<h2>Must know</h2>
<ul>
<li><strong>Microsoft responds to AI data center backlash</strong>: Facing widespread community opposition over energy and water use, Microsoft is pledging a “community-first” approach for its AI infrastructure. The company will pay full electricity grid costs, minimize water consumption, and create local jobs, rejecting property tax reductions. This initiative comes after protests in 24 states and the withdrawal from a Wisconsin data center project, as reported by multiple sources. The move aims to rebuild public trust amid projections that global data center electricity demand will more than double by 2030. <a href="https://app.daily.dev/posts/jqtBl3Jnb" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Critical Windows 0-day actively exploited</strong>: Microsoft has released an urgent patch for CVE-2026-20805, an actively exploited information disclosure vulnerability in Windows. This flaw allows attackers to leak memory addresses, potentially bypassing ASLR as part of larger attack chains. The U.S. Cybersecurity and Infrastructure Security Agency (CISA) has added this to its Known Exploited Vulnerabilities catalog, mandating federal agencies patch by February 3rd. This is part of Microsoft’s January 2026 Patch Tuesday, which addressed 112 CVEs, including several critical remote code execution flaws in Office. <a href="https://app.daily.dev/posts/Kwy6WAbQP" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Meta pivots from metaverse to AI, lays off staff</strong>: Meta is closing three VR studios (Twisted Pixel Games, Sanzaru Games, Armature Studio) and laying off approximately 10% of its Reality Labs division staff, impacting around 1,500 employees. This strategic realignment shifts focus from metaverse and VR initiatives towards AI glasses and wearable technology. Reality Labs has lost over $70 billion since Meta’s 2021 rebrand, with a $4.4 billion loss in Q3 2025 alone, according to multiple reports. The move signals a significant reduction in Meta’s metaverse ambitions after heavy investment. <a href="https://app.daily.dev/posts/7ystcAQPc" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>Anthropic invests in Python open source security</strong>: Anthropic is committing $1.5 million over two years to the Python Software Foundation (PSF) to enhance security across the Python ecosystem, focusing on CPython and PyPI with automated malware detection and a new malware dataset. <a href="https://app.daily.dev/posts/yCHi0TB7q" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>New Linux Mint release with GUI upgrades</strong>: Linux Mint 22.3 “Zena” has been released, featuring a redesigned, customizable application menu, an on-screen keyboard, enhanced Wayland support, and updated tools like the Nemo file manager with pausable operations and regex search. <a href="https://app.daily.dev/posts/ecK1ViBJk" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Node.js releases security fixes</strong>: Node.js versions 20.20.0 (LTS) and 25.3.0 (Current) have been released with security fixes addressing multiple CVEs, including improved error handling, stricter permission checks for symlink APIs, and refactored buffer creation. <a href="https://app.daily.dev/feed-by-ids?id=DwsoMgMKH&amp;id=uw83EGG5r" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Apple launches Creator Studio subscription bundle</strong>: Apple is launching Creator Studio on January 28, 2026, a subscription bundle for creative apps including Final Cut Pro, Logic Pro, and Pixelmator Pro, with AI-powered features and discounted student pricing. <a href="https://app.daily.dev/posts/gpfpGVE8Y" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Advanced Linux malware targets cloud and containers</strong>: Check Point Research identified VoidLink, an advanced malware framework written in Zig, designed for Linux cloud and container environments like Docker and Kubernetes, featuring 37+ modular plugins for credential harvesting and evasion. <a href="https://app.daily.dev/feed-by-ids?id=uJaIfhgHH&amp;id=x9wl564Rc" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Open-weight AI models show strong agentic capabilities</strong>: New open-weight AI models like GLM 4.7 and MiniMax M2.1 successfully built complex CLI task runners autonomously, demonstrating agentic behavior, self-testing, and debugging at significantly lower costs than frontier models. <a href="https://app.daily.dev/posts/p3khbP8jU" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>NVIDIA introduces new CUDA programming model</strong>: NVIDIA’s new Coutile programming model for CUDA simplifies GPU kernel development by allowing developers to partition work into logical blocks and tiles, automating memory management and thread allocation for improved productivity and portability. <a href="https://app.daily.dev/posts/cR0glwi9Z" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>Slackbot reborn as an AI agent, but only for the elite</strong>: Salesforce has ‘reimagined’ Slackbot as an AI agent, now leveraging generative AI for all its core functions like answering questions and scheduling meetings. But don’t get too excited, unless you’re a Business+ or Enterprise+ customer. Apparently, the AI magic is too good for the plebs on lower tiers. No word on transparency, privacy, or actual independent evaluations, just Salesforce’s internal adoption rates. Because, you know, self-reported success is the best kind of success. <a href="https://app.daily.dev/posts/VpXLIsjSp" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Pentagon embraces Grok AI, ignores its past antics</strong>: Defense Secretary Pete Hegseth is pushing to integrate Elon Musk’s Grok AI into the Pentagon’s classified and unclassified networks. This, despite Grok’s recent history of generating ‘inappropriate images’ and other controversies. Apparently, the Pentagon’s AI acceleration strategy means we can just gloss over minor details like ‘how will this not go horribly wrong?’ It’s almost as if they’re more concerned with being ‘cutting edge’ than, you know, actual security or reliability. What could possibly go wrong with an AI that hallucinates in a military context? <a href="https://app.daily.dev/feed-by-ids?id=KoB4l9TcK&amp;id=jRGkCZr4W" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI-generated book spam: the new frontier of ‘knowledge’</strong>: HiTeX Press has managed to publish over 800 technical books in a single year, all generated by AI, completely devoid of human review or fact-checking. One Starlark programming book was found to contain ‘significant fabrication,’ including hallucinated code. These masterpieces are selling for 8€ on platforms like Amazon. It’s a bold new world where quantity trumps quality, and ‘expertise’ is just a prompt away. Good luck finding actual information amidst the AI-generated noise. <a href="https://app.daily.dev/feed-by-ids?id=ILAkYeoup&amp;id=PmK6l9z8J" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
