AICodeKing

daniel.haxx.se

finout

Kotlin Weekly

Adam Argyle

Philipp Lackner

<h2>Must know</h2>
<ul>
<li><strong>China’s massive satellite constellation plans raise concerns</strong>: China has filed requests with the ITU for two new satellite constellations, CTC-1 and CTC-2, totaling an astounding 193,428 satellites—four times larger than SpaceX’s Starlink plans. This move, spearheaded by a newly established entity, has the US FCC concerned about national security implications, accelerating approvals for US satellite networks. While China’s current deployment is limited, these filings could be a strategic signal, though meeting ITU deployment benchmarks will be a significant challenge, as reported by various sources. <a href="https://app.daily.dev/posts/9Dqo8zul1" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Snap Store publisher domains hijacked for malware distribution</strong>: Scammers are exploiting the Canonical Snap Store by registering expired domains previously used by legitimate publishers, then using these hijacked accounts to distribute malicious updates to trusted snap applications. This new tactic, which specifically targets cryptocurrency wallet users, undermines trust in the Snap Store’s publisher reputation signals. Publishers are advised to maintain domain control and enable 2FA, while users should be cautious with crypto apps from Snap, according to cybersecurity reports. <a href="https://app.daily.dev/posts/8RJLmL5a0" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI solves a previously unsolved math problem</strong>: GPT-5.2 Pro has reportedly solved Erdos problem #281, a mathematical challenge that had stumped human mathematicians. Terence Tao, a highly respected mathematician, called this “perhaps the most unambiguous instance” of AI solving an open problem, marking a significant milestone for AI in mathematical research. This achievement, reported on January 18, 2026, highlights the growing capabilities of AI in complex problem-solving. <a href="https://app.daily.dev/posts/pRkd3CPEF" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>Open source models: cathedral, megachurch, and bazaar</strong>: Open source projects are categorized into three models: exclusive ‘cathedrals,’ large ‘megachurches’ like the Linux Foundation, and ‘bazaars’ (mostly single-maintainer projects) which form critical infrastructure but struggle with sustainable funding, a gap the Sovereign Tech Agency is trying to address. <a href="https://app.daily.dev/posts/rr8Vo8FaG" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>New CLI tool integrates AI and developers with Figma</strong>: A new command-line tool, dannote/figma-use, offers full read/write access to Figma, enabling AI agents and developers to programmatically create and modify designs using JSX syntax, with an experimental ‘render’ command promising up to 100x faster operations. <a href="https://app.daily.dev/posts/DfAoRLrdU" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Physical AI and AV industry shifts at CES 2026</strong>: CES 2026 highlighted Physical AI and autonomous vehicles, with Chinese companies dominating and Mobileye acquiring Mentee Robotics for $900 million, while Motional shifted to an ‘AI-first’ AV development and Tesla moved to subscription-only FSD pricing. <a href="https://app.daily.dev/posts/MtQzzuuIN" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Iran disrupts Starlink internet with GPS jamming</strong>: Iran successfully used military-grade GPS jammers to disrupt Starlink internet, cutting performance by up to 80% during a nationwide internet shutdown, raising concerns about Starlink’s reliability in conflict zones. <a href="https://app.daily.dev/posts/VVK9EFJ8w" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Critical RCE vulnerability in n8n automation platform</strong>: A critical unauthenticated remote code execution vulnerability (CVE-2026-21858) has been discovered in the n8n automation platform, allowing attackers to read arbitrary files and execute commands by exploiting a Content-Type confusion bug in the Form Webhook node. <a href="https://app.daily.dev/posts/4XP5JDLv1" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Google’s Antigravity expands to compete with Anthropic’s Co-work</strong>: Google’s Antigravity is now adding productivity features beyond coding, leveraging Gemini 3 Flash and an agent manager to create presentations and organize files directly from the IDE, aiming to compete with Anthropic’s Co-work for developer-centric all-in-one solutions. <a href="https://app.daily.dev/posts/HVlI4ECqg" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Memory-safe languages reduce vulnerabilities</strong>: With 48,000 CVEs published in 2025 and attackers exploiting vulnerabilities within 24 hours, memory-safe languages like Rust show significantly lower vulnerability density (0.2 per million lines) compared to C/C++ (1,000 per million lines), highlighting a critical shift for security in 2026. <a href="https://app.daily.dev/posts/Tk2pS5qPG" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>AI-generated code: still a mess for testability</strong>: Apparently, AI-generated SwiftUI code is still embedding business logic directly into views, making it utterly untestable. Rebeloper points out that this means developers are still stuck refactoring to separate concerns, which is exactly what we’ve been doing for decades. So much for AI making our lives easier; it just generates new ways to create technical debt. <a href="https://app.daily.dev/posts/jCgZBlLZU" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI coding enforcement: because AI needs a babysitter</strong>: Someone actually built a CLI tool called ‘govctl’ to enforce RFC-driven development for AI-assisted coding. Because, you know, AI is so brilliant it needs a strict workflow to prevent ‘undisciplined AI code generation.’ It’s almost like the AI isn’t actually intelligent, and we’re just building more layers of bureaucracy around glorified autocomplete, as if that will magically make it produce good code. <a href="https://app.daily.dev/posts/oBJvqLGNT" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI pricing: because tokens are the new gold</strong>: Finout is out here detailing the intricate pricing models for Vertex AI and Claude in 2026, with Gemini and Claude models priced per million tokens, node-hours, and even web searches. It’s a masterclass in how to nickel-and-dime every single interaction with an AI, proving that while the tech might be advanced, the monetization strategies are as old as time: make it complex, charge for everything, and tell people to ‘optimize’ their usage. <a href="https://app.daily.dev/feed-by-ids?id=AslvqVnql&amp;id=77ufsy3OX" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
