<h2>Must know</h2>
<ul>
<li><strong>Anthropic and Google duke it out over AI coding tools</strong>: Anthropic’s new Cowork tool, exclusive to Claude Max subscribers on macOS, allows non-coders to use agentic AI for file management and data analysis. This comes as Anthropic implemented OAuth restrictions, blocking third-party tools like OpenCode from accessing Claude Pro/Max subscriptions, forcing developers to use their costly pay-per-use API. OpenAI, seizing the opportunity, immediately announced support for ChatGPT subscriptions within OpenCode and other open-source tools, intensifying the rivalry for developer adoption, as reported by TechCrunch. <a href="https://app.daily.dev/feed-by-ids?id=TLkxIySjV&amp;id=8dtXe7P09&amp;id=eRi1VQ5SP" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI data centers are energy hogs, but nuclear might save us</strong>: Hyperscale AI data centers, consuming over a gigawatt each, are driving massive electricity demand, with over 50% currently powered by fossil fuels. This leads to rising energy costs, water shortages, and pollution in hosting communities. However, next-generation nuclear reactors, including small modular reactors (SMRs) and advanced fuels, are emerging as a potential solution to meet this demand safely and efficiently, according to MIT Technology Review. <a href="https://app.daily.dev/feed-by-ids?id=bf0xB32y7&amp;id=qk040xtMi&amp;id=dJDy4tUxK" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Supply chain attacks target automation platforms, data leaks expose job seekers</strong>: Malicious npm packages impersonated n8n workflow automation integrations to steal enterprise OAuth tokens and API keys, marking the first known supply chain attack on the n8n ecosystem. Separately, Interview Coder, an AI-based interview platform, publicly exposed users’ full names, compensation, and offer letters due to improper redaction. These incidents highlight critical vulnerabilities in third-party software trust and lax security practices, as detailed by Endor Labs. <a href="https://app.daily.dev/feed-by-ids?id=E0HoVEZsQ&amp;id=UnvBaUcPK" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>Rust-based malware targets Middle East in espionage campaigns</strong>: The Iranian APT group MuddyWater is deploying new Rust-based malware, ‘RustyWater,’ in spear-phishing campaigns targeting diplomatic, maritime, financial, and telecom sectors in the Middle East, particularly Israel. <a href="https://app.daily.dev/posts/xFuCzijpN" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Linux desktop ecosystem diverges with X11 and Wayland shifts</strong>: IceWM 4.0 continues active development on X11, while Budgie 10.10 fully transitions to Wayland-native operation, and Budgie 11 plans a rewrite using Qt6 and KDE Frameworks, showcasing fragmentation in the Linux GUI space. <a href="https://app.daily.dev/posts/peCpqhbhc" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Google and major retailers launch universal commerce protocol</strong>: Google, in partnership with Shopify, Etsy, Target, Walmart, and Wayfair, announced the Universal Commerce Protocol (UCP), an open standard to enable AI agents to discover products and facilitate purchases across various online retail platforms. <a href="https://app.daily.dev/feed-by-ids?id=xOjrayt7r&amp;id=2S31ZjKRT&amp;id=PLHkMQb0u&amp;id=Ywq5VNAy0" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Neuron AI Laravel SDK brings agentic AI to PHP</strong>: The new Neuron AI Laravel SDK allows PHP developers to create autonomous AI agents within the Laravel framework, supporting multiple AI providers like Anthropic, OpenAI, and Gemini without needing Python. <a href="https://app.daily.dev/feed-by-ids?id=4bt4Q0nem&amp;id=QVuznBqAB&amp;id=pfLq9n48m" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Auto-CPUFreq 3.0 enhances Linux battery and CPU control</strong>: Auto-CPUFreq 3.0 for Linux introduces CPU turbo mode override via CLI and GUI, improved battery device configuration, and specific support for ASUS laptop charge thresholds, enhancing power management. <a href="https://app.daily.dev/posts/umWMqgJxW" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Postman acquires Fern for API documentation and SDK automation</strong>: Postman acquired Fern, a startup specializing in automatically generating API documentation and client SDKs in nine programming languages, aiming to reduce developer friction and expand Postman’s API lifecycle offerings. <a href="https://app.daily.dev/posts/9UqlZV9z7" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Laravel Simplifier plugin released for Claude Code</strong>: Taylor Otwell released a Laravel Simplifier plugin for Claude Code users, adapted from Anthropic’s original, to review code commits and recommend improvements aligning with Laravel standards and PSR-12. <a href="https://app.daily.dev/posts/Z70ay5WSd" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>Google’s AI Overviews: still serving up dangerous health advice</strong>: Google, in its infinite wisdom, removed <em>some</em> faulty AI Overviews health summaries after The Guardian exposed dangerous and inaccurate medical information, specifically regarding liver test results. Apparently, telling people their health is fine when it isn’t is just a minor hiccup. Google claims most AI Overviews are accurate and clinician-reviewed, but only fixed the ones they got caught on. Nothing to see here, folks, just move along. <a href="https://app.daily.dev/posts/fofUhg4zp" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>BreachForums leak: cybercriminals shocked, shocked to find gambling going on</strong>: BreachForums, a marketplace for stolen data, got breached, exposing 324,000 user identities. The hacker, ‘James,’ claims disillusionment with the forum’s members, particularly the ShinyHunters group. Cybersecurity firm Resecurity confirms the data’s authenticity, suggesting it’s enough for law enforcement to identify users who thought they were anonymous. Who knew operating a criminal enterprise might have security risks? Shocking. <a href="https://app.daily.dev/posts/iWBcNtvCe" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI will ‘reshape’ software engineering, but don’t worry, you’ll still have a job… supervising</strong>: Industry leaders are predicting AI will ‘fundamentally reshape’ software engineering roles, shifting us from actually writing code to ‘supervising’ AI systems. So, instead of building, we’ll be babysitting. They’re even hiring junior developers again because AI tools ‘amplify their productivity.’ Sounds like a great way to get more work done with fewer experienced people, while the ‘engineering leadership’ positions become even more uncertain. Progress! <a href="https://app.daily.dev/feed-by-ids?id=MU6fGtVLh&amp;id=hHTJdKyX8" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
