4. 首个零知识证明漏洞出现,其表现与预期不符 The First ZK Exploits Happened, and They Weren''t What We Expected (blog.zksecurity.xyz)
5. 特朗普总统下令联邦机构停止使用Anthropic President Trump orders federal agencies to stop using Anthropic (techcrunch.com)
8. 我让80名幼儿园和一年级学生引导人工智能开发了一款游戏 I had 80 kindergarteners and first graders prompt AI to build a game (serene-licorice-4b5eec.netlify.app)
10. SlimClaw:5分钟即可设置的个人AI助手 SlimClaw: A Personal AI Assistant You Can Set Up in 5 Minutes (ganeshan007.github.io)
11. 潜意识的消亡与潜意识的诞生 The Death of the Subconscious and the Birth of the Subconsciousness (3amto5amclub-wuaqr.wordpress.com)
14. 自建NVR升级方案:树莓派CM5搭载Hailo-8 AI芯片与PoE技术 Self-Hosted NVR Upgrade: Raspberry Pi CM5 with Hailo-8 AI and Poe (www.youtube.com)
15. 互联网上最常见的用户界面?——改版检票闸机与验证页面 The Most-Seen UI on the Internet? Redesigning Turnstile and Challenge Pages (blog.cloudflare.com)
16. 特朗普要求政府停止使用Anthropic的人工智能系统 Trump tells government to stop using Anthropic''s AI systems (www.nbcnews.com)
17. 为何SWE-bench Verified不再衡量前沿编码能力 Why SWE-bench Verified no longer measures frontier coding capabilities (openai.com)
20. 特朗普下令联邦机构停止使用Anthropic的人工智能技术 Trump orders federal agencies to stop using Anthropic''s AI technology (www.cbsnews.com)
21. 特朗普总统禁止在政府系统中使用Anthropic President Trump bans Anthropic from use in government systems (www.npr.org)
22. 马萨诸塞州警察局正大肆采购无人机监控设备 Massachusetts State Police are on a drone surveillance shopping spree (binj.news)
23. 如何在本地运行万亿参数大型语言模型:AMD锐龙AI Max集群指南 How to Run a One Trillion-Parameter LLM Locally: AMD Ryzen AI Max Cluster Guide (www.amd.com)
27. 特朗普下令政府停止使用Anthropic,此前五角大楼僵持不下 Trump Orders Government to Stop Using Anthropic After Pentagon Standoff (www.nytimes.com)
29. 特朗普指示联邦机构停止使用Anthropic Trump directs federal agencies to cease use of Anthropic (www.reuters.com)
31. Spotify的终结:流媒体为何即将过时 The Death of Spotify: Why Streaming Is Minutes Away from Being Obsolete (joelgouveia.substack.com)
32. 美国将在六个月内将Anthropic公司排除在政府合同之外 USA to cut Anthropic from government contracts in six months (www.ft.com)
33. 2011年至2022年间,55岁以下成年人因心脏病发作死亡的人数有所上升。 Heart attack deaths rose between 2011 and 2022 among adults younger than age 55 (newsroom.heart.org)
35. 恶意NPM包利用Pastebin隐写术部署凭证窃取器 Malicious NPM Packages Use Pastebin Steganography to Deploy Credential Stealer (socket.dev)
37. 特朗普下令美国各机构停止使用Anthropic,此前五角大楼发生争执 Trump Orders US Agencies to Drop Anthropic After Pentagon Feud (www.bloomberg.com)
39. 波兰计划禁止儿童使用社交媒体,挑战美国科技巨头 Poland Plans Social Media Ban for Kids in Challenge to US Tech (www.bloomberg.com)
40. TaskForge——OpenClaw机器人的不可变编排系统 TaskForge – immutable, orchiestration for OpenClaw bots (github.com)
41. 当人工智能能帮你完成作业时,上学还有什么意义? What''s the Point of School When AI Can Do Your Homework? (www.404media.co)
42. 特朗普下令联邦机构"立即"停止使用Anthropic人工智能技术 Trump orders federal agencies to stop using Anthropic AI tech ''immediately'' (www.cnbc.com)
43. 特朗普因与五角大楼的人工智能争端将Anthropic列入黑名单 Trump moves to blacklist Anthropic over AI fight with Pentagon (www.axios.com)
45. 美国宇航局在发射后失去了一艘月球飞船。一份新报告详细说明了故障原因。 NASA lost a lunar spacecraft after launch. A new report details what went wrong (www.npr.org)
49. 基础设施即代码引擎的变形测试 [pdf] Metamorphic Testing for Infrastructure-as-Code Engines [pdf] (programming-group.com)
51. 新型"火星GPS"系统使毅力号能精确定位至25厘米范围内 New ''Mars GPS'' lets Perseverance pinpoint its location within 25 centimeters (phys.org)
52. OpenAI解雇一名员工,因其在预测市场中进行内幕交易 OpenAI Fires an Employee for Prediction Market Insider Trading (www.wired.com)
54. 美国宇航局宣布对阿耳忒弥斯登月计划进行重大调整:"我们必须重返月球" NASA announces major overhaul of Artemis moon program: "We''ve got to get back to (spaceflightnow.com)
55. AdaptiveCpp的新Metal后端支持在苹果GPU上运行CUDA方言 AdaptiveCpp''s new Metal backend to support CUDA dialect on Apple GPUs (github.com)
56. 《德军总部:敌军地带》在emscripten上的实现,WS中继用于在线浏览器对战 Wolfenstein: Enemy Territory in emscripten, WS relay for online browser matches (et.klaussilveira.com)
57. 特朗普政府官员着手废除保护美国免受化学灾害的系统 Trump officials move to kill system that protects US from chemical disasters (www.theguardian.com)
58. 美国宇航局宣布阿耳忒弥斯三号任务不再以载人登月为目标 NASA announces Artemis III mission no longer aims to send humans to moon (www.theguardian.com)
59. 即便借助编程助手,你仍无法在周末重现StackOverflow的奇迹 Even with Coding Agents, You Still Can''t Recreate StackOverflow in a Weekend (orischwartz.com)
60. Hyprland 0.54 作为 Wayland 合成器的重大更新发布 Hyprland 0.54 Released as a "Massive" Update to This Wayland Compositor (www.phoronix.com)
64. 为何每次人工智能编程突破在90天内都显得司空见惯 Why every AI coding breakthrough feels normal within 90 days (www.augmentcode.com)
66. 我从不在电话里估算。这是我给自己定下的最佳工程准则。 I never estimate on the call. Best engineering rule I made for myself (read.perspectiveship.com)
67. 在IsDialogMessage内部拦截消息,安装消息过滤器 Intercepting messages inside IsDialogMessage, installing the message filter (devblogs.microsoft.com)
68. 量子摇篮:作为生命量子搜索引擎的生化前矿物 The Quantum Cradle: Prebiotic Minerals as Quantum Search Engines for Life (qspace.fqxi.org)
72. 联邦调查局遭黑客入侵,爱泼斯坦档案中100TB数据被删除[pdf] FBI was hacked and 100tb from the Epstein file deleted [pdf] (www.justice.gov)
73. 或许人们对艺术创作的成功持怀疑态度是有原因的 Perhaps People Are Cynical About Success in the Creative Arts for a Reason (freddiedeboer.substack.com)
79. Xbox更换游戏业务负责人,Xbox发展历程,Xbox未来何去何从 Xbox Replaces Head of Gaming, Xbox History, Whither Xbox (stratechery.com)
80. 在针对中国学生的打压浪潮中,某美国大学似乎在阻挠他们入学 Amid crackdown on Chinese students, one US university appears to block them (www.theguardian.com)
81. Signal推出8.0版本,新增Signal安全备份功能 Signal launches version 8.0 with Signal Secure Backups (aboutsignal.com)
83. 从60 APM到60 Agent:一位不情愿的皈依者对代理式工作流的指南 From 60 APM to 60 Agents: A Reluctant Convert''s Guide to Agentic Workflows (github.com)
84. 国会发现数据经纪商使消费者损失数百亿美元 Congress finds data brokers cost consumers tens of billions of dollars (themarkup.org)
85. 所有权问题:为何我们会变得占有欲强,以及如何应对 The Ownership Problem: Why We Get Territorial and What to Do About It (www.thecaringtechie.com)
88. 华盛顿州热线来电者听到带西班牙口音的人工智能语音——美联社新闻 Washington state hotline callers hear AI voice with Spanish accent – AP News (apnews.com)
89. CognOS——人工智能决策信任验证网关 [开源] CognOS – Trust Verification Gateway for Every AI Decision [Open Source] (github.com)
90. FemtoClaw:面向ESP32和树莓派Pico的OpenClaw/PicoClaw超轻量级移植版 FemtoClaw: Ultralight Port of OpenClaw/PicoClaw for ESP32 and Raspberry Pi Pico (github.com)
91. 双子星卫星连续十天出现不同故障,且需求持续攀升 Gemini''s 10 days of different outages and increasing high demand (aistudio.google.com)
92. 从一个独裁父亲到另一个:莫妮卡在朝鲜的失落童年 From one dictator dad to another: Monica''s lost childhood in North Korea (www.smh.com.au)
94. 代理式工作流时代(以及为何80%的可靠性是失败) The Era of Agentic Workflows (and why 80% reliability is a failure) (project-1960fbd1.doanything.app)
97. 一个副项目被互联网封禁:谷歌安全浏览讨厌302状态码 A side project got banned from the internet: Google safe browsing hates 302 (trysound.io)
98. GitHub Copilot命令行界面下载并执行恶意软件 GitHub Copilot CLI Downloads and Executes Malware (www.promptarmor.com)
100. 人类中心主义者拒绝在人工智能安全保障问题上向五角大楼妥协,争议临近最后期限 Anthropic refuses to bend to Pentagon on AI safeguards as dispute nears deadline (apnews.com)
102. 图形用户界面的历史:WIMP设计的兴起(与衰落?) History of the Graphical User Interface: The Rise (and Fall?) Of WIMP Design (www.uxtigers.com)
104. 儿科协会建议对2至10岁的儿童进行胆固醇筛查 Pediatric society recommends cholesterol screening for kids between 2 – 10 years (www.theglobeandmail.com)
105. 不可饱和大型语言模型基准——评估大型语言模型的技能、可靠性与元认知能力 Unsaturable LLM Benchmark – Rating LLM Skill, Reliability, and Metacognition (unsaturable.com)
106. 由人类脑细胞制成的生物计算机能否运行《毁灭战士》? Could a biocomputer made from human brain cells play DOOM? (gigazine.net)
109. 独立出版社暴君书屋在易主后重返书坛 The indie publisher Tyrant Books is returning, under new ownership (lithub.com)
110. MCP恐怖故事:GitHub提示注入数据劫案[2025] MCP Horror Stories: The GitHub Prompt Injection Data Heist [2025] (www.docker.com)
112. 神经吉他效果器——为Daisy Seed Arm Arm Cortex-M7优化NAM算法 Neural Guitar Pedal – Optimizing NAM for Daisy Seed Arm Cortex-M7 (www.tone3000.com)
113. 杰克·多尔西因Square母公司Block的人工智能技术进步将裁员4000人 Jack Dorsey to cut 4k jobs due to AI advances at Square parent Block (www.theguardian.com)
115. 《反人性卡片》关税退税声明及表格 Cards Against Humanity Statement and Form for Tariff Refunds (www.getyourfuckingmoneyback.com)
117. 现在可以明确地说:特朗普正在策划一场十一月的政变 It Can Now Be Plainly Said: Trump Is Planning a November Coup D''État (newrepublic.com)
119. 欧洲13至15岁女孩的吸烟率在同年龄段中最高 European girls aged 13-15 have highest rate of tobacco use for age group (www.theguardian.com)
120. 汉堡王推出人工智能耳机,类似于《Manna的故事》 Burger King implements AI headsets, similar to Story of Manna (apnews.com)
123. 《超时空》与《卡莉之歌》作者丹·西蒙斯去世,享年77岁 Dan Simmons, author of Hyperion, Song of Kali, dead at 77 (en.wikipedia.org)
124. 一位对AI代理编程持怀疑态度的人尝试了AI代理编程,过程极其详尽 An AI agent coding skeptic tries AI agent coding, in excessive detail (minimaxir.com)
125. 十二月发生逆转:人工智能编程领域六个月的反转 Something Flipped in December: AI Coding''s Six-Month Reversal (medium.com)
127. BeerPAN 35毫米单反相机让全景胶片摄影触手可及 BeerPAN 35mm SLR Makes Panoramic Film Photography Accessible (petapixel.com)
132. 为AI代理提供SSH访问权限而不授予私钥 Giving AI Agents SSH access without giving them your secret keys (patrickmccanna.net)
133. 拜托了,拜托了,拜托了,请停止使用密钥来加密用户数据 Please, please, please stop using passkeys for encrypting user data (blog.timcappalli.me)
135. 杰夫·加洛韦,这位以跑走法激励众人的奥运选手,于80岁高龄辞世 Jeff Galloway, Olympian who inspired people with his run-walk method, dies at 80 (www.nbcnews.com)
136. 一款用于无线地下健康监测的智能地钉平台 A smart nail platform for wireless subsoil health monitoring (www.nature.com)
139. 约束理论:《蓝光》创造无用之能(2007) Theory of Constraints: "Blue Light" creating capacity for nothing (2007) (theoryofconstraints.blogspot.com)
140. 工会称20名教师因邮件被转入垃圾邮件文件夹而遭停职 Emails sent to spam folder lead to licence suspension for 20 teachers says union (www.cbc.ca)
141. PostHog现已开始与Reddit和LinkedIn共享新用户的哈希化电子邮件地址。 PostHog now shares hashed emails of new users with Reddit and LinkedIn (www.uxwizz.com)
145. 无注意力的记忆:来自代数运算符的短语境语言 Memory Without Attention: Short‑Context Language from an Algebraic Operator (www.blogosvet.cz)
146. Sudo-rs 每输入一个字符就回显 *,这会破坏安全措施 Sudo-rs echos * for every character typed breaking security measures (bugs.launchpad.net)
148. Rm -RF Salesforce;我用Git仓库和AI助手替换了我们的客户关系管理系统 Rm -RF Salesforce; I replaced our CRM with a Git repo and an AI Agent (twitter.com)
150. Meta试图阻止律师在庭审中询问扎克伯格关于2310亿美元财富的问题 Meta tried to block lawyers from asking Zuckerberg about 231B fortune at trial (nypost.com)
152. 汉堡王测试人工智能耳机,追踪员工是否使用"请"、"谢谢"等礼貌用语 Burger King testing AI headsets to track if employees say ''please'', ''thank you'' (thehill.com)
154. 世界杯效应是否真实存在?美国职业足球大联盟即将揭晓答案 Is the World Cup bump real? MLS is going to find out (www.theguardian.com)
160. Pydantic Monty:你可能不需要完整的沙盒环境 Pydantic Monty: you probably don''t need a full sandbox (pydantic.dev)
165. Factory推出"任务"模式:多日自主软件开发 Factory introduces "Missions": multi-day autonomous software development (factory.ai)
168. "这显然是假的":奥运冰球明星否认白宫AI合成视频 ''It''s clearly fake'': Olympic hockey star disavows AI-generated White House video (www.politico.com)
169. 俄罗斯6000家国防工业工厂及百万员工数据遭泄露 Data on 6k Russian Defense Industry Factories and Million Employees Leaked (militarnyi.com)
171. "如果真有轮回转世,我愿来世成为债券市场。" "If there is reincarnation, I want to come back as the bond market." (quote.org)
174. 基于二进制幂的量化技术在高效FPGA基GRU架构中的应用 Power-of-Two Quantization for Efficient FPGA-Based GRU Architectures (www.mdpi.com)
175. QuiverAI在Design Arena的SVG基准测试中击败Gemini 3.1 Pro(1502 Elo评分) QuiverAI beats Gemini 3.1 Pro on SVG benchmarks on Design Arena (1502 Elo score) (www.designarena.ai)
183. 爱泼斯坦曾试图在斯坦福大学建立行为工程研究所 Epstein Sought to Establish Behavioral Engineering Institute at Stanford (stanfordreview.org)
184. Vibe编码的Lovable托管应用存在大量基础缺陷,1.8万用户信息遭泄露 Vibe coded Lovable-hosted app littered with basic flaws exposed 18K users (www.theregister.com)
185. Sudo-rs默认启用Resolute Raccoon的密码反馈功能 Sudo-rs enables pwfeedback by default for Resolute Raccoon (discourse.ubuntu.com)
186. ChatGPT在分诊建议结构化测试中的健康表现 ChatGPT Health performance in a structured test of triage recommendations (www.nature.com)
187. Pix – 支持新的Direct3D功能及各类提升用户体验的改进 Pix – support for new D3D features and misc quality-of-life improvements (devblogs.microsoft.com)
188. 科学家称,格陵兰岛最大的冰川可能即将达到临界点 Greenland''s largest glacier could soon reach a tipping point, scientists say (phys.org)
190. 着色器模型6.9零售版及新版D3D12改进 Shader Model 6.9 Retail and New D3D12 Improvements (devblogs.microsoft.com)
191. 破以成:断裂如何塑造组织与器官 Break It to Make It: How Fracturing Sculpts Tissues and Organs (www.quantamagazine.org)
192. 人工智能通过注重隐私的手部图像精准识别医疗疾病 AI accurately spots medical disorder from privacy-conscious hand images (www.kobe-u.ac.jp)
194. 我家猫总往台面上跳……于是我造了个机器人[视频] My Cat Kept Jumping on the Counter..So I Built a Robot [video] (www.youtube.com)
197. 一项旨在缩短人工智能生成代码验证差距的实验性项目 An experimental project to shorten the verification gap in AI generated code (www.kurrent.io)
201. 美国宇航局宣布对阿耳忒弥斯计划进行重大调整,因存在安全隐患及延误问题 NASA announces major overhaul of Artemis program amid safety concerns, delays (www.cbsnews.com)
203. 哲学上只有一个严肃的问题,那就是自杀。 There is but one serious philosophical problem, and that is suicide (blog.danielyj.com)
204. 苹果公司表示"本周将有大事发生"。以下是我们预期的内容: Apple says it has "a big week ahead." Here''s what we expect to see (arstechnica.com)
207. 新一代航天器如何压垮我们的通信网络 How Next-Gen Spacecraft Are Overwhelming Our Communication Networks (atempleton.bearblog.dev)
208. 赋予你的OpenClaw智能体内在生命——情感、梦想、自我进化 Give your OpenClaw agent an inner life – emotions, dreams, self-evolution (github.com)
211. 美国联邦通信委员会(FCC)就体育赛事直播从电视广播转向流媒体平台展开调查 FCC opens inquiry into movement of live sports from broadcast TV to streaming (www.nbcsports.com)
212. 金融科技专家称,Perplexity克隆了彭博终端机,其年费高达3万美元 Finance techie says cloned Bloomberg''s 30k/year Terminal with Perplexity (www.tomshardware.com)
214. 宝马在德国生产基地部署人形机器人AEON BMW deploys the humanoid robot AEON in production sites in Germany (robotics.hexagon.com)
215. 电磁反馈:制作电子声音雕塑 [视频] Electromagnetic Feedback: Making an electronic sound sculpture [video] (www.youtube.com)
217. 我们对7款开源AI代理的安全性进行了审计——以下是我们的发现 We Audited the Security of 7 Open-Source AI Agents – Here Is What We Found (grith.ai)
218. 中国军队的内部清洗比你想象的更深入 The Purges Within China''s Military Are Even Deeper Than You Think (chinapower.csis.org)
219. 沙盒环境还是裸机环境?人工智能代理部署的统计与研究 Sandboxed or bare metal? Statistics and study on AI agent deployment (internetwarte.eu)
220. 拾荒天才坂茂:用纸建造大教堂与抗震避难所 Scavenger genius Shigeru Ban: building cathedrals and quake shelters with paper (www.theguardian.com)
221. 匿名认证:为访客账户创建访问令牌 Anonymous Authentication: Creating access tokens for guest accounts (github.com)
222. 专家敲响警钟:ChatGPT Health无法识别医疗紧急情况 Experts sound alarm after ChatGPT Health fails to recognise medical emergencies (www.theguardian.com)
223. 巴基斯坦国防部长称该国与阿富汗之间存在"公开战争" Pakistan''s defense minister says that there is an ''open war'' with Afghanistan (apnews.com)
226. 开源计算器固件DB48X因年龄验证功能禁止使用CA/CO键。 Open source calculator firmware DB48X forbids CA/CO use due to age verification (github.com)
227. 特斯拉中国交付周期骤降至1-3周,同时再次延长融资期限 Tesla China delivery times collapse to 1-3 weeks as it extends financing again (electrek.co)
228. 蒙德里安作品进入公共领域,其遗产管理方对此表示异议。 Mondrian Entered the Public Domain. The Estate Disagrees (copyrightlately.com)
230. 美国宇航局调整阿耳忒弥斯计划以加速重返月球 NASA shakes up its Artemis program to speed up lunar return (arstechnica.com)
232. Snyk Agent Scan:面向AI智能体、MCP服务器及智能体技能的安全扫描工具 Snyk Agent Scan: Security scanner for AI agents, MCP servers and agent skills (github.com)
234. 中国DeepSeek在英伟达顶级芯片上训练AI模型,尽管面临美国禁令 China''s DeepSeek trained AI model on Nvidia''s best chip despite US ban (www.reuters.com)
235. Canonical今年大力推广RISC-V架构,推出Ubuntu 26.04 LTS版本 Canonical Talks Up RISC-V This Year with Ubuntu 26.04 LTS (www.phoronix.com)
236. 加州新法规要求所有操作系统账户必须进行年龄验证 New California law requires age verification for all OS accounts (www.pcgamer.com)
238. 政府机构对埃隆·马斯克旗下Grok聊天机器人的使用发出警告 Government Agencies Raise Alarms About Use of Elon Musk''s Grok Chatbot (www.wsj.com)
239. 第十巡回法院:第四修正案不支持对示威者设备进行广泛搜查 Tenth Circuit: 4th Amendment Doesn''t Support Broad Search of Protesters'' Devices (www.eff.org)
240. 五角大楼威胁Anthropic是错误之举 The Pentagon is making a mistake by threatening Anthropic (www.understandingai.org)
241. 山姆·奥尔特曼表示,OpenAI在五角大楼争端中与Anthropic共享红线 Sam Altman says OpenAI shares Anthropic''s red lines in Pentagon fight (www.axios.com)
246. OpenAI完成110亿美元融资(亚马逊、英伟达、软银参与投资) OpenAI''s 110B funding round (investments from Amazon, Nvidia, SoftBank) (www.reuters.com)
248. 使用Renovate和Vendir实现仓库间文件同步自动化 Automating the syncing of files between repos with Renovate and Vendir (www.jvt.me)
250. Block公司因拥抱人工智能裁员40% Block Cuts 40% of Its Work Force Because of Its Embrace of A.I (www.nytimes.com)
252. OpenAI创始人山姆·奥尔特曼呼吁与赫格塞斯在人类优先原则之争中缓和对峙 OpenAI''s Sam Altman Calls for De-Escalation in Anthropic Showdown with Hegseth (www.wsj.com)
255. 数据中心邻避主义者正扼杀价值1万亿美元的人工智能基础设施 Data Center NIMBYs Are Killing 1T in AI Infrastructure (garryslist.org)
258. 宝马利用专利和商标使汽车DIY维修变得困难重重 BMW using Patents and Trademark to Make DIY Car Repairs Much Harder (www.jalopnik.com)
260. 人工智能生成产品不会引发SaaS末日 AI-Generated Products Won''t Trigger a SaaSpocalypse (pawelbrodzinski.substack.com)
262. OpenAI完成110亿美元融资,创下最大规模私募融资纪录 OpenAI closes 110B funding round in largest private financing (www.cnbc.com)
263. IBM研究发现:人工智能加速网络攻击,应用程序漏洞利用激增44% 44% Surge in App Exploits as AI Speeds Up Cyber-Attacks, IBM Finds (www.infosecurity-magazine.com)
265. Bash命令在处理复杂问题时优于向量搜索 Bash commands outperform vector search for complex questions (chrisweves.com)
266. 纳米香蕉2——谷歌免费4K人工智能图像生成器 Nano Banana 2 – Google''s Free 4K AI Image Generator (nanobananaflash.io)
268. 山姆·奥尔特曼:我们从亚马逊、英伟达和软银获得了110亿美元融资。 Sam Altman: We raised a 110B round from Amazon, Nvidia, SoftBank (twitter.com)
269. 重定向巨细胞病毒免疫反应对抗胰腺癌以实现免疫治疗 Redirecting cytomegalovirus immunity against pancreas cancer for immunotherapy (jitc.bmj.com)
272. 氛围研究,或我如何在四天内写完一篇学术论文 Vibe Research, or How I Wrote an Academic Paper in Four Days (vincent.codes.finance)
273. 五角大楼着手开发针对中国网络行动的人工智能工具 Pentagon moves to build AI tools for China cyber operations (www.ft.com)
274. 评估人工智能能力的专家级学术问题基准——HLE A benchmark of expert-level academic questions to assess AI capabilities – HLE (www.nature.com)
276. 我们理应拥有更优质的 JavaScript 流 API We deserve a better streams API for JavaScript (blog.cloudflare.com)
279. 亚马逊故障——无法查看订单历史或下单 Amazon Down – unable to view order history or make orders (downdetector.co.uk)
281. MediaFM:Netflix媒体理解的多模态人工智能基础模型 MediaFM: The Multimodal AI Foundation for Media Understanding at Netflix (netflixtechblog.com)
283. 伊丽莎白·巴伦出任PHP基金会新任执行董事 Elizabeth Barron Is the New Executive Director of the PHP Foundation (thephp.foundation)
284. 迪士尼指控谷歌利用人工智能实施版权侵权 Disney Accuses Google of Using AI to Engage in Copyright Infringement (variety.com)
286. 地球观测图像数据集应采用表格形式,而非文件夹形式。 Earth Observation image datasets should be tables, not folders (blog.terrafloww.com)
290. Banuba的AI语音照片API将静态图像转化为栩栩如生的会说话视频 Banuba''s AI Talking Photo API Turns Static Images into Lifelike Talking Videos (www.banuba.com)
291. Apache Iggy 向基于 io_uring 的每核线程设计迁移之旅 Apache Iggy''s migration journey to thread-per-core design powered by io_uring (iggy.apache.org)
292. 生成式人工智能使用与美国成年人的抑郁症状 Generative AI Use and Depressive Symptoms Among US Adults (jamanetwork.com)
293. 破碎的令牌:马拉雅拉姆语模型的令牌化技术 The Broken Token: Tokenization for Malayalam Language Models (thottingal.in)
296. 西班牙工程师发现"智能"吸尘器存在漏洞,成功控制7000台设备 Spanish engineer reports flaw in ''smart'' vacuums, takes control of 7k devices (www.theguardian.com)
298. EEGFrontier——从零开始打造的开源脑电波板 EEGFrontier – an open-source EEG board built from scratch (github.com)