<h2>Must know</h2>
<ul>
<li><strong>Cisco patches critical zero-day RCE exploited by China-linked group</strong>: Cisco has released emergency patches for a critical remote command execution (RCE) vulnerability, CVE-2025-20393, affecting 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 the Spam Quarantine feature is internet-exposed. Cisco’s fix also removes persistence mechanisms installed by attackers, as reported by multiple security outlets. <a href="https://app.daily.dev/posts/Faoj4Wg3E" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AWS CodeBuild webhook regex flaw exposed global cloud supply chain risk</strong>: Security researchers at Wiz discovered a critical vulnerability, dubbed “CodeBreach,” in AWS CodeBuild due to improperly anchored regex patterns in webhook filters. This flaw could have led to unauthorized access, admin-level repository control, and global supply chain attacks on AWS environments, potentially affecting 66% of global cloud environments. AWS was notified in August 2025 and patched the vulnerability within 48 hours, implementing global hardening measures, according to Wiz and other tech outlets. <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>Ransomware gangs weaponize compliance violations for extortion</strong>: Ransomware groups like Anubis and RansomHub are now threatening victims, particularly in highly regulated sectors like healthcare, with regulatory reporting of compliance violations. This new tactic uses AI-powered tools to analyze stolen data for compliance gaps and generate formal legal complaints, creating a dual threat of ransom demands and regulatory punishment. This strategy is becoming standard, increasing reputational and financial risks for victim organizations. <a href="https://app.daily.dev/posts/OsQyB9bxn" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>TSMC revenue surges on AI demand, Nvidia becomes top client</strong>: TSMC reported a 25.5% year-over-year revenue growth in Q4 2025, forecasting 30% growth for FY 2026, primarily driven by persistent AI infrastructure demand, with Nvidia reportedly surpassing Apple as TSMC’s largest client. <a href="https://app.daily.dev/posts/aTc9ls9yL" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>IBM to launch sovereign core platform on Red Hat OpenShift</strong>: IBM is releasing a technical preview of IBM Sovereign Core, a platform built on Red Hat OpenShift, designed to rapidly deploy isolated workloads and address data sovereignty concerns by keeping authentication, encryption keys, and access management under internal IT control. <a href="https://app.daily.dev/posts/eWD0cjmPm" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Tailscale’s Terraform module simplifies multi-cloud deployments</strong>: Tailscale has released an open-source Terraform module that streamlines VM provisioning across various cloud providers by automating Tailscale installation via cloud-init, addressing common bootstrap issues and supporting multi-cloud, ephemeral machines, and route configurations. <a href="https://app.daily.dev/posts/sCY8mByyk" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>RavenDB introduces Kubernetes operator for database orchestration</strong>: RavenDB has launched a Kubernetes Operator to replace Helm charts, automating cluster formation, storage binding, and networking, and handling Day 2 operations like rolling upgrades and certificate rotation with declarative custom resource definitions. <a href="https://app.daily.dev/posts/BIJJTYGeQ" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>GitHub Copilot introduces agentic memory in Visual Studio</strong>: GitHub Copilot now includes a “memories” feature that learns and documents coding preferences and project guidelines, aiming to reduce code review friction and streamline onboarding by sharing learned patterns across development workflows. <a href="https://app.daily.dev/posts/nCmOnsQS3" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Wikimedia Foundation partners with major AI companies</strong>: The Wikimedia Foundation has launched commercial partnerships with Amazon, Meta, Microsoft, Mistral AI, and Perplexity through Wikimedia Enterprise, monetizing AI companies’ reliance on Wikipedia data for training large language models and creating a new revenue stream. <a href="https://app.daily.dev/posts/QmEAqVcYI" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Palo Alto Networks issues emergency patch for firewall DoS flaw</strong>: Palo Alto Networks released emergency patches for CVE-2026-0227, a denial-of-service vulnerability in PAN-OS firewalls with GlobalProtect enabled, which allows unauthenticated remote attackers to force firewalls into maintenance mode, disrupting network availability. <a href="https://app.daily.dev/posts/MtreEtTLc" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>OpenAI invests in Altman’s brain-computer interface startup</strong>: OpenAI, facing projected operating losses of $74 billion by 2028 and $1.4 trillion in infrastructure commitments, somehow found $250 million to invest in Merge Labs, a non-invasive BCI startup co-founded by its CEO, Sam Altman. While the tech is decades away, the “synergy” of OpenAI’s investment potentially increasing the value of Altman’s other company is just a delightful coincidence, isn’t it? Nothing to see here, just good old-fashioned ethical governance at play. <a href="https://app.daily.dev/posts/rYkFwNLz3" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Raspberry Pi AI HAT+ 2: massive upgrade or just more marketing fluff?</strong>: Raspberry Pi released the AI HAT+ 2, boasting a Hailo-10H neural network accelerator and 8GB RAM for on-device generative AI. They claim 40 TOPS for LLMs and 26 TOPS for computer vision, which is a bit confusing, but hey, it runs small LLMs locally! Despite some sources noting unchanged computer vision performance from the previous version and a higher price, it’s still being touted as a “massive upgrade.” Because nothing says innovation like conflicting performance figures and a price hike for marginal gains. <a href="https://app.daily.dev/posts/nbjHwDlw3" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Enterprise AI integration: just add more certifications and frameworks!</strong>: IBM’s CIO is pushing an enterprise AI transformation strategy, complete with “AI fusion teams,” “AI license to drive” certifications, and a “Builder’s Path framework.” Because what every engineer needs is more certifications and frameworks to navigate, rather than just, you know, actually building things. Apparently, the key to success is changing organizational culture to reward “AI-enabled productivity over traditional hard work.” So, if you’re still doing “hard work,” you’re doing it wrong, peasant. <a href="https://app.daily.dev/feed-by-ids?id=XM781gSqJ&amp;id=pjUEOTNDx" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
