<h2>Must know</h2>
<ul>
<li><strong>Meta pivots from metaverse to AI infrastructure</strong>: Meta is closing three VR studios and laying off 10% of its Reality Labs staff, totaling about 1,500 employees, after losing over $70 billion since 2021. The company is shifting its focus and massive investments towards AI-powered glasses and wearable technology, with CTO Andrew Bosworth scheduling an urgent meeting to address the strategic reorientation. This move signals a significant departure from its previous metaverse-centric vision, as reported by multiple sources. <a href="https://app.daily.dev/posts/7ystcAQPc" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Cloudflare fined by Italy over DNS blocking refusal</strong>: Italy’s communications regulator, AGCOM, has 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 argues compliance is technically infeasible without impacting all users and is seeking U.S. government support, even threatening to withdraw its servers from Italy. CEO Matthew Prince has also notified the International Olympic Committee about potential cybersecurity risks for the 2026 Winter Olympics in Italy, framing the dispute as a broader issue than just copyright enforcement. <a href="https://app.daily.dev/posts/Q0IbcQd4P" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Microsoft’s AI data center expansion faces backlash</strong>: Microsoft’s ‘Community-First AI Infrastructure’ initiative, promising to cover power costs and minimize water usage for its U.S. data centers, is a direct response to widespread public backlash. Activist groups in 24 states are protesting the environmental and social impacts of these centers, leading Microsoft to abandon projects in Wisconsin and face resistance in Michigan and Ohio. The IEA projects global data center electricity demand will more than double by 2030, intensifying concerns about rising residential energy rates and strained local resources. <a href="https://app.daily.dev/posts/jqtBl3Jnb" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>Anthropic commits $1.5 million to Python security</strong>: Anthropic has pledged $1.5 million over two years to the Python Software Foundation (PSF) to enhance the Python ecosystem’s security, focusing on automated malware detection for PyPI packages and supporting CPython core development. <a href="https://app.daily.dev/posts/yCHi0TB7q" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>New Linux malware targets cloud and containers</strong>: Check Point Research identified VoidLink, a new Linux malware framework written in Zig, designed to infiltrate cloud and container environments like AWS, Azure, GCP, Docker, and 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>AI agents for enterprise Slack users</strong>: Salesforce has relaunched Slackbot as an AI agent for Business+ and Enterprise+ customers, leveraging context from Slack and third-party tools to answer questions, organize work, create content, and schedule meetings, with future updates planned for voice interaction and internet browsing. <a href="https://app.daily.dev/posts/VpXLIsjSp" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Node.js releases address security vulnerabilities</strong>: Node.js 20.20.0 (LTS) and 25.3.0 (Current) were released with security updates addressing multiple CVEs, focusing on permission handling, disabling futimes when the permission model is enabled, and tightening symlink API requirements. <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>WordPress 7.0 to drop PHP 7.2/7.3 support</strong>: The WordPress core development team is preparing for the 7.0 release, which will discontinue support for PHP 7.2 and 7.3, and is calling for volunteers to participate in the release process. <a href="https://app.daily.dev/posts/xho5u0zys" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Quantum computing sees product launches and security focus</strong>: The quantum computing industry is buzzing with new product launches like SuperQ’s ChatQLM, leadership changes at Xanadu, and significant security concerns highlighted by Penn State researchers regarding cloud-based quantum systems, alongside IBM’s release of the 120-qubit Nighthawk processor. <a href="https://app.daily.dev/feed-by-ids?id=EJu78wwgZ&amp;id=J8FTmD4cu&amp;id=JmIuGeLwy&amp;id=Lx6jVDI9P&amp;id=MkzC4PEYB&amp;id=QPb7M1IyK&amp;id=RHyOC3Vzq&amp;id=RONvWGnCv&amp;id=dtnIdUSBt&amp;id=hhNibwkwa" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>GCC 16 and JDK 27 bring security and architecture updates</strong>: GCC 16, expected in March-April 2026, will introduce an Algol 68 frontend, support for AMD Zen 6 and Armv9.6-A CPUs, and make C++20 the default, while JDK 27, due in September 2026, will feature post-quantum hybrid key exchange support for TLS 1.3. <a href="https://app.daily.dev/feed-by-ids?id=MZNI2dhHP&amp;id=fybiaWluM" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>AI verification bottleneck: more toil, less trust</strong>: Apparently, AI code generation has created a ‘verification bottleneck’ where developers spend more time reviewing unreliable AI-generated code than they save writing it. While 72% of developers use AI coding tools daily, 96% don’t fully trust the code. So, developer toil hasn’t shrunk; it’s just shifted from creation to endless babysitting, with power users reporting increased technical debt and correction work. Who needs productivity when you can have more ‘toil’? <a href="https://app.daily.dev/feed-by-ids?id=SVMT3qR0j&amp;id=1lxfdifYi" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Anthropic’s token crackdown: ‘No way’ to use our AI, unless you pay more</strong>: Anthropic, in a move that screams ‘we need more money,’ has started strictly enforcing token restrictions for Claude Code, preventing its use with third-party tools like Cursor and Open Code. Developers, who had built workflows around this ‘leniency,’ are now told to use Claude Code exclusively or pay significantly higher API prices. This, of course, is framed as a strategic effort to ‘lock developers into its entire stack,’ rather than, say, a desperate attempt to cover unsustainable infrastructure costs. The developer community is, naturally, thrilled. <a href="https://app.daily.dev/feed-by-ids?id=DEqFrm707&amp;id=jcpFkjQkx" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Agentic AI: the next big thing, or the next big failure?</strong>: Gartner, the oracle of tech predictions, says 40% of agentic AI projects will be canceled by 2027, not because the AI is bad, but because companies can’t implement ‘adequate guardrails.’ So, we’re building these autonomous systems that can apparently do everything, but we can’t manage them. It’s like giving a toddler a chainsaw and then being surprised when things go sideways. But don’t worry, ‘outcome-focused testing’ and ‘phased deployment’ will save us, probably. <a href="https://app.daily.dev/feed-by-ids?id=9X1ifJmCu&amp;id=QnCK3ryB4" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
