<h2>Must know</h2>
<ul>
<li><strong>AI agent security vulnerabilities are a growing concern</strong>: Snyk’s research, ToxicSkills, found that 13.4% of AI agent skills contain critical security issues, with 76 confirmed malicious payloads designed for credential theft and data exfiltration. This highlights a significant supply chain risk for enterprises using third-party LLMs and agents. Operant AI launched Agent Protector to provide real-time security for autonomous AI agents, offering inline protection against privilege escalation and data exfiltration, emphasizing that platform providers should be responsible for agent security, not customers. <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 escalate AI chatbot ad war</strong>: Anthropic plans Super Bowl commercials mocking OpenAI’s decision to introduce advertising in ChatGPT, positioning its Claude AI as ad-free. OpenAI CEO Sam Altman responded on X, calling Anthropic’s ads dishonest and defending OpenAI’s advertising 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% of users paying. <a href="https://app.daily.dev/posts/TmM6mNSSP" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Microsoft appoints engineering quality czar amid ongoing issues</strong>: Microsoft CEO Satya Nadella appointed Charlie Bell as engineering quality czar, reporting directly to him, to address persistent problems like Azure outages, Windows patch issues, and low Copilot adoption. This move signals a serious effort to improve engineering quality across all Microsoft products, with Hayete Gallot returning to lead security. <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>OpenAI launches enterprise AI agent platform ‘Frontier’</strong>: OpenAI introduced ‘Frontier,’ an end-to-end platform for enterprises to build, deploy, and manage AI agents, allowing them to connect to external data and applications with controlled access and permissions. <a href="https://app.daily.dev/posts/TGZyay3jK" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Anthropic releases Claude Opus 4.6 with agent teams and expanded context</strong>: Anthropic launched Opus 4.6, featuring ‘agent teams’ for parallel task execution, a 1 million token context window, and new PowerPoint integration, aiming to broaden its appeal beyond developers to knowledge workers. <a href="https://app.daily.dev/feed-by-ids?id=Ijw4mCG7B&amp;id=G7Sre82A1" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Laravel launches AI SDK for application development</strong>: Laravel released 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, structured outputs, and streaming. <a href="https://app.daily.dev/posts/ND9RAWZ9f" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AWS launches new EC2 instances with Intel Xeon 6 processors</strong>: AWS introduced new EC2 C8id, M8id, and R8id instances powered by custom Intel Xeon 6 processors, offering up to 43% higher performance and 3.3x more memory bandwidth than previous generations, targeting compute, balanced, and memory-intensive workloads. <a href="https://app.daily.dev/posts/JPt4mjvME" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Substack discloses user data breach from October 2025</strong>: Substack revealed a data breach from October 2025 that exposed user email addresses, phone numbers, and internal metadata, remaining undetected for four to five months, 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>ScyllaDB integrates USearch for vector search capabilities</strong>: ScyllaDB now includes vector search capabilities using the open-source USearch library, allowing storage of vector embeddings alongside structured data and leveraging ScyllaDB’s shard-per-core architecture for optimized performance in real-time AI applications. <a href="https://app.daily.dev/posts/pKNE7iWlw" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Amazon ECS adds NLB deployment support for linear and canary strategies</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 and integrating with CloudWatch alarms for automatic rollbacks. <a href="https://app.daily.dev/posts/4MU5wceHx" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>AI bot traffic surges, threatening publishers and reshaping internet use</strong>: Apparently, AI bots are now one of the internet’s biggest users, with one bot visit for every 31 human visits by Q4 2025, mostly from RAG bots. Publishers are up in arms because clickthrough rates from AI applications plummeted to 0.27%, threatening their ‘financial viability.’ It’s almost like building a business model entirely on ad impressions from human eyeballs might be a tad fragile when the internet starts talking to itself. <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’ according to an 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 screams ‘unbiased, rigorous research’ like an anonymous study from a major tech company that just happens to perfectly align with your promotional narrative. Sure, Elixir’s great, but let’s not pretend a magic Tencent study makes it the undisputed AI champion. <a href="https://app.daily.dev/posts/1D7tGyRwk" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Unmonitored AI agents pose significant security risk</strong>: A survey of IT executives warns that 1.5 million AI agents are running around without monitoring, leading to security incidents. Security experts are comparing this to the ‘Titanic disaster,’ because apparently, every tech problem needs a dramatic, over-the-top analogy. It’s almost as if deploying AI agents without proper oversight is a bad idea, who would’ve thought? <a href="https://app.daily.dev/posts/RMzBzQnM3" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
