<h2>Must know</h2>
<ul>
<li><strong>Chrome zero-day under active exploitation</strong>: Google released an urgent update for Chrome to patch CVE-2026-2441, a high-severity use-after-free vulnerability in its CSS font feature parsing that is actively being exploited. This is Chrome’s first zero-day of the year, and it could lead to arbitrary code execution within the browser sandbox. The fix involves preventing modification of the underlying map during traversal, and users are advised to update immediately. <a href="https://app.daily.dev/posts/ISqaRw1RF" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Microsoft Copilot Chat exposed confidential emails</strong>: Microsoft confirmed a security bug in Copilot Chat (CW1226324) that allowed the AI to bypass confidentiality labels and Data Loss Prevention policies, summarizing emails from users’ Sent Items and Drafts folders without authorization. The flaw was active since late January, affecting Microsoft 365 business customers, and highlights ongoing enterprise anxiety about AI tools accessing sensitive corporate data. <a href="https://app.daily.dev/posts/jSmIdhAQU" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Critical vulnerabilities in popular VS Code extensions</strong>: OX Security disclosed critical vulnerabilities in four popular VS Code extensions—Live Server, Code Runner, Markdown Preview Enhanced, and Microsoft’s Live Preview—totaling 128 million installs. These flaws enable file theft, remote code execution, and local network reconnaissance. While Microsoft silently patched Live Preview, three other maintainers never responded, leaving millions of developers at risk. <a href="https://app.daily.dev/posts/IuDICegHN" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>AI agent orchestration and security are top concerns</strong>: The CNCF Observability Summit will feature sessions on AI and Model Context Protocol (MCP) in observability, while Cisco is developing AI Defense for model scanning and agent identity frameworks, highlighting the growing focus on securing and managing AI agents. <a href="https://app.daily.dev/feed-by-ids?id=DKt3wvmEv&amp;id=B8WCrtkB4&amp;id=QrNlwTjqt" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI productivity gains remain elusive for most businesses</strong>: An NBER survey of nearly 6,000 executives found over 80% reported no measurable impact from AI on productivity or employment over the past three years, despite 69% of businesses using some form of AI. <a href="https://app.daily.dev/posts/rfmYnNc4e" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Google launches Lyria 3 AI music generation in Gemini</strong>: Google integrated Lyria 3, an AI music generation model, into Gemini Tools and the Gemini app, allowing users to create 30-second songs with lyrics from text, photos, or videos, complete with SynthID watermarking. <a href="https://app.daily.dev/posts/eN6VW0Hvt" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>India pushes for AI superpower status with massive investments</strong>: India is investing over $100 billion by 2035 to build 5 GW of renewable-powered AI datacenters and develop sovereign AI infrastructure, with companies like Sarvam AI launching open-source multilingual LLMs and deploying edge AI models on various devices. <a href="https://app.daily.dev/feed-by-ids?id=8IlGUVGaS&amp;id=JaYsn9TtR&amp;id=NDuZu47ZF&amp;id=TI4zH4l6O&amp;id=W4mWB5Hp1" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Databricks launches custom agents for production AI workflows</strong>: Databricks released Custom Agents (formerly Agent Framework), enabling developers to build, test, and deploy production-quality AI agents as fully managed Databricks Apps on serverless compute, with built-in memory and enterprise data access. <a href="https://app.daily.dev/posts/QJn2roYcs" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Tailscale introduces peer relays and security integrations</strong>: Tailscale announced general availability of Peer Relays for improved network performance and launched integrations with Huntress EDR and Fleet MDM to automate device posture enforcement in access policies. <a href="https://app.daily.dev/feed-by-ids?id=Q0uzlfbeJ&amp;id=L7iJSIIDS&amp;id=VukD3HNtV" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>New developer tools for cloud deployment and streaming</strong>: Deno Deploy introduced <code>deno sandbox</code> for instant Linux microVM provisioning, TinyIce offers a lightweight Icecast2-compatible audio streaming server, and Konta brings GitOps workflows to Docker Compose on low-resource VPS environments. <a href="https://app.daily.dev/feed-by-ids?id=7Ar3LQ3uM&amp;id=BHwbWnzAf&amp;id=QRFLXVuPF" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>Anthropic’s Claude C compiler marketing is a masterclass in spin</strong>: Anthropic’s marketing video claims Claude built a C compiler ‘from scratch’ with ‘no human intervention’ that could compile Linux. Turns out, it used 37 years of GCC test suites, an online GCC oracle, required human restarts, cost $20,000 in API calls, and couldn’t even boot Linux. But hey, 16 AI agents running for two weeks to produce a 100,000-line compiler is still pretty neat, if only they’d just said that. <a href="https://app.daily.dev/posts/0csgEkRMy" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI metrics: measuring tool usage instead of actual value</strong>: A developer at an unnamed company is calling out a new internal metric mandating 80% of epics use AI tooling, arguing it measures tool usage rather than actual business outcomes. Because nothing says ‘innovation’ like forcing AI into every corner, regardless of whether it actually helps or just generates more ‘AI slop’ for everyone else to clean up. <a href="https://app.daily.dev/posts/8WhnSQAtX" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>The Pentagon’s crazy reason for punishing Anthropic</strong>: The Pentagon is threatening to label Anthropic a national security risk because, after Claude was allegedly used in a military operation in Venezuela, someone at Anthropic dared to ask the US government about it. Apparently, asking questions about your AI being used in a military operation is now considered ‘condemnation’ and grounds for a feud. Who knew transparency was such a threat? <a href="https://app.daily.dev/posts/9eGp6hlhZ" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
