<h2>Must know</h2>
<ul>
<li><strong>Laravel Wayfinder brings end-to-end type safety to PHP and TypeScript</strong>: Laravel has launched Wayfinder, an open-source tool that automatically generates TypeScript types from Laravel applications. This provides end-to-end type safety between PHP backends and TypeScript frontends, covering routes, model attributes, validation rules, and enums. It can even sync types across separate repositories, making your PHP and TypeScript play nice together, as reported by Laravel News. <a href="https://app.daily.dev/posts/rcdxam3W8" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Ransomware gangs weaponize compliance violations</strong>: Ransomware groups like Anubis and RansomHub are now threatening victims, especially in healthcare, with regulatory reporting of compliance violations if ransoms aren’t paid. They’re using AI to analyze stolen data for compliance gaps and even draft legal complaints. This creates a double whammy: pay up or face regulatory fines and reputational damage, according to multiple reports. <a href="https://app.daily.dev/posts/OsQyB9bxn" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Cisco patches critical zero-day RCE exploited by China-linked group</strong>: Cisco has released emergency patches for a critical remote code execution (RCE) vulnerability (CVE-2025-20393) in AsyncOS for Secure Email Gateway and Secure Email and Web Manager. This flaw, actively exploited by the China-linked APT group UAT-9686 since late November 2025, allows arbitrary command execution as root. If you’re running these, patch now, as reported by multiple security outlets. <a href="https://app.daily.dev/posts/Faoj4Wg3E" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>GitHub Copilot gets agentic memory in Visual Studio</strong>: GitHub Copilot now learns and remembers your coding preferences and project guidelines, aiming to reduce code review friction and streamline onboarding, with memories expiring after 28 days to prevent stale knowledge. <a href="https://app.daily.dev/posts/nCmOnsQS3" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>TSMC revenue surges on AI demand, Nvidia now top client</strong>: TSMC reported a 25.5% year-over-year revenue growth in Q4 2025, forecasting 30% growth for 2026, driven by persistent AI infrastructure demand, with Nvidia reportedly surpassing Apple as their largest client. <a href="https://app.daily.dev/posts/aTc9ls9yL" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Raspberry Pi AI HAT+ 2 enables on-device generative AI</strong>: Raspberry Pi released the AI HAT+ 2, an add-on board with a Hailo-10H neural network accelerator and 8GB RAM, enabling local execution of smaller LLMs and computer vision models without cloud dependency. <a href="https://app.daily.dev/posts/nbjHwDlw3" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Wiz uncovers critical AWS CodeBuild supply chain flaw</strong>: Security researchers at Wiz found a critical vulnerability in AWS CodeBuild, dubbed “CodeBreach,” that could have led to global supply chain attacks by bypassing webhook filters, though AWS patched it within 48 hours with no evidence of in-the-wild exploitation. <a href="https://app.daily.dev/feed-by-ids?id=q7gLeN4EH&amp;id=MZnlHztN0&amp;id=PTMQTWkW2" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Wikimedia partners with major AI companies for data access</strong>: The Wikimedia Foundation has launched commercial partnerships with Amazon, Meta, Microsoft, Mistral AI, and Perplexity via Wikimedia Enterprise, monetizing structured access to Wikipedia data for training large language models. <a href="https://app.daily.dev/posts/QmEAqVcYI" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Palo Alto Networks patches firewall DoS flaw</strong>: Palo Alto Networks issued an emergency patch for a high-severity denial-of-service vulnerability (CVE-2026-0227) in PAN-OS firewalls with GlobalProtect, allowing unauthenticated remote attackers to force devices into maintenance mode. <a href="https://app.daily.dev/posts/MtreEtTLc" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Eurail BV discloses customer database breach</strong>: Eurail BV announced a cybersecurity incident exposing personal information, including names, contact details, and passport data, with DiscoverEU participants potentially having additional sensitive data compromised, though no misuse has been reported yet. <a href="https://app.daily.dev/posts/SkuYRBlp7" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>Oracle’s debt drama: because transparency is for suckers</strong>: Oracle is facing a class-action lawsuit from bondholders who claim the company conveniently forgot to mention an extra $38 billion in debt for OpenAI datacenter investments. Apparently, bond offering documents were just ‘misleading’ by omitting this minor detail. Now, Oracle’s bonds are tanking, and investors are left holding the bag. Who needs full disclosure when you’re in a ‘multi-trillion-dollar arms race’? <a href="https://app.daily.dev/posts/9fgRHJtn9" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>GenAI tools: the corporate data leak you didn’t ask for</strong>: Harmonic Security found that 2.6% of 22.4 million prompts to generative AI tools contained sensitive corporate data, with ChatGPT being the biggest culprit at 71%. They’re warning that using personal accounts and training public models with this data is a ticking time bomb. So, while everyone’s hyping AI for productivity, it seems it’s also a fantastic new way to accidentally broadcast your company’s secrets to the world. <a href="https://app.daily.dev/feed-by-ids?id=F9SzKFotn&amp;id=mZZUloj80" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Microsoft, Sony, Nintendo: still committed to ‘safety’ (in principle)</strong>: Microsoft, Sony, and Nintendo have released a joint update on their ‘player safety’ initiative, first announced in 2020. The update outlines principles like ‘prevention,’ ‘partnership,’ and ‘responsibility,’ focusing on empowering players and working with regulators. What it doesn’t include, however, are any actual new policies, features, or technical updates. It’s a masterclass in saying a lot without saying anything at all, proving that PR statements are still alive and well in 2026. <a href="https://app.daily.dev/posts/CTtMMpBVE" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
