<h2>Must know</h2>
<ul>
<li><strong>OpenAI goes all-in on ads and subscriptions</strong>: OpenAI has launched ChatGPT Go, an $8/month subscription tier with ads, globally, following its India debut. This aggressive monetization strategy, which aims for $1 billion in ad revenue in 2026, is a direct response to Google Gemini’s growing market share, as reported by ‘The A in AGI stands for Ads’ and ‘wtf is going on with ChatGPT?’. Critics are raising privacy concerns, noting the irony of AGI development being funded by user data monetization. <a href="https://app.daily.dev/feed-by-ids?id=ArnlILqap&amp;id=ep7p1hnQy" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>China’s massive satellite internet ambitions and Starlink’s vulnerabilities</strong>: China has filed plans with the ITU for two mega-constellations totaling nearly 200,000 satellites, dwarfing Starlink’s plans and signaling an intent to dominate orbital spectrum. This move has prompted the U.S. FCC to accelerate approvals for domestic networks due to national security concerns. Meanwhile, Iran successfully jammed Starlink internet during a national blackout, degrading performance by up to 80% and raising questions about the system’s reliability in conflict zones, according to reports. <a href="https://app.daily.dev/feed-by-ids?id=9Dqo8zul1&amp;id=VVK9EFJ8w" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI accelerates software development but raises authenticity and accountability concerns</strong>: AI is speeding up software development, but human review remains critical for correctness and security. The ‘Dead Internet Theory’ suggests much online content, including code, is AI-generated, leading to authenticity doubts. A project called OpenSlopware, which documented AI-contaminated open-source code, was harassed offline but continues under new maintainers, highlighting community concerns about code quality, copyright, and the environmental impact of AI-generated content. <a href="https://app.daily.dev/feed-by-ids?id=JjrjV7tSv&amp;id=ZeB3UcRxG&amp;id=alpyX7cTa&amp;id=m1Q81Q3xQ" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>AI agents for SRE: multi-agent incident response</strong>: Multi-agent AI systems are emerging to assist SRE teams in incident response by automating investigation tasks, with centralized architectures showing promise, though production readiness and human oversight remain crucial. <a href="https://app.daily.dev/posts/T6l4Ggvwh" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Wine 11.0 stable release brings major architectural improvements</strong>: Wine 11.0 stable is out with 6,300 code changes, including a new NTSYNC kernel module, a complete WoW64 architecture overhaul, and EGL as the default OpenGL backend, significantly improving Windows application compatibility on Linux. <a href="https://app.daily.dev/posts/5LDn1X8et" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Wine enables direct Adobe Photoshop installation on Linux</strong>: New Wine patches allow direct installation of Adobe Photoshop 2021 and 2025 on Linux by fixing XML parsing issues related to legacy Internet Explorer dependencies, though these patches are not yet in the official mainline. <a href="https://app.daily.dev/feed-by-ids?id=WW9lhhDTb&amp;id=jcudSc8Dd" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Curl drops OpenSSL-QUIC backend</strong>: The curl project is removing support for the OpenSSL-QUIC backend in version 8.19.0 due to its inadequate API, poor performance, and excessive memory usage, consolidating HTTP/3 support to nghttp2+nghttp3 and quiche. <a href="https://app.daily.dev/posts/As2f3e0AW" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Jakarta EE 12 development and community events update</strong>: The Jakarta EE Platform project is progressing towards Milestone 2 for Jakarta EE 12, targeting January 27, 2026, with numerous community conferences scheduled to foster engagement in the enterprise Java ecosystem. <a href="https://app.daily.dev/posts/YrhkKGKoN" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Memory-safe languages and AI-driven security for 2026</strong>: With 48,000 CVEs in 2025 and AI accelerating exploitation, prioritizing memory-safe languages like Rust (0.2 vulnerabilities per million lines vs. C/C++'s 1,000) and AI-driven monitoring will be crucial for security in 2026. <a href="https://app.daily.dev/posts/Tk2pS5qPG" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Vertex AI pricing for 16 services in 2026</strong>: Vertex AI pricing varies significantly across 16 core services, with Gemini models ranging from $0.10 to $120 per million tokens and GPU training instances from $0.64 to $105 per hour, emphasizing the need for cost optimization strategies. <a href="https://app.daily.dev/posts/AslvqVnql" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>AI coding assistants: the illusion of productivity</strong>: Apparently, AI-powered code generation can deliver outputs in minutes, creating pressure for increased throughput without matching human comprehension. Yet, research suggests these coding assistants may actually <em>slow</em> developers down, despite perceived gains. So, while your AI might churn out code faster than you can blink, you’ll still be stuck debugging its statistical hallucinations, as noted by various community efforts tracking AI-generated code. <a href="https://app.daily.dev/feed-by-ids?id=JjrjV7tSv&amp;id=ZeB3UcRxG&amp;id=alpyX7cTa&amp;id=m1Q81Q3xQ" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI’s persistent memory management bug: a feature, not a bug?</strong>: It seems AI code generation tools consistently replicate a memory management bug involving buffer allocation, returning allocated buffer size instead of actual data. This isn’t a one-off; it’s a systemic limitation where LLMs statistically mirror flawed patterns from their training data. So, don’t expect AI to magically fix our coding woes when it’s busy perfecting the art of repeating our mistakes, as highlighted by articles discussing algorithmic influence and AI code generation flaws. <a href="https://app.daily.dev/feed-by-ids?id=mY0RAoEVz&amp;id=x9XQt7sEo" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>The joy of coding: now automated away by AI</strong>: A developer, after 20 years of Vim/Neovim, is now questioning if learning fundamental tools like awk and sed still makes sense because AI tools are ‘good enough.’ It’s a touching reflection on the tension between efficiency gains and the ‘loss of joy’ from the craft of coding itself. Because, you know, nothing says ‘progress’ like replacing the satisfaction of mastery with the bland efficiency of automation, as lamented on Hacker News. <a href="https://app.daily.dev/posts/vFzyyNK3z" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
