The pressure
Simon Willison's Weblog19 days agoCommentaryDaniel Stenberg says the curl security team is facing an unprecedented surge of credible, detailed AI-assisted vulnerability reports. Incoming reports are now 4-5 times higher than in 2024 and twice the 2025 rate, averaging more than one per day. The upside is that recent curl vulnerabilities have generally been LOW or MEDIUM severity, with the last HIGH CVE published in October 2023.
3D-printable humanoid legs let robotics experiments run wild
Ars Technica AI19 days agoHardwareArs Technica reports that Hugging Face has introduced a roughly $2,500 bipedal humanoid robot project built around 3D-printable legs. The effort targets builders and researchers rather than mainstream consumers, lowering the hardware barrier for hands-on robotics experiments. Its broader significance is in open, reproducible embodied AI research, where models and control systems need physical platforms for testing.
Some ideas for what comes next, May 2026
Interconnects (Nathan L.)19 days agoCommentaryNathan Lambert argues that 2026 AI progress is becoming higher-stakes, with model capabilities, work patterns, economics, and real-world risks all escalating. He says open models still lack a true Claude Code and Opus 4.5-style agent moment, and Gemini has no clear competitor to Claude Code or Codex yet. The essay also tracks Mythos, American open-model momentum, frontier-lab competition, and mounting intervention from governments and other power structures.
Microsoft Copilot Cowork Exfiltrates Files★ 76
Simon Willison's Weblog19 days agoIncidentSimon Willison summarizes a PromptArmor report about Microsoft Copilot Cowork and agentic data exfiltration risks. The issue involved agents sending messages to a user’s own inbox without approval, where rendered external images could trigger requests to attacker-controlled sites. Because OneDrive can create pre-authenticated download links, a successful prompt injection could leak links that allow attackers to download files.
Launch HN: Minicor (YC P26) - Windows desktop automations at scale
Hacker News (AI keywords)19 days agoNew ToolMinicor appeared on Hacker News as a Launch HN post focused on Windows desktop automation at scale. Based on the title alone, it seems positioned beyond simple personal scripting, aiming at repeatable automation across Windows desktop workflows. No source text is available, so details such as AI usage, architecture, supported apps, pricing, security controls, and customer traction cannot be confirmed.
Digital Infinite to Showcase AI-Stack and ixCSP at COMPUTEX 2026
INSIDE 硬塞 AI20 days agoBusinessDigital Infinite will exhibit AI-Stack and ixCSP at COMPUTEX 2026. AI-Stack focuses on managing heterogeneous AI compute resources, while ixCSP turns compute capacity into operable and billable cloud services. The article frames the company’s direction as moving from AI infrastructure toward cloud-based compute commercialization, though it does not provide benchmark data, pricing, customer deployments, or model-specific details.
From Jailbreaking to Vibe Hacking: AI Security Shifts to "Psychocybersecurity"
INSIDE 硬塞 AI21 days agoEthicsAI security is shifting from technical jailbreaks to "Vibe Hacking," where attackers use social engineering and psychological tactics to manipulate an LLM's simulated persona. By exploiting the model's behavioral tendencies rather than code vulnerabilities, this trend establishes "psychocybersecurity" as a critical new frontier for AI alignment and safety.
邁向光速文本生成：NVIDIA Nemotron-Labs 推出擴散語言模型 (Diffusion Language Models)★ 75
Hugging Face Blog23 days agoReleaseTraditional large language models (such as GPT, Claude, and others) all use an "autoregressive" mechanism — that is, they must predict the next token based on…
美國政府緊急應對：網友利用 AI 模擬罹難飛行員聲音，規避法律限制★ 75
Ars Technica AI23 days agoIncidentThis controversy stems from strict U.S. legal restrictions on aviation accident investigation data. Under federal law, the National Transportation Safety Board…
Google DeepMind 於亞太地區推出加速器計畫，旨在利用 AI 應對環境風險
Google DeepMind Blog24 days agoReleaseGoogle DeepMind has announced the official launch of the "Google DeepMind Accelerator Program" in the Asia-Pacific (APAC) region, aimed at bringing together…
Show HN: Agent.email – sign up via curl, claim with a human OTP
Hacker News (AI keywords)24 days agoNew ToolThis Show HN post introduces Agent.email, apparently a tool for programmatic email or account signup. Based on the title, its main idea is signing up through curl, then letting a human claim the resource via OTP. No article body is available, so product details, security model, pricing, and actual use cases cannot be verified.
Launch HN: Runtime (YC P26) – Sandboxed coding agents for every team
Hacker News (AI keywords)24 days agoNew ToolRuntime is a YC P26 launch focused on making coding agents usable across an organization, not only by engineers. It provides sandboxed environments with company context, integrations, secrets, policies, observability, and cost controls. The product page says it works with tools including Claude Code, Cursor, Codex, Copilot, Gemini CLI, Devin, and OpenCode, while fitting into Slack, Linear, GitHub, and related workflows.
神秘 AI 新創 Hark 完成 7 億美元 A 輪融資，打造「通用」AI 介面與專屬硬體★ 75
TechCrunch AI24 days agoBusinessThe mysterious AI startup Hark has announced the successful completion of a Series A funding round totaling $700 million (approximately NT$22 billion), capital…
黃仁勳預測 Nvidia 的下一個「全新」2000 億美元市場：用於 AI Agent 的 CPU★ 85
TechCrunch AI25 days agoBusinessNvidia CEO Jensen Huang has recently put forward a major market prediction, stating that Nvidia has its sights set on a brand-new market worth as much as $200…
Railway：打造 Agent 原生雲端平台 — 專訪執行長 Jake Cooper★ 75
Latent Space25 days agoOpinionIn a recent interview on the well-known AI podcast Latent Space, Jake Cooper, CEO of Railway — a next-generation PaaS (Platform as a Service) platform — shared…
網路瘋傳！Figure AI 推出人形機器人 24 小時搬包裹直播，引發社群熱烈關注
Ars Technica AI25 days agoReleaseHumanoid robot startup Figure AI recently launched a highly buzzworthy technology showcase: a 24-hour uninterrupted live stream depicting its latest humanoid…
Vercel Chat SDK 現在整合 AI SDK Tools，簡化聊天介面的工具調用開發★ 75
Vercel Changelog26 days agoReleaseVercel's official Changelog announced that its Chat SDK has now formally incorporated AI SDK Tools functionality. This update is aimed at simplifying the often…
Vercel 推出 WordPress 專用的 Vercel AI Gateway 插件★ 70
Vercel Changelog26 days agoReleaseVercel officially announced the launch of the Vercel AI Gateway plugin for WordPress. Vercel AI Gateway is a service designed to help developers better manage…
Vercel Chat SDK 新增訊息主題（Message Subjects）與直接 SDK 存取功能★ 72
Vercel Changelog26 days agoReleaseThe Vercel AI SDK (specifically its Chat SDK component), which is widely popular among developers, has recently received a feature update aimed at providing…
AI2 發表 OlmoEarth v1.1：更高效的開源地球觀測模型家族★ 70
Hugging Face Blog26 days agoReleaseThe Allen Institute for AI (AI2) has officially released OlmoEarth v1.1 on Hugging Face. This is a brand-new family of open-source foundation models designed…
Hugging Face 推出 Ettin Reranker 重排模型家族：大幅提升 RAG 檢索精度與效率★ 80
Hugging Face Blog27 days agoReleaseIn building Retrieval-Augmented Generation (RAG) systems, accurately locating the most relevant information from a vast document collection has always been the…
Google DeepMind 透過 Co-Scientist 加速基因研究，成功發現逆轉細胞衰老的新型因子★ 85
Google DeepMind Blog27 days agoReleaseGoogle DeepMind has announced a major breakthrough by its AI scientific assistant "Co-Scientist" in the biomedical field. Biologists using the system have…
PaddleOCR 3.5 發布：支援 Transformers 後端，輕鬆執行 OCR 與文件解析任務★ 75
Hugging Face Blog27 days agoReleaseThe well-known open-source OCR (Optical Character Recognition) toolkit PaddleOCR has long been celebrated for its high accuracy, lightweight models, and strong…
烏克蘭無人機創辦人 Yaroslav Azhnyuk 談自主無人機技術棧與無人機經濟學：西方國家正處於昏睡狀態
Latent Space27 days agoCommentaryIn this episode of the Latent Space podcast, the hosts and guest host Noah Smith (author of the well-known economics and technology blog Noahpinion)…
Import AI 457：AI 版 Stuxnet 震網病毒、神祕的 Muon 優化器，以及積極對齊（Positive Alignment）★ 78
Import AI (Jack Clark)27 days agoCommentaryThis issue of Import AI 457, written by Jack Clark, delves into three forward-looking and stylistically distinct topics in the field of artificial…
Google DeepMind 推出 Project Genie 新功能：結合 Street View 街景模擬真實世界，並向全球 Google AI Ultra 訂閱者開放★ 85
Google DeepMind Blog28 days agoReleaseGoogle DeepMind has officially announced that access to its much-anticipated generative interactive environment model, "Project Genie," is being expanded to…
讓使用者更輕鬆了解網頁內容的建立與編輯來源：Google 擴大推廣內容憑證與 SynthID 技術★ 78
Google DeepMind Blog28 days agoReleaseAs generative AI technology becomes more widespread, the internet is increasingly flooded with images and information that are difficult to distinguish as real…
最新開放模型動態 (#21)：開放模型大爆發！Gemma 4、DeepSeek V4、Kimi K2.6、MiMo 2.5、GLM-5.1 等，以及 CAISI V4 評估分析★ 85
Interconnects (Nathan L.)29 days agoCommentaryThis is Issue #21 of the "Open Artifacts" column by well-known AI commentator Nathan Lambert, exploring the explosive growth in the open-weights and…
Google DeepMind 與新加坡建立國家級合作夥伴關係，共同應對醫療、教育與永續發展挑戰★ 75
Google DeepMind Blog30 days agoBusinessGoogle DeepMind has officially announced a new national-level partnership with the Singapore government, aimed at leveraging the most advanced frontier AI…
Cerebras 迎來 600 億美元 IPO：起步緩慢，隨後一飛沖天★ 85
Latent Space30 days agoBusinessAI chip design unicorn Cerebras Systems officially entered its long-awaited initial public offering (IPO) at a valuation of $60 billion. This company…