons of water in 2025, coinciding with drought conditions affecting 63% of the US. UN scientists warn that by 2030, AI’s water needs could match those of 1.3 billion people, while its electricity demand may triple that of 650 million people across Pakistan, Bangladesh, and Nigeria. https://www.barchart.com/story/news/2339834/ai-data-centers-water-consumption-breaks-264-billion-gallons-in-2025-as-devastating-drought-hits-nearly-63-of-u-s https://unu.edu/inweh/news/environmental-cost-of-AIs-Enrgy-use-carbon-water-and-land-footprints Bots Now Outnumber Humans on the Internet For the first time in history, machines generate more web traffic than people, with bots now accounting for 57.5% of all HTTP requests. This ‘platform shift’ threatens the economic foundation of the digital advertising and SaaS industries, which were built on the assumption of human attention. ...

Meta’s Secret Face-Recognition for Smart Glasses Meta has quietly shipped face-recognition code to millions of phones via its AI app, despite publicly claiming the technology was still under consideration. The system, internally called ‘NameTag,’ can transform faces into biometric signatures and alert wearers when a recognized person is identified, effectively turning users into a distributed surveillance network. https://www.wired.com/story/meta-smart-glasses-face-recognition-nametag-connections/ Google Pays SpaceX $11B+ for AI Compute Google will pay SpaceX $920 million per month from 2026 to 2029 to access 110,000 NVIDIA GPUs. This ‘bridge capacity’ deal highlights the extreme scarcity of AI infrastructure and SpaceX’s pivot into a high-revenue compute landlord for AI giants like Google and Anthropic. ...

Bots Now Outnumber Humans on the Web For the first time in history, bot traffic has eclipsed human traffic online. This shift threatens the traditional ad-supported business model of the internet, as bots do not click ads, potentially forcing a transition toward charging AI services for access to human-generated content. https://www.nbcnews.com/tech/tech-news/bot-web-traffic-overtaken-human-web-traffic-data-shows-rcna348522 Anthropic Calls for Global AI Development Freeze Anthropic, now the world’s most valuable AI startup, is urging a global pause on frontier AI development. The company warns that the technology is approaching a tipping point where systems could self-evolve, creating an ‘arms control problem’ that could lead to catastrophic outcomes if not coordinated internationally. ...

AI-Driven Recursive Self-Improvement Accelerates Coding Anthropic reports a massive acceleration in AI development as agents move from simple code generation to autonomous execution and research. Internal data shows engineers are merging 8x more code per day than in 2024, and AI is now capable of solving complex, open-ended engineering tasks in hours that previously took humans days. https://www.anthropic.com/institute/recursive-self-improvement The Return of DDR4: AI Memory Crisis Forces Hardware Regression Unprecedented DRAM shortages driven by AI data centers are forcing PC manufacturers to restart DDR4 production. With DDR5 prices skyrocketing and supply vanishing, motherboard vendors and chipmakers like AMD and Intel are reviving legacy DDR4 platforms to keep entry-level PCs affordable through 2027. ...