<h2>Must know</h2>
<ul>
<li><strong>AI-driven cyber threats escalate with new attack vectors</strong>: The cybersecurity landscape is getting wilder, with economic pressures increasing insider risks and autonomous AI agents creating new vulnerabilities like goal hijacking and prompt injection. According to the World Economic Forum’s 2026 report, cyber fraud, not ransomware, is now the top threat, with 87% of companies seeing AI vulnerabilities as the fastest-growing risk. Meanwhile, state-sponsored groups like UAT-8837 are exploiting zero-days in critical infrastructure, and botnets like RondoDox are actively targeting unpatched HPE OneView systems, as reported by Check Point. It seems everyone’s getting in on the AI action, even the bad guys. <a href="https://app.daily.dev/feed-by-ids?id=31oswUgqu&amp;id=96JMSYGET&amp;id=zsNUfuuUT&amp;id=59RHDgM8n&amp;id=D4Ey8eaYF&amp;id=DfwQyGbyK&amp;id=Kx8Jb4Xns&amp;id=QSNdMLSLa&amp;id=hgS5v4ArI&amp;id=xjKL13D2C" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>OpenAI’s ad push and brain-computer interface bet</strong>: OpenAI is rolling out targeted ads for free and Go ChatGPT users in the U.S. to fund continued free access, promising ads won’t influence responses or sell user data. Simultaneously, OpenAI participated in a $252 million seed round for Merge Labs, a brain-computer interface (BCI) venture co-founded by Sam Altman. Merge Labs is developing non-invasive BCI using molecular neuron interfacing, with commercialization potentially decades away, according to reports. This dual strategy highlights OpenAI’s pursuit of diverse revenue streams and long-term, high-risk AI investments. <a href="https://app.daily.dev/feed-by-ids?id=DQh9IAXNl&amp;id=rYkFwNLz3" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Elon Musk’s lawsuit against OpenAI and Microsoft heads to jury trial</strong>: A federal judge has rejected motions to dismiss Elon Musk’s lawsuit against OpenAI and Microsoft, clearing the way for a jury trial in late April. Musk alleges OpenAI violated its nonprofit founding mission by prioritizing profits after accepting billions from Microsoft. The jury will determine if OpenAI breached its original promises and if Microsoft aided in those alleged breaches, focusing on the company’s shift to a for-profit model. <a href="https://app.daily.dev/posts/pwpD2QRjm" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>TSMC sees ‘endless’ AI chip demand, plans massive investment</strong>: TSMC reported record Q4 earnings and forecasts 30% revenue growth for 2026, driven by what its CEO calls ‘endless’ demand for AI chips, planning $52-56 billion in capital expenditures for advanced process nodes. <a href="https://app.daily.dev/posts/aTc9ls9yL" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Google previews BigQuery AI for natural language to SQL</strong>: Google is testing a ‘Comments to SQL’ feature in BigQuery, powered by Gemini, which translates natural language instructions in SQL comments into executable queries to accelerate data analysis. <a href="https://app.daily.dev/posts/5AcBUpc2i" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>GitHub Copilot introduces agentic memory for paid plans</strong>: GitHub Copilot now offers agentic memory in public preview for paid plans, allowing it to learn and retain repository-specific insights to improve code assistance, with memories expiring after 28 days. <a href="https://app.daily.dev/posts/nCmOnsQS3" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>ClickHouse acquires Langfuse, expands AI observability</strong>: ClickHouse acquired Langfuse, an open-source LLM observability platform, to enhance its Agentic Data Stack for monitoring non-deterministic AI applications, while Langfuse remains open-source and a standalone cloud service. <a href="https://app.daily.dev/feed-by-ids?id=7DDF3YQda&amp;id=kU2RvmlCw&amp;id=BqVJ6XPkS&amp;id=Ycyvr9ct2" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Cloudflare boosts resilience, acquires Astro Technology</strong>: Cloudflare launched ‘Code Orange: Fail Small’ to prevent global outages through phased configuration rollouts and acquired Astro Technology Company, an open-source framework for content-centric websites, which will remain MIT-licensed. <a href="https://app.daily.dev/feed-by-ids?id=OdNHxfcwg&amp;id=Q4Q57JyYL&amp;id=yRAOJZmiR" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Modular DS WordPress plugin critical vulnerability</strong>: A critical privilege escalation vulnerability (CVE-2026-23550) in Modular DS WordPress plugin versions 2.5.1 and below allows unauthenticated admin access, with active exploitation detected and a patch (v2.5.2) now available. <a href="https://app.daily.dev/posts/D6wKV3WK1" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Eurail customer data breach exposes sensitive info</strong>: Eurail BV reported unauthorized access to its customer database, exposing names, contact details, and passport data for general users and DiscoverEU participants, with no evidence of misuse yet. <a href="https://app.daily.dev/posts/SkuYRBlp7" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>AI coding tools: productivity booster or maintainer’s nightmare?</strong>: Linus Torvalds apparently used ‘vibe coding’ with AI for a hobby project, and now everyone’s doing it for ‘micro apps.’ But hold on, research suggests developers are 19% slower with AI tools, and open-source maintainers are drowning in low-quality, AI-generated pull requests. So, while AI might help you whip up a quick app to track your cat’s naps, don’t expect it to write your next production-ready, bug-free enterprise system. It seems the ‘AI empowerment’ narrative conveniently glosses over the maintainability and security headaches it creates. <a href="https://app.daily.dev/feed-by-ids?id=H6Mp947gD&amp;id=h6ZJabZ0D&amp;id=laMxMdAYV&amp;id=ETkKEZbZ4&amp;id=dpcmbDROy&amp;id=S37WCeeOE&amp;id=dc9HNh5yX&amp;id=EuNu5at2M&amp;id=9IenhmVcQ&amp;id=nLHgst6Y1" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Google’s AI search: ‘unavoidable’ errors or corporate accountability dodge?</strong>: Google’s AI search is busy making up criminal accusations against Norwegian politicians, causing ‘severe reputational damage.’ Google’s defense? These errors are ‘unavoidable’ given the scale of the internet and AI. Apparently, the tech giant that organizes the world’s information can’t be bothered to ensure its AI doesn’t libel people. It’s a classic move: tout AI as revolutionary, then shrug when it causes harm, effectively shifting the burden of misinformation onto users and victims. Accountability, anyone? <a href="https://app.daily.dev/feed-by-ids?id=507BvKbPR&amp;id=QiN4er8h3" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI firms and copyrighted data: ‘operational cost’ or blatant theft?</strong>: Schneier on Security is calling out AI firms for systematically scraping copyrighted material without consent, comparing it to the Aaron Swartz case. Anthropic’s $1.5 billion settlement for using copyrighted books is framed not as a penalty, but as an ‘operational cost’ for these companies. It seems the ‘democratization of knowledge’ only applies when it benefits their bottom line, while the actual creators get a raw deal. Who needs ethics when you have a multi-billion dollar valuation? <a href="https://app.daily.dev/feed-by-ids?id=ETkKEZbZ4&amp;id=dpcmbDROy" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
