<h2>Must know</h2>
<ul>
<li><strong>Meta pivots from metaverse to AI, cuts Reality Labs staff</strong>: Meta is closing three VR studios and laying off 10% of its Reality Labs workforce, approximately 1,500 employees, after the division lost over $70 billion since 2021. This strategic shift moves investment from VR and metaverse initiatives towards AI-powered glasses and wearable technology, acknowledging sustained financial losses and slow market growth for VR hardware. CTO Andrew Bosworth has scheduled an urgent meeting with Reality Labs staff to address the changes, as reported by multiple sources. <a href="https://app.daily.dev/posts/7ystcAQPc" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Microsoft’s AI data center expansion faces backlash</strong>: Microsoft is expanding its AI data center footprint, but faces significant community backlash in 24 U.S. states over environmental and social impacts. In response, Microsoft launched a “Community-First AI Infrastructure” initiative, promising to cover all power costs, minimize water usage, create local jobs, and pay full property taxes. This move aims to prevent residential electricity bill increases due to rising grid demand, a concern highlighted by the IEA’s projection that global data center electricity demand will more than double by 2030, as detailed in various reports. <a href="https://app.daily.dev/posts/jqtBl3Jnb" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Anthropic restricts Claude Code tokens for third-party tools</strong>: Anthropic has begun strictly enforcing its terms of service, preventing the use of Claude Code subscription tokens with third-party tools like Cursor and Open Code. This means developers must now use Claude Code exclusively or pay significantly higher API prices when integrating with other platforms. The move has sparked controversy among developers who had built workflows around these integrations, with many speculating it’s a strategic effort by Anthropic to lock users into its ecosystem and manage infrastructure costs, according to technical reporting. <a href="https://app.daily.dev/feed-by-ids?id=DEqFrm707&amp;id=jcpFkjQkx" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>Apple launches Creator Studio subscription for creative pros</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, priced at $12.99/month or $129/year, with a student/educator discount. <a href="https://app.daily.dev/posts/gpfpGVE8Y" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Slackbot re-emerges as an AI agent for enterprises</strong>: Salesforce has relaunched Slackbot as a fully redesigned AI agent for Business+ and Enterprise+ customers, capable of answering questions, organizing work, and scheduling meetings using natural language, integrating with third-party tools like Microsoft Teams and Google Drive. <a href="https://app.daily.dev/posts/VpXLIsjSp" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>New Linux malware targets cloud and container infrastructure</strong>: Check Point Research identified VoidLink, a new Linux malware framework written in Zig, designed to infiltrate cloud and container environments (AWS, Azure, GCP, Docker, Kubernetes) using advanced rootkit techniques and a modular plugin system for credential harvesting and lateral movement. <a href="https://app.daily.dev/posts/uJaIfhgHH" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>WordPress 7.0 to drop PHP 7.2 and 7.3 support</strong>: The WordPress core development team plans to discontinue support for PHP 7.2 and 7.3 in the upcoming WordPress 7.0 release, with a meeting scheduled for January 14, 2026, to finalize details and call for volunteers. <a href="https://app.daily.dev/posts/xho5u0zys" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Node.js releases security updates</strong>: Node.js 20.20.0 (LTS) and 25.3.0 (Current) were released on January 13, 2026, addressing multiple security vulnerabilities, including disabling futimes with the permission model enabled and tightening symlink API permissions. <a href="https://app.daily.dev/feed-by-ids?id=DwsoMgMKH&amp;id=Wy4br4Zyw&amp;id=Zr0Dv9dwZ" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Bun 1.3.6 introduces tarball handling and performance boosts</strong>: Bun v1.3.6, released January 13, 2026, adds Bun.Archive for tarball creation/extraction, a JSONC parser, and significant performance improvements, including 3.5x faster Response.json() and 20x faster Bun.hash.crc32. <a href="https://app.daily.dev/feed-by-ids?id=Wx4br4Zyw&amp;id=Yy5NFIIOM" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Laravel Simplifier plugin for Claude Code released</strong>: Taylor Otwell released the Laravel Simplifier plugin for Claude Code, an open-source tool adapted from Anthropic’s Code Simplifier, designed to review code commits for adherence to Laravel standards and PSR-12 guidelines. <a href="https://app.daily.dev/posts/Z70ay5WSd" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>AI agents are the future, unless they fail 40% of the time</strong>: Gartner, in its infinite wisdom, predicts 40% of agentic AI projects will be canceled by 2027, not because the AI is inherently flaky, but because companies can’t be bothered to implement ‘guardrails.’ Apparently, treating an AI like a human employee with ‘strict governance’ is the secret sauce. Who knew that unchecked automation could lead to costly errors, like a chatbot agreeing to sell a $76,000 car for $1? Shocking. <a href="https://app.daily.dev/feed-by-ids?id=9X1ifJmCu&amp;id=QnCKryB4" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Tailwind CSS revenue drops 80% due to AI, because who needs docs when you have hallucinations?</strong>: Tailwind CSS is apparently seeing an 80% revenue drop because AI models are scraping and caching their documentation, directly impacting their monetization. And their brilliant solution? Rejecting a pull request that would improve LLM parsing, because protecting direct traffic is more important than, you know, adapting to the future. It’s almost like they didn’t see this coming when everyone started feeding their entire internet to large language models. <a href="https://app.daily.dev/feed-by-ids?id=2a8HRxzCb&amp;id=Ml47Y2Hd6&amp;id=g50WJmopi" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Cloudflare fined €14.2 million for not blocking pirate sites, because ‘no way’ is not an option</strong>: Italy’s AGCOM fined Cloudflare €14.2 million for refusing to block pirate sites on its 1.1.1.1 DNS resolver, citing the Piracy Shield law. Cloudflare CEO Matthew Prince is now crying to the U.S. government and the IOC, claiming filtering would increase latency and threaten Olympic cybersecurity. It’s a classic move: when faced with a regulation you don’t like, threaten to pull out and warn of impending doom. Because apparently, a DNS resolver is too delicate to handle a simple block list. <a href="https://app.daily.dev/posts/Q0IbcQd4P" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
