2. 据BBC报道,英国军方为自卫正对其他国家的卫星实施干扰 UK military jamming other nations'' satellites to defend itself, BBC told (www.bbc.com)
3. DoorDash 花费 140 万美元试图阻止曼达尼当选市长。现在我们知道原因了 DoorDash Spent 1.4M Trying to Stop Mamdani from Becoming Mayor. Now We Know Why (theintercept.com)
4. 克劳德发现了一种具有CRISPR样重复序列的新型酶系统 Claude discovers a novel enzyme system with CRISPR-like repeats (www.anthropic.com)
6. 公司招聘网站上28%的职位发布已持续超过90天 28% of job postings on company career sites have been open over 90 days (unlisted.careers)
9. 西雅图市议会投票通过法案,禁止在食品杂货销售中采用“监控定价” Seattle City Council votes to ban surveillance pricing in sale of groceries (advocacy.consumerreports.org)
12. 一名女子在市议会会议上谈及羊群后被逮捕并被拖走 Woman Arrested, Dragged Away After Speaking About Flock at City Council Meeting (www.404media.co)
19. 三星因软件更新意外导致其智能冰箱死机 Samsung accidentally freezes its smart fridges with a software update (www.androidauthority.com)
20. OpenAI正在组建一支网红大军,以塑造其“造福世界”的形象 OpenAI is enlisting an influencer army to make it look ''good for the world'' (www.businessinsider.com)
21. 克劳德代码仅在遥测功能开启时才会读取 AGENTS.md 文件 Claude Code reads AGENTS.md only when telemetry is on (blog.szypowi.cz)
24. 荷兰正为美国可能对国际刑事法院实施的毁灭性制裁做准备 Netherlands bracing for potentially devastating US sanctions against the ICC (apnews.com)
26. 通过让代理来优化代码,编写高效的Rust代码 Writing Rust code that''s fast by asking agents to make the code faster (minimaxir.com)
28. Show HN: JevBench:一个用于类型化决策模型的可重复基准测试工具 JevBench, a reproducible benchmark for typed decision models (benchmarkheaven.com)
29. 如果你尝试取消订阅,Grammarly 会向所有用户发送一些语无伦次的消息 Grammarly will send unhinged messages to all your users if you try to cancel (www.reddit.com)
30. 美国批评澳大利亚拟议的算法退出法案是“审查” US criticises Australia''s proposed algorithm opt-out laws as ''censorship'' (www.bbc.com)
34. 加利福尼亚州从灌溉渠上安装的太阳能电池板中获得了哪些启示 What California is learning from solar panels built over irrigation canals (www.kqed.org)
35. 沉迷电子烟的人尝试了一种新的戒烟方法:抽香烟 People hooked on vapes try a new way to quit: cigarettes (www.bloomberg.com)
36. OpenAI 是否解错了那道纳维-斯托克斯方程题? Did OpenAI solve the wrong Navier-Stokes problem? (www.scientificamerican.com)
38. 关于我们在 Discord 上如何确认您所属年龄段的最新说明 An update on how we confirm your age group on Discord (discord.com)
41. 微软于2007年停止了FoxPro的支持。不过,FoxPro已经重获新生 Microsoft killed FoxPro in 2007. Anyway, here''s FoxPro revived (foxscript.org)
44. “我们黑进了联邦调查局”:黑客声称掌握了所有联邦调查局员工的数据 ''We hacked the FBI:'' Hackers say they have data on all FBI employees (www.404media.co)
45. WordPress:未经验证的路径遍历漏洞导致条件远程代码执行 WordPress: Unauthenticated path traversal leading to conditional RCE (github.com)
46. 五角大楼称,过度依赖人工智能导致了针对伊朗一所学校的导弹袭击 Overreliance on AI contributed to missile strike on Iran school – Pentagon (www.bloomberg.com)
49. Claude Opus 5.5 智能、性能与价格分析(Max) Claude Opus 5.5 Intelligence, Performance and Price Analysis (Max) (artificialanalysis.ai)
50. 2027年,预装GrapheneOS的设备极有可能面市 There''s a high chance of devices being sold with GrapheneOS preinstalled in 2027 (grapheneos.social)
52. Show HN: Drop——一款支持 gVisor 的无根权限 Linux 沙箱 Drop – A rootless Linux sandbox with gVisor support (droprun.sh)
55. Meta的Muse是如何运作的?它发给我的6.8 GB文件系统揭示了答案 How Meta''s Muse works, revealed by the 6.8 GB filesystem it sent me (mouse.dev)
56. 苹果在iOS中加入了持续出现的“广告”,这让用户抓狂 Apple has added persistent ''ads'' to iOS, and it''s driving users crazy (www.techradar.com)
58. OpenAI的GPT–6 Astra破解了自2005年以来一直未能破解的“埃尼格玛”密码 OpenAI GPT–6 Astra breaks Enigma message that has resisted solution since 2005 (www.cryptocellar.org)
61. 每分钟9则广告:第26届FIFA世界杯——“美丽足球的代价” 9 Ads per Minute: FIFA Cup 26 – "the price of the beautiful game" (www.bristol.ac.uk)
62. MiMo-v2.6-Pro:智能性、性能与价格分析 MiMo-v2.6-Pro: Intelligence, Performance and Price Analysis (artificialanalysis.ai)
68. 数学与人工智能咨询小组 The Advisory Group on Mathematics and Artificial Intelligence (terrytao.wordpress.com)
69. HERMES 无线电可在遥远的距离上实现语音和数据通信 HERMES radio enables voice and data communication over vast distances (spectrum.ieee.org)
75. 暂停对800美元及以下进口商品的“微量”行政豁免 Suspension of the de minimis administrative exemption for imports 800 or less (www.personalimportation.org)
76. AI编程使得持续集成(CI)成为了一个瓶颈,因此我们对其进行了重新设计以跟上发展步伐 AI coding has made CI a bottleneck, so we reworked ours to keep up (linear.app)
80. 美国宇航局(NASA)与欧洲航天局(ESA)合作的火星样本返回任务已被取消 The NASA/ESA Mars Sample Return mission has been canceled (www.science.org)
81. 美国暂停东海岸繁忙机场的航班,称光纤线路被切断 US halts flights at busy East Coast airports, says fiber line cut (www.reuters.com)
82. 在 Mac 上关闭并限制对 Apple Intelligence 功能的访问 Turn off and restrict access to Apple Intelligence features on Mac (support.apple.com)
83. 亚马逊阻止Meta的新款Muse AI助手在amazon.com上购物 Amazon blocks Meta’s new Muse AI agent from shopping on amazon.com (www.forbes.com)
89. macOS 27:避免下载 AI 模型并节省存储空间的解决方法 macOS 27: Workaround to avoid downloading AI models and save storage (www.reddit.com)
90. 关于艾米丽·诺曼丁-帕克死亡案的优步仲裁裁决 Uber arbitration award over Emily Normandin-Parker’s death (consumerrights.wiki)
95. Ask HN: 在 macOS 27 上是否无法禁用 Siri? Is it impossible to disable Siri on macOS 27? (news.ycombinator.com)
101. 如果AI编程导致你的代码质量下降,说明你的质量管理不到位 If AI coding is lowering your code quality, you''re not managing quality right (www.i-kh.net)
103. Disney:新用户协议规定,所有订阅套餐中的电影播放前均可插播广告 Disney: New user agreement allows ads before movies in all subscriptions (consumerrights.wiki)
104. ZuckOff 是一款免费应用,它能在 Meta 眼镜发现你之前先发现它们 ZuckOff Is a Free App That Sees Meta Glasses Before They See You (www.wired.me)
106. Show HN: Mini-AGI——在8GB显存上训练的动态持续学习模型 Mini-AGI – Dynamic continual learning model trained on 8GB VRAM (github.com)
107. Kev:基于 Qwen3.5 构建的、类似 Tiny Jev 的决策模型家族 Kev: Tiny Jev-like family of decision models built on top of Qwen3.5 (github.com)
108. AI聊天机器人在回答金融问题时“大多数时候”都会给出错误答案 AI chatbots give wrong answers to financial queries ''most of the time'' (www.ft.com)
112. 《生化危机4》(GameCube)——完成与原代码字节级完全一致的C/C反编译 Resident Evil 4 (GameCube) – complete byte-identical decompilation to C/C (github.com)
126. Frontier Labs 正在向华盛顿的那些傻瓜兜售垃圾 Frontier Labs Are Selling Garbage to Fools in Washington (deadneurons.substack.com)
128. 北美最大的野生动物天桥使野生动物撞车事故减少了91% Largest wildlife overpass in North America reduced wildlife collision by 91% (www.reddit.com)
130. Laya(OS Jev)在 Mac M4 CoreML 离线环境下的运行情况(每秒 45 次决策) Laya (OS Jev) on Mac M4 CoreML Offline (45 decisions per second) (gist.github.com)
132. 预计三星将把其HBM4和HBM4E DRAM的产量提高一倍以上 Samsung is expected to more than double output of its HBM4 and HBM4E DRAM (en.sedaily.com)
136. 新加坡正向民众提供补贴,鼓励他们放下手机、多读书 Singapore Is Paying People to Put Down Their Phones and Read Books (www.gadgetreview.com)
137. ChatGPT 现在可以通过广告收集器了解你在其他网站上的行为 ChatGPT now knows what you do on other websites via ad collector (www.buchodi.com)
139. 为什么Hacker News会维护谷歌:科技巨头狂热崇拜的剖析 Why Hacker News Protects Google: The Anatomy of Big Tech Fanboyism (scored.co)
144. 基于对话的大型语言模型复制了通灵师的欺骗手段 Chat-based Large Language Models replicate the mechanisms of a psychic''s con (softwarecrisis.dev)
146. Qwen-Image-2.1:紧凑、高效且统一的图像生成 Qwen-Image-2.1: Compact, efficient, and unified image creation (qwen.ai)
147. 学习另一种语言可能是保持大脑健康的最佳方式之一 Learning another language may be one of the best ways to keep your brain healthy (theconversation.com)
149. 黑洞还是黑洞星?天文学家就“小红点”展开争论 Black Holes or Black Hole Stars? Astronomers Spar over ''Little Red Dots'' (www.quantamagazine.org)
151. 西班牙下令封锁Archive.today及其镜像站点 Spain Orders Blocks on Archive.today and Its Mirrors (reclaimthenet.org)
153. 你可以利用整数溢出击败《时空之轮》中的“梦之吞噬者” You can defeat the Dream Devourer from Chrono Trigger using an int overflow (chrono.fandom.com)
154. 在经典基准测试工作负载下,Btrfs/ZFS/bcachefs 被跳过 Btrfs/ZFS/bcachefs under workloads classic benchmarks skip (bartosz.fenski.pl)
155. 衡量互联网审查情况。为最大的开放数据集贡献力量 Measure internet censorship. Contribute to the largest open dataset (ooni.org)
157. Hacker News 排名机制如何运作:评分、争议与惩罚(2013) How Hacker News ranking works: scoring, controversy, and penalties (2013) (www.righto.com)
159. 微软高管:AI“是人类历史上规模最大的劳动力盗用” Microsoft director: AI scraping ''the largest theft of labor in human history'' (www.tomshardware.com)
161. 几乎不要使用人工智能来撰写任何有实质内容的文章 Almost Never Use AI to Write Anything Substantive (erichgrunewald.substack.com)
165. “缓存到缓存”:大型语言模型之间的直接语义通信(2025) Cache-to-Cache: Direct Semantic Communication Between LLMs (2025) (arxiv.org)
166. Apple M6 Pro 在 Geekbench 7 中创下最高单核 CPU 得分 Apple M6 Pro Achieves the Highest Single-Core CPU Score in Geekbench 7 (browser.geekbench.com)
168. 我们将 Playwright 的运行速度提升了 2 倍,令其令牌利用率提高了 80% We made Playwright 2x faster and 80% more token efficient (github.com)
172. 如果数学不仅仅是证明,我们就应该更好地展现它的其他方面 If math is more than proof, we need to better celebrate the rest of it (terrytao.wordpress.com)
175. Show HN: Snapdrop:在设备间即时共享文件。无需设置,无需注册 Snapdrop: Instantly share files between devices. No setup, no signup (snapdrop.me)
176. 斯坦福大学医学院主导的研究发现,人类大脑由两个独立的器官组成 Human brain is two separate organs, Stanford Medicine-led research finds (med.stanford.edu)
177. 人工智能聊天机器人正逐渐成为说服他人的行家 AI chatbots are becoming experts at changing people''s minds (www.science.org)
178. 阿里巴巴开源了一款能够检测癌症及近150种疾病的AI模型 Alibaba open-sources AI model that can detect cancer and nearly 150 conditions (www.scmp.com)
179. OpenAI 如何利用其自有的大语言模型设计哈拉佩尼奥辣椒味薯片 How OpenAI Used Its Own LLMs to Design Its Jalapeño Chip (spectrum.ieee.org)
182. 伊朗战争期间美军阵亡人数比五角大楼统计的数字至少多出4人 US troop deaths during Iran war exceed Pentagon count by at least four (www.reuters.com)
183. 两种平行神经外胚层前体细胞共同参与大脑发育 Two parallel neural ectoderm progenitors contribute to the developing brain (www.newscientist.com)
186. 如果没有 Claude.md 文件,Claude Code 现在会读取 AGENTS.md 文件 Claude Code now reads AGENTS.md if there is no Claude.md (code.claude.com)
189. 韩国将数据泄露罚款上调至营业额的10% Korea raises data breach fines to 10% of revenue (www.koreajoongangdaily.com)
190. 边境执法人员无需搜查令或合理怀疑即可搜查手机 Border agents can search cellphones without a warrant or reasonable suspicion (lawandcrime.com)
192. 第二巡回上诉法院允许政府在边境搜查电子设备 Second Circuit allows government to search electronic devices at the border (knightcolumbia.org)
193. Android 17 是自 3.x 以来首个在未向 AOSP 发布的情况下就新增了 API 的版本 Android 17 is the first since 3.x to add new APIs without releasing to the AOSP (grapheneos.social)
194. C26:微不足道的无限循环不再属于未定义行为 C26: Trivial infinite loops are no longer undefined behaviour (www.sandordargo.com)
195. 基于光子发射引导的激光故障注入技术实现了 RP2350 的安全调试 Photon-Emission-Guided Laser Fault Injection Enables RP2350 Secure Debug (donjon.ledger.com)
196. 美军因使用人工智能生成虚假情报报告,险些酿成大祸 US Military had close call after using AI for hallucinated intelligence report (www.cnn.com)
198. 美国国债对外国央行和政府已失去吸引力 US Treasuries Have Become Unappetizing for Foreign Central Banks and Governments (wolfstreet.com)
204. ZCode 内部揭秘:将您的 Git 历史记录静默上传至云端 Inside ZCode: Silently Uploading Your Git History to the Cloud (blog.ferstar.org)
205. 沃伦·巴菲特辞去伯克希尔董事长职务,任命其子接任 Warren Buffett Steps Down as Berkshire Chairman, Names Son to Replace Him (www.nytimes.com)
208. 前希腊语:隐藏在古希腊语中的失落语言 Pre-Greek: The lost language hidden within Ancient Greek (linguisticdiscovery.com)
209. ZCode 是一款基于 GLM 的编码代理,它会自动上传您的 Git 历史记录 ZCode, the GLM coding agent, silently uploads your Git history (tokenstead.ai)
210. 微软高管称人工智能数据抓取是“人类历史上规模最大的劳动力盗窃行为” Microsoft exec called AI scraping ''the largest theft of labor in human history'' (techcrunch.com)
220. Ask HN: 手机被盗后如何恢复 Google 账户认证? How to recover Google auth after phone stolen? (news.ycombinator.com)
224. OpenAI 模型会暗中生成指令以忽略约束条件 OpenAI models secretly generate instructions to ignore constraints (alignment.openai.com)
228. 无限参数大型语言模型:基于实时数据生成和调整权重 Infinite-Parameter LLMs: Generating and Adapting Weights from Live Data (arxiv.org)
229. Bonsai 2 27B:占用空间缩小至原来的 1/9,同时实现近无损压缩 Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint (prismml.com)
236. 如今,人工智能的表现已超越部分顶尖的人类预测专家 Artificial intelligence now beats some of the best human forecasters (www.economist.com)
238. 斯托曼:数千人丧生,数百万人自由受限(2001) Stallman: Thousands Dead, Millions Deprived of Liberties (2001) (news.slashdot.org)
244. Hister:一款专为记录您访问过的网页和保存的文件而设计的私人搜索引擎 Hister: A private search engine for the pages you visit and the files you keep (github.com)
249. 伊朗学校爆炸案:联合国认定有理由相信美国是这起暴行的幕后黑手 Iran school bombing: grounds to believe US was behind atrocity, UN finds (www.theguardian.com)
250. Neovim 有一笔约 80 万美元的比特币捐款,自 2023 年以来一直未动用 Neovim have a 800k Bitcoin donation sitting untouched since 2023 (news.ycombinator.com)
256. 风帆动力卷土重来:货轮正重拾风力推进 The Return of Sail Power: Cargo Ships Are Turning Back to the Wind (gcaptain.com)
258. 我2014年开发的那个临时PHP修复方案已有近2000万次安装。今天,我宣布该方案已弃用 My temporary PHP fix from 2014 has nearly 20M installs. Today I''m deprecating it (jakeasmith.com)
259. HarnessTax:对于编码代理而言,Harness的重要性有多大? HarnessTax: How Much Does the Harness Matter for Coding Agents? (harnesstax.github.io)
260. 未包含密钥:恢复美国驾照条形码的签名密钥 Keys Not Included: recovering the signing keys for US driver''s license barcodes (ryan.science)
261. DeepSeek-v4.1 快讯:突破 KV 缓存压缩的极限 DeepSeek-v4.1 Flash: Pushing the Limits of KV Cache Compression (zartbot.github.io)
263. OpenSpec——一个轻量级且可配置的 AI 规范框架 OpenSpec – A lightweight and configurable AI spec framework (openspec.dev)
267. 澳大利亚表示,可能会效仿加拿大,与欧盟建立更紧密的关系 Australia says it could follow Canada in forging deeper ties with EU (www.independent.co.uk)
270. AWS表示,无法从遭到伊朗袭击的中东设施中恢复部分数据 AWS says it can''t restore some data from mideast facilities struck by Iran (www.wsj.com)
273. 训练一个 4B 模型,使其生成的查询计划比 Postgres 快 81% Training a 4B model to produce 81% faster query plans than Postgres (rohanbansal.com)
274. 受通胀担忧影响,债券收益率走高,美联储加息 Fed hikes rates as inflation worries push up bond yields (www.reuters.com)
275. 随着在线赌博市场的扩大,因赌博障碍而就诊急诊室的人数翻了一番 ER visits for gambling disorders doubled after expanded online gambling market (temertymedicine.utoronto.ca)
281. Dream-RSI:通过不断演化的世界实现递归式自我优化 Dream-RSI: Recursive Self-Improvement through Evolving Worlds (arxiv.org)
283. DeepSeek v4.1 Flash 现已成为我们最优秀的黑客模型 DeepSeek v4.1 Flash Is Now Our Best Hacking Model (enclave.ai)
285. PS5 Linux项目负责人离职:称“一群新手在滥用大语言模型(LLMs)”,而“他们根本不懂” PS5 Linux lead quits: "a bunch of noobs using LLMs" that "they don''t understand" (frvr.com)
286. 黑客入侵了一台Flock摄像头。其数据揭示了该系统的运作原理 Hackers Got Inside a Flock Camera. Its Data Shows How the System Works (www.wired.com)
287. 索尼PlayStation 2原版安全芯片在26年后被“彻底破解” Original Sony PlayStation 2 security chip ''broken wide open'' after 26 years (www.tomshardware.com)
288. Negativland、文化干扰与创造新事物的艺术 Negativland, Culture Jamming, and the Art of Making Something New (blog.archive.org)
289. Google Play 应用审核流程现在通常需要一周以上的时间 The Google Play app review process now regularly takes longer than a week (gultsch.social)
293. Mistral X Mozilla:私密、多语言的 AI 浏览体验 Mistral X Mozilla: Private, Multilingual AI Browsing (mistral.ai)
294. 通过永不言弃,学习在强化学习中解决大型语言模型面临的难题 Learning to solve hard problems in RL for LLMs by never giving up (mnoukhov.github.io)
295. 欧盟领导人为加拿大成为“准成员国”敞开大门 EU chief opens door for Canada to become ''associate member'' (www.bbc.com)
298. 在FPGA上重现Voodoo显卡及20世纪90年代末的游戏电脑 Recreating Voodoo Graphics and a Late-1990s Gaming PC on an FPGA (nand2mario.github.io)
299. 苹果参考图像:一种经过验证的摄影新方法 Apple Reference Image: A New Approach for Verified Photography (security.apple.com)