<h2>Must know</h2>
<ul>
<li><strong>European Parliament bans cloud AI over data security fears</strong>: The European Parliament has banned cloud-based AI features on lawmakers’ devices due to data security and privacy concerns, highlighting a growing institutional distrust of AI tools, especially those relying on US-based cloud services. This move, reported by sources like Article #1PAt3KUHq, signals a pragmatic approach to AI adoption, prioritizing security over convenience, despite the EU’s own comprehensive AI Act. <a href="https://app.daily.dev/posts/NW6nIRqB9" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Code signing policies get stricter</strong>: Major tech companies are hardening code signing policies, enforcing a new 460-day maximum validity period for certificates to reduce supply chain attack windows. This change, driven by the CA/Browser Forum, requires organizations to transition from file-based keys to FIPS-compliant hardware and automate signing workflows to avoid disruptions, as reported by Latest Hacking News. <a href="https://app.daily.dev/posts/VDBUnXRHh" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Cloud outages are the new normal</strong>: Cloud outages are becoming more frequent due to staffing challenges, enterprise complacency, and the increasing complexity of hyperscale platforms. InfoWorld highlights the recent 10-hour Azure outage, caused by a storage policy misconfiguration, as an example of how human error in complex systems can lead to widespread disruption, emphasizing the need for better operational excellence and architectural redundancy. <a href="https://app.daily.dev/posts/V2F1hpMgh" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>Meta deploys Nvidia grace cpus for AI</strong>: Meta is deploying Nvidia’s Grace CPUs as standalone processors at scale for general-purpose and agentic AI workloads, marking a shift from the typical Grace-Hopper Superchip configuration. <a href="https://app.daily.dev/posts/tohNH8UoB" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Alibaba launches Qwen3.5 multimodal AI</strong>: Alibaba released Qwen3.5, a new multimodal AI model designed for enterprise agent workflows, claiming benchmark improvements over GPT-5.2, Claude 4.5 Opus, and Gemini 3 Pro. <a href="https://app.daily.dev/posts/UOocLbfK4" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Temporal raises $300m for AI workflows</strong>: Temporal secured $300M Series D funding at a $5B valuation, driven by demand for Durable Execution in AI and long-running workflows, with companies like OpenAI and ADP using their platform. <a href="https://app.daily.dev/posts/HqHJFpyjl" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Linux Foundation moves to real-time analytics with Tinybird</strong>: The Linux Foundation replaced a 12-hour Snowflake batch sync with Tinybird for real-time analytics on their LFX Insights platform, tracking over 6,000 open-source projects. <a href="https://app.daily.dev/posts/LqfC5JQvs" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Laravel migration cuts infrastructure costs by half</strong>: PyleSoft migrated six Laravel applications from Laravel Vapor to Laravel Cloud, reducing infrastructure costs by 50% while handling 1.5 million daily requests. <a href="https://app.daily.dev/posts/cy0axOtZo" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>OWASP highlights smart contract governance risks</strong>: OWASP released its Smart Contract Top 10 2026 risk framework, identifying governance failures and access control issues as dominant security risks in blockchain systems. <a href="https://app.daily.dev/posts/U7Z7c8wKd" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AWS launches new high-performance computing instances</strong>: AWS launched EC2 Hpc8a instances, powered by 5th Gen AMD EPYC processors, delivering up to 40% higher performance for tightly coupled HPC workloads. <a href="https://app.daily.dev/posts/mfaqQ5NVR" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>AI accidentally exposes DJI robot vacuum vulnerability</strong>: Apparently, a hobbyist, with the help of Claude AI, “accidentally” found a vulnerability in DJI’s MQTT broker, exposing 7,000 robot vacuums. So, AI is now so good it can stumble upon critical IoT security flaws, making manufacturers look even worse for shipping products with basic security failures. Maybe they should ask an AI to secure their stuff. <a href="https://app.daily.dev/posts/sfYExhzMO" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Toilet maker Toto is now an AI play, apparently</strong>: Activist investor Palliser Capital is trying to convince everyone that Japanese toilet manufacturer Toto is an “AI play” because its advanced ceramics division makes parts for memory chips. Yes, you read that right. A toilet company is now an AI stock. This is peak financial engineering, turning porcelain into AI gold. <a href="https://app.daily.dev/posts/08TkDSLKo" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Quest software launches AI-powered data platform with unverified claims</strong>: Quest Software launched an “AI-powered” data management platform that supposedly creates data products from natural language prompts and offers up to 54% faster delivery. Because nothing says “trustworthy data” like letting an AI whip it up from a vague prompt, and then claiming massive savings without any independent verification. What could possibly go wrong? <a href="https://app.daily.dev/posts/wlE5eAAsI" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
