<h2>Must know</h2>
<ul>
<li><strong>Meta builds gigawatt-scale AI clusters</strong>: Meta is building gigawatt-scale AI clusters named Prometheus, using a Backend Aggregation (BAG) layer to interconnect tens of thousands of GPUs with petabit-range bandwidth. This massive infrastructure investment highlights the intense competition and resource demands in the AI space, with companies like Anthropic also finalizing significant funding rounds to fuel their own AI development, as reported by sources like Import AI. This scale of investment and infrastructure is critical for advancing AI capabilities and will heavily influence future hardware trends and resource prioritization. <a href="https://app.daily.dev/feed-by-ids?id=L6vqc64qy&amp;id=eKCtso6St&amp;id=r0F63t9O4&amp;id=a2pgFNMBq" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>EU accuses Meta of WhatsApp AI antitrust violation</strong>: The EU is accusing Meta of antitrust violations for blocking third-party AI assistants like ChatGPT and Microsoft Copilot from WhatsApp, following policy changes that made Meta AI the sole assistant. This move could seriously harm the AI assistant market, and the European Commission is considering emergency measures. This situation underscores the ongoing regulatory scrutiny of tech giants and their control over platform ecosystems, especially as AI integration becomes more pervasive. <a href="https://app.daily.dev/posts/w9RLQf8YH" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Discord implements global age verification</strong>: Discord is rolling out global age verification, defaulting unverified accounts to “teen-by-default” settings that restrict access to sensitive content and direct messages from strangers. This follows lawsuits alleging minor exploitation on the platform and mirrors similar moves by Roblox, which is also facing government pressure in Australia over child safety concerns. These actions reflect increasing pressure on platforms to implement robust safety measures, impacting user experience and privacy. <a href="https://app.daily.dev/feed-by-ids?id=J7jpcebnc&amp;id=LeKezdNaM" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>AI coding assistants write 90% of code</strong>: AI coding agents like Composer 1.5, Claude Opus 4.6, and GPT-5.3 Codex are now capable of writing up to 90% of code, significantly boosting developer productivity but also raising concerns about cognitive atrophy and the need for “Digital Shokunin” practices. <a href="https://app.daily.dev/feed-by-ids?id=DwLEeopez&amp;id=FTpnNlc2S&amp;id=FX1ggxaUN&amp;id=GfBuGFYT2&amp;id=KExlc4RU1&amp;id=QdPD1orhu&amp;id=ZkpPhHe3M&amp;id=aSRhxb4Ut&amp;id=acBBuIfYQ&amp;id=doxSL4jOM" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI-enabled social engineering attacks on FinTech</strong>: North Korean threat actor UNC1069 used AI-generated deepfake video, compromised Telegram accounts, and fake Zoom meetings in a sophisticated attack against a FinTech entity, deploying seven malware families. <a href="https://app.daily.dev/feed-by-ids?id=wgv654vDT&amp;id=3qL5mblsY&amp;id=4NFnRjHAX&amp;id=7hFHkMPwg&amp;id=8wZDCxvNp&amp;id=9ObJeT7O1&amp;id=A8iF37C5E&amp;id=AaCpDyacF&amp;id=ByPKunTEa&amp;id=CZgK5a0B2" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Zscaler acquires SquareX for browser security</strong>: Zscaler acquired SquareX to integrate browser-native threat protection into its Zero Trust Exchange platform, aiming to secure unmanaged BYOD devices against phishing and AI data leakage. <a href="https://app.daily.dev/posts/cYMDggZgc" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Toyota releases open-source game engine Fluorite</strong>: Toyota Connected North America announced Fluorite, an open-source game engine built with C++ and Flutter, designed for automotive digital cockpits and optimized for low-end hardware. <a href="https://app.daily.dev/posts/FTwlzGL8G" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Rust Foundation addresses infrastructure costs</strong>: The Rust Foundation’s January 2026 update highlights strategic goals for 2026-2028, addressing infrastructure challenges like higher-than-expected AWS costs for <a href="http://crates.io" target="_blank" rel="noopener nofollow">crates.io</a> CloudFront traffic, and prioritizing an accreditation program for Rust training providers. <a href="https://app.daily.dev/posts/Rvs7vPPfT" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Redshift enhances automatic optimizations</strong>: Amazon Redshift now supports allocating extra compute resources for automatic optimization features like Automatic Table Optimization, Auto Vacuum, and Auto Analyze, allowing them to run without impacting user workloads. <a href="https://app.daily.dev/posts/7o3DKKn6E" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>MIT develops stretchable thermal material</strong>: MIT researchers discovered an olefin block copolymer that can reversibly double its thermal conductivity in 0.22 seconds when stretched, offering potential for adaptive thermal management in electronics and clothing. <a href="https://app.daily.dev/feed-by-ids?id=gjUxIUjHY&amp;id=pn7MCCqlT" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>Claude co-work plugins “wipe out” tech stocks</strong>: Anthropic introduced Claude Co-work with industry-specific plugins, and the article title claims this “Wiped Out $400 Billion In Tech Stocks.” While the body text merely mentions “triggering market concerns,” it seems a bit dramatic to attribute such a massive market shift to a plugin release. Perhaps the market just needed an excuse to panic, and Claude’s new toys were it. <a href="https://app.daily.dev/posts/QnkkrfiEd" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>OpenAI embraces ads for ChatGPT</strong>: OpenAI is reportedly introducing advertising to ChatGPT, a complete reversal of Sam Altman’s previous stance, because apparently, losing billions annually isn’t a sustainable business model. They’ve even hired a former Facebook mobile advertising executive. So, get ready for your AI to start subtly recommending products, because nothing says “unbiased intelligence” like monetized recommendations. <a href="https://app.daily.dev/feed-by-ids?id=5kN0YzGRy&amp;id=BoDPkFq8Y&amp;id=cX7pVnhKE" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI chatbots fail to improve medical advice</strong>: A new study from the University of Oxford found that AI chatbots like GPT-4o and Llama 3 don’t actually improve medical advice compared to search engines, often giving inconsistent recommendations. So, while everyone’s hyping AI for critical applications, it turns out for something as important as health, it’s still just a fancy search bar that sometimes makes things up. Shocking. <a href="https://app.daily.dev/feed-by-ids?id=1kwDX3PnJ&amp;id=1mNE3s7VJ&amp;id=DLtzBsMsI&amp;id=KxYxbzZtb&amp;id=L6vqc64qy&amp;id=PmuBwjJNz&amp;id=QTyq32sGV&amp;id=eu7ShA5er&amp;id=kYVouXXxj&amp;id=lk77RUHW7" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
