<h2>Must know</h2>
<ul>
<li><strong>AI agent security vulnerabilities exposed</strong>: Snyk’s ‘ToxicSkills’ research uncovered critical security flaws in the AI Agent Skills ecosystem, with 13.4% of scanned skills containing severe issues and 76 confirmed malicious payloads designed for credential theft and data exfiltration. This highlights a significant supply chain risk for enterprises adopting AI agents, affecting platforms like OpenClaw and Claude Code. Operant AI launched ‘Agent Protector’ to address these risks with real-time security for autonomous AI agents, focusing on shadow agent discovery and inline protection against privilege escalation and data exfiltration. <a href="https://app.daily.dev/feed-by-ids?id=0BdgmYNfg&amp;id=0e28QTlYb&amp;id=2vNCw5dB4&amp;id=dL9ys7HAk&amp;id=the2EenMO" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Anthropic and OpenAI clash over AI advertising</strong>: Anthropic is planning Super Bowl ads to mock OpenAI’s move to introduce advertising in ChatGPT, positioning its Claude AI as ad-free. OpenAI CEO Sam Altman called Anthropic’s ads ‘dishonest,’ defending OpenAI’s ad principles. This public spat highlights differing business models and competitive tension, with Anthropic relying on enterprise contracts while OpenAI reportedly burns $9 billion annually with only 5% paying users. <a href="https://app.daily.dev/posts/TmM2mNSSP" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Microsoft appoints engineering quality czar</strong>: Microsoft CEO Satya Nadella appointed Charlie Bell as the new engineering quality czar, reporting directly to him. This move aims to address ongoing issues like Azure outages, Windows patch problems, and low Copilot adoption rates, signaling a significant push to improve product reliability and user experience across Microsoft’s offerings. <a href="https://app.daily.dev/posts/NGytU6zYE" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>AI coding agents advance with new models and frameworks</strong>: Anthropic released Claude Opus 4.6 with ‘agent teams’ for parallel task execution and expanded context windows, while OpenAI launched ‘Frontier,’ an enterprise platform for building and managing AI agents, and GPT-5.3-Codex demonstrated autonomous coding for full deployment lifecycles. <a href="https://app.daily.dev/feed-by-ids?id=G7Sre82A1&amp;id=Ijw4mCG7B&amp;id=TGZyay3jK&amp;id=RtOeT8YDy&amp;id=YPOCeTWKC&amp;id=gdolSqfBx&amp;id=kPp1hKs76&amp;id=n83HYCggC&amp;id=tWZGSplHo&amp;id=vu5lYpI1U" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AWS launches new EC2 instances</strong>: AWS introduced new EC2 High Memory U7i-6TB instances with custom Intel Xeon Scalable Processors for in-memory databases, and C8id, M8id, R8id instances powered by custom Intel Xeon 6 processors, offering significant performance and memory bandwidth improvements. <a href="https://app.daily.dev/posts/JPt4mjvME" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Laravel releases official AI SDK</strong>: Laravel launched an official AI SDK in beta, providing a unified API for integrating AI capabilities from multiple providers like Anthropic, Gemini, and OpenAI into Laravel applications, supporting features like AI agents, text generation, and vector stores. <a href="https://app.daily.dev/posts/ND9RAWZ9f" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Substack discloses user data breach</strong>: Substack revealed a data breach from October 2025 that exposed user email addresses, phone numbers, and internal metadata, affecting an unknown number of its 50+ million active subscriptions, though passwords and financial information were not compromised. <a href="https://app.daily.dev/posts/myrJIoXYs" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Valve delays hardware due to component shortages</strong>: Valve has delayed pricing and shipping announcements for its Steam Machine and Steam Frame VR headset, attributing the delays to memory and storage shortages driven by AI data center demand, which has more than doubled RAM prices since October 2025. <a href="https://app.daily.dev/posts/iXYPYa4qJ" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Amazon ECS adds NLB deployment support</strong>: Amazon ECS now supports linear and canary deployment strategies for services using Network Load Balancers (NLB), enabling controlled, incremental traffic shifting for TCP/UDP-based applications with integration for CloudWatch alarms and automatic rollback. <a href="https://app.daily.dev/posts/4MU5wceHx" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Intel ISPC 1.30 released with AMX support</strong>: Intel released ISPC (Implicit SPMD Program Compiler) version 1.30, a C variant for parallel processing, now including AMX (Advanced Matrix Extensions) support in its standard library to simplify targeting Intel CPUs and GPUs. <a href="https://app.daily.dev/posts/SaAG9DKe2" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>AI bot traffic surges, publishers panic</strong>: Apparently, AI bots now account for one web visit for every 31 human visits, and clickthrough rates from AI apps to publishers have plummeted to 0.27%. Publishers are ‘deploying increasingly aggressive defenses,’ because who needs actual human readers when you can have an ‘arms race’ with bots? It’s almost like the internet is becoming a giant, self-referential AI echo chamber. <a href="https://app.daily.dev/posts/5O2tzW5oh" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Elixir is the ‘best language for AI’ says unverified study</strong>: An article boldly claims Elixir is the ‘best language for AI,’ citing an unnamed ‘Tencent study’ that supposedly found Elixir achieved a 97.5% LLM completion rate. Because nothing says ‘scientific rigor’ like an anonymous study from a major tech company, conveniently unlinked and unverified, to push a programming language. Sure, Elixir’s great, but let’s not pretend it’s magic. <a href="https://app.daily.dev/posts/1D7tGyRwk" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Unmonitored AI agents pose ‘Titanic disaster’ risk</strong>: A survey of IT executives warns that 1.5 million AI agents are running without monitoring, leading to 88% of execs experiencing or suspecting security incidents. Security experts are comparing this to the ‘Titanic disaster,’ because apparently, ignoring known risks is a new phenomenon. Who knew that deploying ‘invisible’ AI agents via low-code tools without oversight could be a problem? Shocking. <a href="https://app.daily.dev/posts/RMzBzQnM3" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
