<h2>Must know</h2>
<ul>
<li><strong>New lawsuits target AI companies for copyright infringement</strong>: A group of authors, including John Carreyrou, has filed a lawsuit against major AI firms like Anthropic, Google, OpenAI, Meta, xAI, and Perplexity. They allege unauthorized use of copyrighted books to train AI models, rejecting a prior $1.5 billion settlement offer as insufficient. This legal action highlights growing concerns over data sourcing practices in the AI industry and the perceived exploitation of intellectual property, as reported by various sources. <a href="https://app.daily.dev/posts/Lxx8mMcVQ" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>UEFI firmware vulnerability exposes major motherboard brands</strong>: A critical UEFI firmware vulnerability affects motherboards from Gigabyte, MSI, ASUS, and ASRock, allowing direct memory access (DMA) preboot attacks via malicious PCI devices. The flaw causes firmware to incorrectly report DMA protection as enabled, even when the memory management unit isn’t properly initialized. Riot Games’ Vanguard anti-cheat software identified this issue, leading to Valorant blocking affected systems until manufacturers release patches. <a href="https://app.daily.dev/posts/MlW9kdUo9" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Sophisticated credential theft campaigns target major platforms</strong>: Multiple credential theft campaigns are exploiting supply chains and platform trust. The U.S. Department of Justice seized a domain involved in a $14.6 million bank takeover scheme, while a malicious npm package, ‘lotusbail,’ exfiltrated WhatsApp data for six months. Additionally, ‘Phantom Shuttle’ Chrome extensions have been stealing credentials since 2017, and MacSync Stealer bypassed Apple’s notarization, highlighting persistent security gaps across various platforms. <a href="https://app.daily.dev/feed-by-ids?id=Nr5aLncd4&amp;id=V5siJoBfE&amp;id=Vvy6yzAJF&amp;id=pPs6THkFj&amp;id=w6fKBEdwQ" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>Linux kernel security vulnerabilities persist in cloud native platforms</strong>: The Linux kernel remains a significant security vulnerability in cloud native infrastructure, with containers sharing a common kernel and new CVEs constantly emerging, leading to slow patching and persistent exposure. <a href="https://app.daily.dev/feed-by-ids?id=DzBRqDVDY&amp;id=EFBPr81el&amp;id=FsbObvStu" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Texas sues smart TV makers over ACR tracking and privacy violations</strong>: Texas has filed lawsuits against LG, Hisense, and TCL, alleging unauthorized surveillance through Automated Content Recognition (ACR) on smart TVs, claiming manufacturers concealed data collection within lengthy terms of service. <a href="https://app.daily.dev/posts/N0oS44vYI" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Meta adopts Valve’s Linux scheduler and expands ExecuTorch for AI</strong>: Meta is integrating Valve’s Steam Deck Linux scheduler into its server infrastructure for performance optimization and expanding ExecuTorch, a unified PyTorch framework, to power on-device AI across Instagram, WhatsApp, and Quest 3. <a href="https://app.daily.dev/feed-by-ids?id=KhY7fDm4w&amp;id=i4MTohvIi" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Intel prepares Panther Lake H SoCs for next-gen laptops</strong>: Intel Linux driver updates indicate preparation for up to 13 different Panther Lake H SoC variants, targeting next-generation premium and high-end performance laptop markets, with a market introduction expected in 2025. <a href="https://app.daily.dev/posts/VF6MYJgOi" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI coding agents reshape developer roles and productivity</strong>: AI coding agents are shifting developer roles towards higher-order design and decision-making, with studies showing that while AI boosts individual productivity, organizational ROI requires holistic workflow redesign and strong developer experience foundations. <a href="https://app.daily.dev/feed-by-ids?id=2Ocmx4zuF&amp;id=AERV9p9ZK&amp;id=bEXnPrZ9F&amp;id=hbJeadAQC&amp;id=ylH8lM3L5" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AWS announces major AI infrastructure and tooling updates</strong>: AWS re:Invent 2025 featured new ultra servers with Nvidia GB300 chips, Nova Forge for custom AI model building, reinforcement learning for Bedrock, serverless model customization in SageMaker AI, and updates to AWS Agent Core for safer AI agent development. <a href="https://app.daily.dev/feed-by-ids?id=YQgj0lyer&amp;id=g5wVaTnGE" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Remote DoS attacks against NAT networks discovered</strong>: Research from NDSS 2025 uncovered critical vulnerabilities in NAT networks, allowing remote attackers to identify NAT devices and terminate TCP connections by exploiting weaknesses in Path MTU Discovery and inadequate packet validation. <a href="https://app.daily.dev/posts/inUJxcIuR" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>AI safety and regulation: still a wild west, apparently</strong>: Despite LLMs being used for sensitive subjects like mental health, there’s still no industry-wide safety benchmark or regulatory oversight. Social media, particularly X, continues to amplify exaggerated AI claims, like GPT-5 solving math problems it just ‘retrieved’ solutions for. So, while the hype machine churns, actual safety and reliable performance are still playing catch-up, leaving us all to wonder what could possibly go wrong. <a href="https://app.daily.dev/feed-by-ids?id=LOqeM5yNl&amp;id=ae2tFYiIl" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI-generated ‘slop’ is the new art form, say creators</strong>: Apparently, the internet is being flooded with AI-generated video ‘slop’ from tools like Sora and Veo, leading to a ‘degradation’ of online content. But don’t worry, some creators are calling it an ‘emergent creativity form,’ comparing it to historical resistance against new artistic technologies. Because nothing says ‘art’ like mass-produced, algorithmically-generated content that critics say diminishes quality and threatens traditional jobs. It’s just like photography, but with more existential dread. <a href="https://app.daily.dev/feed-by-ids?id=wTuAZcV1H&amp;id=xDVCiJUDK" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Nvidia’s AI pivot: good for them, bad for your wallet</strong>: Nvidia’s gaming revenue has plummeted to less than 8% of its total, with AI data centers now raking in the big bucks. This shift, driven by insatiable AI demand, means rising memory prices and a reduced supply of gaming GPUs, pushing future RTX cards above $2,000. So, while Nvidia makes bank on AI, gamers get to pay more for less, or just switch to cloud gaming. Thanks, AI, for making PC gaming even more exclusive! <a href="https://app.daily.dev/feed-by-ids?id=T86dDQ8DE&amp;id=Xqr5NpNtt" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
