<h2>Must know</h2>
<ul>
<li><strong>Chrome zero-day under active exploitation</strong>: Google has released an urgent update for Chrome (versions 145.0.7632.75/76 for Windows/Mac and 144.0.7559.75 for Linux) to patch CVE-2026-2441, a high-severity use-after-free vulnerability in its CSS font feature parsing. This is Chrome’s first zero-day of 2026 and is actively being exploited in the wild, allowing for arbitrary code execution within the browser sandbox. Update immediately. <a href="https://app.daily.dev/posts/ISqaRw1RF" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Microsoft Copilot Chat exposes confidential emails</strong>: A security bug (CW1226324) in Microsoft Copilot Chat 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, active since late January, highlights ongoing enterprise concerns about AI tools accessing sensitive corporate data, with a fix rolling out in early February. <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. Developers are advised to update Live Preview to 0.4.16+, disable unused extensions, and avoid browsing untrusted sites with localhost servers active. <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>Anthropic Sonnet 4.6 released with performance claims</strong>: Anthropic released Claude Sonnet 4.6, a mid-tier model priced significantly lower than Opus 4.6, claiming it matches or exceeds Opus performance on key enterprise benchmarks, though independent analysis suggests it uses over 4x more tokens, potentially negating cost savings. <a href="https://app.daily.dev/feed-by-ids?id=3TzVcx3Vs&amp;id=i4O6sQBIw" 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 and AI-generated cover art from text prompts, photos, or videos. <a href="https://app.daily.dev/posts/eN6VW0Hvt" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI agent development platforms receive major updates</strong>: LangSmith Agent Builder released updates including a central ‘Chat’ agent for workspace tools and unified tool registry, while Databricks launched Custom Agents for building, testing, and deploying production-quality AI agents on serverless compute. <a href="https://app.daily.dev/feed-by-ids?id=4idREqj8d&amp;id=QJn2roYcs" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>GitHub Copilot expands AI model and Windows support</strong>: Claude Opus 4.6 is now generally available within GitHub Copilot across various IDEs, and the Copilot coding agent, previously Linux-only, now supports Windows development environments. <a href="https://app.daily.dev/feed-by-ids?id=49jCLhThU&amp;id=xntn7IAsM" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Deno Deploy introduces instant Linux microVMs</strong>: Deno Deploy launched <code>deno sandbox</code>, a new subcommand for provisioning cloud-hosted Linux microVMs and SSH-connecting in under a second, complete with a secrets mechanism for secure outbound requests. <a href="https://app.daily.dev/posts/7Ar3LQ3uM" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Laravel releases new Svelte starter kit</strong>: Laravel 12 now offers an official Svelte starter kit, joining React, Vue, and Livewire options, utilizing Inertia.js, Svelte 5, TypeScript, Tailwind CSS, and shadcn-svelte components. <a href="https://app.daily.dev/posts/J0baIl5WX" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI agent ecosystem navigates policy shifts and data challenges</strong>: Anthropic clarified its Claude Code documentation regarding OAuth token usage in third-party tools, while OpenAI partnered with Indian fintech Pine Labs to embed AI APIs, and Amazon shared a framework for evaluating agentic AI systems. <a href="https://app.daily.dev/feed-by-ids?id=1UIQguzNO&amp;id=3lsZ5Vula&amp;id=D45i7bkfM&amp;id=EZOdc00Xc&amp;id=FQxtQgg9R&amp;id=FTWE8YzDG&amp;id=JlixRkBbI&amp;id=JqU9WLooM&amp;id=LKtTxw05m&amp;id=R2tIcO414" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>AI productivity impact: much ado about nothing</strong>: An NBER survey of nearly 6,000 executives found over 80% reported no measurable impact from AI on productivity or employment, despite 69% of businesses using some form of AI. So, all that hype about AI revolutionizing the workplace? Turns out it’s mostly just… hype, for now. But don’t worry, they still anticipate AI will eliminate 1.75 million jobs by 2028, so there’s that to look forward to. <a href="https://app.daily.dev/posts/rfmYnNc4e" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Anthropic’s C compiler: a triumph of marketing over reality</strong>: Anthropic proudly announced Claude built a C compiler ‘from scratch’ with ‘no human intervention,’ capable of compiling Linux and Doom. What they conveniently left out was the $20,000 in API calls, 37 years of GCC test suites, human restarts, and the fact it couldn’t actually boot Linux. It’s a great achievement for AI, but the marketing department clearly got a little too creative with the truth. <a href="https://app.daily.dev/posts/0csgEkRMy" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI-generated code: the dumbest metric is back</strong>: Apparently, AI is making Lines of Code (LOC) a ‘dangerous comeback’ as a metric, despite being discredited by computer science pioneers. Because nothing says ‘progress’ like generating mountains of ‘instant legacy code’ that nobody fully understands, all while accelerating development without improving judgment. It’s like giving a toddler a chainsaw and calling it efficiency. <a href="https://app.daily.dev/posts/ySpIGlBP7" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
