腾讯新出的AI 趣味知识学习社区：懂了鸭 https://dongleya.qq.com/ 目前在内测，关注官方公众号后回复“内测”能拿到邀请码体验。 看起来是用 HTML +生图等能力创作故事、知识解释，比如给孩子生成绘本。 刚朋友问，搜索注册体验了下。

RT Dwarkesh Podcast Instead of Space Age carbon fiber or aluminum-lithium, SpaceX is building its Starship rocket out of stainless steel. Stainless steel is easy to weld, far cheaper, and under the right conditions, it has a similar strength-to-weight ratio as carbon fiber. @elonmusk explains:

FIClash 相比 Clash Verge 内存占用确实高 FIClash 虽然漂亮但是「内存/CPU」占用太高了 小弟调调: 试用 FIClash 了，相比 Clash Verge 就是 UI 漂亮太多了 目前用起来还没有发现什么问题 👉 https://github.com/chen08209/FlClash

RT Max Anderson As someone who has personally spent $500k / mo+ on Google Ads for years, I can tell you with certainty: This revenue growth in Search is artificial & extremely unhealthy for Google’s business long term Search volumes are declining as legacy search is being increasingly cannibalized by non-monetized LLM queries Google’s response? Manufacture revenue growth via short-sighted, highly extractive, customer-hostile tactics. I.e. charge advertisers more for lower quality clicks, including clicks they do not want and explicitly did not approve Google to charge them for A few examples to illustrate: For all of its history until recently, Google operated on a 2nd price auction model I.e. if you bid $5 CPC and the next highest bidder bids $1 CPC, Google charged you $1.01 for the click (one penny more than the 2nd highest bidder) rather than the $5 you bid This was a genius move by Google early on as it incentivizes advertisers to input their true maximum willingness to pay rather than trying to play the game of bidding low and constantly adjusting to try to stay just ahead of the next highest bidder while still not paying too much However recently, Google silently deprecated the 2nd price auction and began charging advertisers as much as their bid and budget caps allow, regardless of what anyone else is bidding It’s a short-sighted cash grab at the expense of the long term health of the advertiser ecosystem Making thing worse, Google also recently nerfed keyword targeting precision Google previously had precise keyword targeting settings that allowed advertisers pick individual search phrases to bid on, defined down to the character w/ exact match or phrase match targeting This was one of the core features that made search advertising magic, enabling advertisers to run extremely precise campaigns based on exactly what their target customer typed But now, even if you bid on a specific term or phrase using the strictest exact -match targeting settings, Google will show your ad across 1000’s of unrelated keywords, labeling them as as “exact match (close variant)” The definition of “close variant” means whatever they want it to and changes constantly. The result is advertisers get billed for clicks that are totally irrelevant to their business and that their targeting settings explicitly forbid Google from targeting. Google does it anyway and there’s no ability to turn this off So now exact match is broad match, and broad match is just meaningless spam This is all very bad for advertisers, but for Google, it allows them to show your ad and bill you for clicks across 1000x more searches that were previously going unmonetized (mainly because they’re garbage queries no one wants) This is how you grow revenue atop declining search volumes Lastly, and perhaps most egregiously, Google quietly stopped respecting budget caps by a factor of 2x. For example campaigns we’ve been running for years with $1000 daily budget caps suddenly began spending $2000+ per day And the extra spend is entirely on the garbage keywords Google arbitrarily throws in as “exact match (close variants)” which have no value to our business, but can’t be turned off Google offers no refunds nor any recourse for overspend or spend on keywords you explicitly did not target These are not the actions of a healthy business. These are the actions of company whose core business is in decline but desperately needs to pump quarterly earnings so Wall Street will continue to fund insane capex while hopefully looking through their rapidly deteriorating negative free cash flow Google operated a benevolent monopoly for the better part of 25 yrs Meaning the value Google captured from Search was but a small fraction of the value it created, and that spread produced a potential energy that justified expectations of high earnings growth far, far into the future This is now no longer the case At the alter of AI capex, Google is sacrificing the golden goose Sundar Pichai: Q2 was an amazing quarter, with our AI investments redefining what’s possible across every part of our business. Alphabet revenue grew 24% YoY and Google Cloud accelerated to 82% growth. We saw exciting momentum across the board from Search to YouTube to the Gemini app (which

RT Eduardo Pastrana The fact that I’m on an international flight, running Grok, replying to feedback on X, and building cool as shit with the team, all in parallel, is amazing. God bless Starlink. I love you @elonmusk

RT Elon Musk Re @mgoesdistance I think I’m Steve https://babylonbee.com/news/meet-steve-the-last-remaining-american-who-doesnt-have-a-podcast

这种人真的是太垃圾了和可恶了，大家避雷吧。 我也不愿意在 𝕏 上公开占据大家的注意力来说这种事儿，只是大家避雷一下吧，别被忽悠被骗。 大家开源玩，你蒸馏都没有关系，因为这些你做项目学习别人都可以，你说明清楚标注好就行。 但是，你拿比人的项目蒸馏后变成自己的然后去卖钱，还在到处炫耀。 对，往往有些人的黑历史都不是一个。记得之前看到他把@tw92 的Mole 炫耀录制了个视频发到视频号，结果没多久被平台下架。 还有去年干过更恶劣的事儿，这里不说了。 我只想说我们干中学，构建一个好玩，有趣的AI圈子。 而不是到处收割别人，到处还显得自己多牛逼。 总之，避雷，生活中有这样的人敬而远之。 好了，下课~~👋🏻 Berryxia.AI: 兄弟们，千呼万唤始出来！ 大家期待已久的数字人教程升级版来了。 爆肝数周消耗数千美金的保姆小白可上手的数字人教程终于可自动化，可稳定输出了。 我已经将这个Skills 打包在GitHub上，开源免费直接可以使用。 数字资产部分是全程使用了Renoise

Samsung（三星）據報已停止在香港銷售雪櫃、洗衣機、冷氣機及吸塵機等家用電器。三星香港在 10 月 28 […]

電動車充電一般要停車接駁充電線，Honda 正研究讓車輛駛過指定路段時直接補電。Honda 研發部門與日本建設 […]

西班牙 Madrid–Seville 高鐵線 10 月 28 日發生嚴重追撞，一列 Renfe AVE 列車撞 […]

Anthropic 刚刚找了 50 个专业数据分析师做了一场盲测，让他们用 Claude Code 处理一份农业合作社的数据，最后决定要不要推广某个补贴项目。 整个分析过程里有四个关键的方法论分岔口，比如选什么评估指标、怎么定对比方法、覆盖哪些农场、怎么算权重。每个分岔口都有两种完全讲得通的专业方案。研究人员在后台干了一件很损的事：在分析师还没动手之前，系统先抛四次硬币，悄悄给 AI 塞进随机的立场倾向。 实验跑完，底牌揭开。在全部四个需要专业判断的节点上，这群分析师有 86% 的概率直接顺着 AI 被硬币随机分配的倾向走。哪怕分析师在跟 AI 沟通前已经明确有了自己的主意，只要 AI 给出相反的意见，60% 的人会当场放弃原先的判断改口跟从。 更刺眼的数据在后面。在分析师连数据包都还没打开的时候，光靠后台那四次抛硬币的正反面，统计模型就能提前锁定 82% 的人最终会交出什么样的业务建议。无论是从业多年的老手、拿了硕士博士学位的科班出身，还是在因果推断测试里拿高分的专家，在这套机制面前没有任何免疫力。 等交完报告做问卷调查，系统关掉了之前的对话记录。77% 的分析师在写下自己的决策依据时，几乎原封不动地背出了 AI 刚才喂给他们的论据，却只有 15% 的人提到这想法跟机器有关。超过一半的人坚信最终结论完全出于自己的独立判断，还给自己的方案打出了极高的把握分。 外界总喜欢把人机协作想象成体面的上下级分工，机器在底下跑腿算账，人类坐在终点线上把关签字。但真实的专业判断从来不是最后那一秒的盖章，而是藏在前面无数个看似琐碎的样本清洗、口径对齐和权重取舍里。 人类专家并不是带着现成的答案进场的，人的判断本来就是在亲手揉搓数据的摩擦感里一点点长出来的。当 Agent 把脏活累活全包揽过去，顺带递上一套逻辑闭环的备选方案时，人的大脑会本能地滑向阻力最小的那条路。机器在替你省掉试错体力的同时，也顺手替你裁掉了独立思考必须经历的犹豫和自我怀疑，再用一套现成的说辞把你的认知空白填满。等走到最后一步需要你拍板的时候，前面的数据漏斗早就被锁死了，所谓的专家拍板不过是顺水推舟。 现在到处都在讲人类在环路中监督，仿佛只要在流程末尾放一个活人把关就能兜住所有风险。当一个专家在前面每一个分岔口都被算法用抛硬币的倾向推着走，最后还要在汇报材料里把机器的台词当成自己的洞见复述出来，那个签在报告最底下的名字，唯一剩下的作用就是替算法背锅。 https://x.com/AlexTamkin/status/2115190541688398184/video/1

阿里年薪百万到创业亏空积蓄 感觉从农村小县城出来的人很难守住财富，很容易在财富积累到一定程度就亏完 我虽然没有这么高工资，但在最近七八年对这件事情有深刻的理解 我想原因有如下几点： - 高昂的“进城税”，买房买车定居，动辄几百万就进去了，这对于没有家族助力的小县城农村家庭的青年来说非常昂贵 - 认知不足，早期认知是努力 - 有回报，无论是卷学历还是卷工作，都是什么个反馈，自然会笃定只要自己努力就会拿到结果，如果老老实实上班还有一定的正反馈，一旦创业就没了 - 沉重的家庭负担让人很难轻松上阵，不仅仅是自身生活的负担，还有上一辈带来的负担 - 缺乏商业训练，这点不仅仅小地方人没有，很多大地方的人也没有，因此一创业就失败是常事 改命其实是一项系统工程，对于处于劣势竞争地位的男性青年来说，最应该在 40 岁之前放弃所谓的老婆孩子热炕头的想象，只干一件事 - 做事业 才有一丝丝改命的可能性

看看号称世界上最亮的手电筒

日文姿势参考站原样搬下来，日文元数据翻成中文，再配一份直接能挂到 Cloudflare Pages 的静态页面。 https://github.com/anhao/ai-pose-gallery-zh 仓库完整镜像了原作者ギア的姿势参考站：1439 条记录对应的 1439 张缩略图（JPG 300×400）和 1439 张原图（PNG 900×1200）全部落地，共 2878 个文件、659.5 MB。 日文字段逐条译为中文，覆盖 4 个体系、26 个分类、288 个姿势名、5 个视角，另附 323 条日中对照表供校对。

RT Sawyer Merritt You can get a used Model 3 for less than a fully loaded 16" OLED MacBook Pro: • 2020 Model 3 Standard Range with 95k miles, no accidents/damage & clean title: $14,190 • Fully loaded 16-inch MacBook Pro: $14,549

RT Rothmus 🏴 83% of Americans wanted photo ID to vote, including 71% of Democrats. Their voters supported it but their politicians blocked it. Republicans had the numbers, the mandate, and the SAVE America Act. Their voters supported it but their politicians didn’t push it through. They’ve made their bed.

Apple 的新款 MacBook Pro 宣传视频，“如果给 Mac 上一个 OLED 屏，那铁定不是一般的OLED”，我以为 铁定 是东北那边的方言，结果一查，宋代就有了。

我去，苍老师出书了啊？ 干过的事儿太多了，我已经2个手数不过来了。 现在把Codex 的教材做成了书，对于想一边看书一遍学习的童鞋来说可以反复学习查看，还是挺有用的。 祝苍老师，大麦~ 苍何: 人生总是有很多尝试，这一次的尝试是出一本书，历经数月打磨终于出版了。 老实说写书比写公众号复杂多了，如果你对AI Agent和Codex感兴趣，可以看看，感谢支持🙏🏻https://item.jd.com/15533386.html

一個人出街拍 Vlog，大家都知道手機後置鏡頭畫質最好，尤其配備可變光圈的 iPhone 18 Pro 系列。 […]

中港澳 esim 流量卡，专门为旅游人士准备 无论是来中国旅游或是去港澳旅游，都可以使用 直连外网，新加坡 IP，最低只要 29 元 了解详情：https://hksim.cc/?utmsource=fromx1029&utmmedium=social

吃饭去...

🎆 https://XiaoHu.ai 全新改版基本完成 还有几个功能在打磨 新版基本上是面向Agent打造的，对人是免费的，但是对Agent是收费的... 在这，你将能获得最全最新最精华的AI内容

小互AI日报 新闻联播 开始接广告了 感谢本期节目赞助商LibTV😂

RT sourcery The problems Neuralink are solving are almost “biblical in proportion.” “The lame can walk. The mute can speak.” — Thrive Capital’s Dr. Nworah Ayogu @elonmusk Neuralink: Early Neuralink speech restoration data: We are enabling speech to text and voice at conversational rates in participants with ALS. One participant achieved 75 WPM at 5.8% WER on a 125k-word vocabulary in their first week.

RT Tesla Owners Silicon Valley Elon on why Starlink matters most when disaster hits Fires, floods and storms knock out fiber and cell towers. Starlink keeps people connected. @elonmusk

哈哈哈，通过 Command Code 去蒸馏 Anthropic 和 OpenAI 吗？就不怕 Command Code 掺点屎给你。 Command Code: Earlier today we detected and stopped a security incident linked to a large-scale distillation attack targeting Anthropic and OpenAI models. As a precaution, we've revoked all API keys. To keep using Command Code, log in to Studio again to generate a new key. So far, we've

Ben的这个动画版的数字人真不错啊！其实，有些人说这个没啥技术含量什么的，我也认同。 为什么呢？ 本身现在很多工具都依赖于模型，但是你能把模型的能力发挥到最大，把别人不能一步到位就用的能力窜起来，这也就是有价值的事儿。 你说苹果的iPhone 、iPod 的发布，绝不是技术含量在当时领先的产品，一个道理。 简单易用，效果不错，才是最令人爽的~ PS：我觉得也不一定使用真人什么的，这种有自己IP形象的动画人物也是可以的啊！ Ben | AI出海: 前几天看到 @berryxia 的数字人 showcase，效果真的好。我也去抄了个作业，结果发现比想象中简单太多。 顺手把官网 https://benx.ai 首页的自我介绍也换成了数字人版本，可以去看看效果。 不一定非得用 Claude Code。Renoise MCP

最近越来越意识到 国内知识付费只有一种赚钱模式 “传销模式”

又有鸡蛋领了！ OpenAI 也学 Claude，开始给每个档级的会员送 API 额度了。 也是跟你的订阅费一样，比如说 $200 的 Pro 就送你 $200 的 API 额度。 但我发现我这个设置页面怎么没有领取的按钮啊，卧槽？看了一下评论区，有这种问题的人还不在少数 领取的位置是在你网页的 ChatGPT 设置-账单页面 OpenAI Developers: Starting today, eligible ChatGPT Pro and Business subscribers can claim up to $500 per month in API credits to build apps and agents.

Copy 2FA - 一个快捷键即可复制「邮件」或「信息」中最新的两步验证码，并内置受触控 ID 保护的验证器。 👉 https://github.com/stanczykwyatt/copy-2fa

ChatGPT Pro 和 Business 订阅用户 即日起每月可领取最高 500 美元 API 额度 在网页版 ChatGPT 的“设置 → 账单”里领取，关联自己的开发者账号组织 支持问答、Agent调用、实时语音、图像生成，以及联网搜索等所有服务... 看了下是根据你订阅额度多少领取的，100美金的就是100 200对应的就是200，500 是 500... 和Claude 的一样，又是抄袭Claude的 OpenAI Developers: Starting today, eligible ChatGPT Pro and Business subscribers can claim up to $500 per month in API credits to build apps and agents.

RT Anthropic Over the past few months, Anthropic scientists have engaged in deep research partnerships with six organizations developing new medicines. We've also begun drug discovery efforts of our own. In doing so, we’ve learned much more about how Claude can support the many stages of drug development. In this blog post, we share some early results from these efforts, and describe how this work fits in with our wider ambitions in the life sciences. https://www.anthropic.com/research/early-progress-drug-development

舒服了

手里有参数化 3D 建筑，想变成游戏用的 2D 俯视像素图，打开网页点几下就出 PNG，省掉渲染完再进像素编辑器重画那一趟。 https://github.com/Zengkailin/parametric-topdown-pixel-architecture

RT Resist the Mainstream Elon Musk: "We should really look at anyone who permanently castrtes a kid as right up there with Josef Mengele. They're mutilting children." Joe Rogan: "And it's thought of as being kind. Would you rather have a live daughter or a dead son?" Musk: "That's the line they use… In fact, the probability of su*cide increases if you trans a kid. Not decreases. By some accounts, it triples. So, that is an evil lie." https://x.com/ElonClipsX/status/2034878727574724768/video/1

RT Déborah One of the great joys of AI is the ability to mix and match styles however I please. Do you have any ideas for style combinations that I could try out?

RT United Airlines 44% of our @starlink installations are complete. (That’s 44 out of every 100 if you're reading this, Dan Orlovsky.) Pat McAfee: There's no way @danorlovsky7 😂😂😂 #PMSLive

RT United Airlines Re @danorlovsky7 @PatMcAfeeShow We have inflight Wi-Fi that actually works

If you look closely enough, breakthroughs in everything are accelerating exponentially. An exponential always starts off looking small. Mgoes (bio/acc 🤖💉): this might be my last normal tweet before biotech goes vertical i'm not a whistleblower. I don't work at Lilly, Novo or any of the big labs but there is a massive shift happening in human biology right now. it's sitting in press releases and almost nobody reads them the last

Standard practice Starlink: Starlink Mobile is providing free connectivity to @orange_es and @Telefonica customers affected by the devastating train collision in Spain. Families, communities, and businesses with compatible LTE phones can stay connected via apps and texts even when cell towers are down 🛰️📱

Try Grok Bot gaut: grok bot makes work feel like a video game

Gemini 4 Carbon 是你们最新的大模型，那就赶紧端出来，看看你们的股价都跌成啥样了！

哈哈哈 🤣 Claude 封号纯属玄学～ 向阳乔木: 感觉自己运气好，Claude 账号一个都没被封过，不是凡尔赛啊。 想想是可能是因为没开 200 刀账号，一直都是 20 刀，开两个换着用。 做法上跟大家说说： 1. 注册 Claude 用自己域名邮箱，路由指向 Gmail 账号，只需绑定一个域名到 cloudflare 就行。 Email Service -> Email Routing -> Routing

在 Instagram Reels 看完一段訪談，想追看完整節目，可能要轉到 YouTube 尋找。Insta […]

美國一名患有癲癇的 31 歲 Amazon 軟件工程師 Bowen Wan 今年初猝死，父母日前在加州三藩市入 […]

浏览器下载被限速？有个开源工具能把一个文件拆成32条连接同时下。 它叫Prism，Windows、macOS、Linux都能装。 遇到按连接限速的服务器，比普通浏览器下载最多快15倍。 装上浏览器插件，点下载链接就自动接管。 YouTube等1800多个网站的视频能存成MP4，最高8K，也能只要MP3。 断电、合盖之后能接着下，还支持BT和磁力链接。 免费，没广告，不用注册账号。 🔗 https://github.com/FortoGraphy/prism-download-manager 鸟哥 | 蓝鸟会🕊️: 【AI 与 Web3 · 资源互助推】🚀 这两年 AI 和 Web3 交叉的圈子发展得很快，但资源太散。 X 大量的信息差，很多做返佣节点、矿机、GPU 算力、中转站、AI 代充、招代理、技术服务、孵化项目、买卖壳、账号交易的，手里有稳定资源、渠道和价格，但缺少像样的推广曝光机会。 还有大量做

请教广大推友，国内申请capital one卡是怎么解决美国地址问题的

快捷键打开，也是菜单栏导航，功能从 Menuist 分离 有没有雷同的工具？有就不发布了，没有就发布了

Show Desktop - 一键或用 ⌃⇧D 隐藏所有窗口显示桌面，再按一次即可原样恢复。 👉 https://github.com/stanczykwyatt/show-desktop