<h2>Must know</h2>
<ul>
<li><strong>Google releases Gemini 3.1 Pro, claims top spot in AI benchmarks</strong>: Google has launched Gemini 3.1 Pro in public preview, touting it as their most powerful agentic and coding model. It scored 77.1% on the ARC-AGI-2 benchmark, significantly outperforming previous versions and competitors like Anthropic’s Opus 4.6 and OpenAI’s GPT-5.2. The model features a 1 million token context window, multimodal inputs, and a new configurable thinking level, with pricing set at $2 per million input tokens, making it more cost-effective than some rivals. Early adopters are already reporting gains in problem-solving and code generation. <a href="https://app.daily.dev/posts/PjE6Xo1XG" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>MySQL community demands independent governance from Oracle</strong>: Over 248 database professionals, including engineers from major tech companies, have signed an open letter urging Oracle to transfer MySQL governance to an independent non-profit foundation. Concerns include declining commit activity, a shrinking contributor base, lack of AI-era features, and an opaque roadmap. The community points to PostgreSQL’s success with community-led governance as a model, highlighting growing fragmentation and the need for a more open development process. <a href="https://app.daily.dev/posts/GY7vcPtOu" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI agents reshape software development, bring new challenges</strong>: Agentic AI is rapidly changing software development, with tools like Kiro and Claude Code boosting engineer productivity. However, this shift introduces new roles like Agent Orchestration Engineer and LLMOps Specialist, and raises concerns about bias, compliance, and transparency. While individual productivity soars, organizations face new bottlenecks in coordination and review processes, leading to an ‘AI Curse’ where speed outpaces quality control. <a href="https://app.daily.dev/feed-by-ids?id=037a5YIjx&amp;id=4CjaWv3pP&amp;id=7XxAvY7Df&amp;id=Bl6akHr3q&amp;id=DqRWRtC6t&amp;id=F8eGl7g0G&amp;id=LCg5ueVO7&amp;id=SBASpAo7A&amp;id=UJqBwwcg1&amp;id=Uw1m3wdti" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>AI infrastructure demands drive massive power plant project</strong>: SoftBank’s SB Energy plans a $33 billion, 9.2 gigawatt natural gas power plant in the Ohio-Kentucky border region, potentially the largest in the U.S., to meet the surging energy demands of AI data centers. <a href="https://app.daily.dev/feed-by-ids?id=IYsqAR4Cy&amp;id=VBQEBjah6" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>OpenAI and Anthropic intensify rivalry with India expansion</strong>: OpenAI and Anthropic are aggressively expanding into India with new partnerships and offices, highlighting a fierce rivalry as both companies vie for market dominance in a critical region. <a href="https://app.daily.dev/feed-by-ids?id=JlixRkBbI&amp;id=JodTqUZtB&amp;id=6fxZc9j6p&amp;id=KUvvYNNG9" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI memory and agent management solutions emerge</strong>: New solutions like Mengram’s open-source AI memory API and Reload’s Epic platform are addressing AI agent memory limitations, offering semantic, episodic, and procedural memory types for individual agents and shared long-term memory for teams. <a href="https://app.daily.dev/feed-by-ids?id=0tYkAozTe&amp;id=MQxvV1b9u&amp;id=dtY8IcMQN&amp;id=qBKhvArRq" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI and robotics industrial adoption accelerates</strong>: Toyota is deploying Agility Robotics’ Digit humanoid robots in its RAV4 plant, and NVIDIA’s survey shows 90% of telecom companies report AI increasing revenue, indicating a strong trend of AI and robotics moving from pilot to industrial deployment. <a href="https://app.daily.dev/feed-by-ids?id=WnnenuOfb&amp;id=p1OVwpWSR" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI security tools and vulnerability management developments</strong>: New AI-powered SAST tools are emerging to improve vulnerability detection, while BARBIE offers a novel backdoor detection method for deep learning models, and Anubis provides bot protection against aggressive AI scraping. <a href="https://app.daily.dev/feed-by-ids?id=2TCivC1ph&amp;id=9x8XnEM3d&amp;id=CBTU3pEb0&amp;id=zsAWpz3GM" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI agent security, memory, and performance challenges</strong>: New tools like Azazel are emerging for kernel-level tracing of AI agent actions, while Anthropic’s Sonnet 4.6 shows a ‘split personality’ in benchmarks, excelling in agentic coding but regressing on one-shot tasks, and sandboxing is being implemented to improve agent security. <a href="https://app.daily.dev/feed-by-ids?id=cbOVi24pM&amp;id=iGzH52Vzz&amp;id=k9XThzl85" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Cisco and NVIDIA partner on secure AI factory</strong>: Cisco and NVIDIA announced the Cisco Secure AI Factory, a modular architecture integrating AI compute, networking, storage, and security, aiming to capture a projected $267 billion partner-delivered AI services market by 2030. <a href="https://app.daily.dev/feed-by-ids?id=IYsqAR4Cy&amp;id=VBQEBjah6" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>Anthropic’s Claude Pro now integrates with PowerPoint, because that’s what we all needed</strong>: Anthropic announced Claude Pro now integrates with Microsoft PowerPoint, allowing users to ‘pull context from their daily tools directly into presentations.’ Because clearly, the biggest problem with presentations was the lack of AI-generated bullet points, not, you know, actual content or engaging delivery. One source even claims Claude Pro is ‘getting bigger every week,’ which sounds suspiciously like a marketing department trying to justify its existence. <a href="https://app.daily.dev/posts/oLQ4V0bPw" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI management dashboard for crawler traffic control: solving problems we didn’t know we had</strong>: Link11 launched an AI Management Dashboard to give companies ‘dedicated visibility and control over AI crawler traffic.’ Apparently, we now need a special dashboard to see what our AI is doing, separate from all the other bot traffic. It’s designed to help with ‘compliance, governance, and economic assessments of AI access,’ because nothing says ‘cutting-edge AI’ like more dashboards and audit trails for things that used to just be ‘web traffic.’ <a href="https://app.daily.dev/posts/zxo3wKZNk" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI will replace tasks, not jobs, say CEOs whose companies use AI to replace tasks</strong>: CEOs of Read AI and Lucidya are confidently stating that AI replaces tasks, not entire human roles, citing their own lean teams as proof. Read AI’s five-person support team serves millions of users, thanks to AI. So, while AI isn’t ‘replacing’ jobs, it’s definitely making it possible to do the work of hundreds with a handful of people. Just a subtle distinction, I’m sure. <a href="https://app.daily.dev/posts/QMBkqUnaK" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
