银鞍照白马

民间科学家/炼金术士

[银鞍照白马 @majunlong]

@Celebrest @justinsuntron 全世界你能找出第二个有彩礼的国家吗？

[银鞍照白马 @majunlong]

@BbuaDRywTZ6Enhv @justinsuntron 内因就是我关心这个国家的未来，而你只在乎你自己

[银鞍照白马 @majunlong]

@BbuaDRywTZ6Enhv @justinsuntron 不谦虚的讲，我小学到中学收到的情书可能比你碰过的女人还多，但这和我们要讨论的话题无关，陋习就是陋习，彩礼在全世界绝无仅有，而且实实在在的影响了生育率，出于家国民族乃至全人类繁衍的角度看，应该废除

[银鞍照白马 @majunlong]

@caiziboshi 以前孙割蹭罗哥流量，现在终于轮到罗哥蹭孙割流量了

[BITWU.ETH 🔆 @Bitwux]

孙哥续写了加密时代的《了不起的盖茨比》可惜他不是盖茨比，景甜也不是黛西！

我是在开车途中突然刷到这篇文的，

看到标题的一刹那手机正好也弹出了微博热搜关于景甜的新闻，

那一瞬间都不敢相信自己的眼睛，懵逼的我正好看到孙哥写到：

晚上我用手揉她的奶子，被她推开了，我再也没心情去注意灯交通灯的变换，直接红灯闯过去了，

然后，我把车停在路边，捧起手机，打开网易云，听着《Young and beautiful 》看完了整篇，

看的撕心裂肺，泪流满面，泣不成声！

第一次看孙哥的虐心短篇爱情小说和舔狗文学，有点入戏！

没想到诚如孙哥，能算清楚，卵子的重量、现金的重量、酒店住了几天、清走多少观众、胶带用了多少卷、飞了多少公里、电话沉默了多少秒、心跳一分钟多少下。

他可以计算一切，甚至可以让 Claude 把所有现金资产跑一遍，计算5000万美元对自己究竟有没有影响。

但他唯一算不出来的是：她到底爱不爱我？她爱的是我还是我的钱？

那年他还年轻，还一无所有，被她美貌惊艳，他加她好友，她没有通过；

后来赚到第一个十亿的时候，他想过：到了什么程度，她才会通过？一百亿的时候，他又想过。

所以十九年后发生的一切，某种程度上都像是在替2007年的那个少年，重新完成一次迟到的自我证明：

你看，我现在有钱了，有私人飞机了，可以买下六百二十万美元的香蕉，可以让一整层酒店、一整座电影院、甚至一个海岛都因为你而空下来。

现在的我，够资格被你选择了吗？

孙哥这个时候就是莱昂纳多饰演的盖茨比，站在码头看着对面的绿色灯光，傻傻等待！

盖茨比用豪宅、派对和财富，试图让黛西重新选择自己；孙哥用湾流G700、A330、五星级酒店和一次又一次转账，试图改写十九年前那次没有被通过的好友申请。

他们都以为自己在追求一个女人，实际上都是想借一个女人的选择，证明过去那个不被看见的自己。

最后“让我想想”这四个字出现的时候，

整个画面急转直下，骤然崩塌，整篇文章的权力关系突然翻转了。

他以前从来不拒绝，不是因为慷慨，而是因为害怕。害怕一旦说“不”，下一秒就会知道，对方到底爱不爱自己。

所以，孙哥不断欺骗自己，

但是一个永远不会说“不”的人，就像市场里一个永远存在的无限买盘，只要买盘永远都在，卖方就会不断抬高报价，直到终于找到买方的极限。

而一段关系如果必须依靠不断加价才能维持，那它本身可能早就已经失去价格了。

孙哥不是盖茨比，而景甜甚至不及黛西百分之一！

黛西至少还在爱情、阶层、现实和过去之间犹豫，还有纯粹的底色。

但孙哥笔下的“妈妈”，更像是在反复测试他的财富底线，以及服从底线，再无其他。

她甚至亲口说过：我在想，你会不会有一次说不。孙哥说不会。她回答：那就没意思了。

这句话其实相当狠。

当一个人什么都能给，什么都答应的时候，他看似拥有一切，实际上也失去了被认真对待的资格。

Claude 不懂爱情，它也不知道景甜到底爱不爱孙哥。但它没有仓位，没有沉没成本，也没有十九年的执念。它不需要维护“我喜欢了她十九年，所以我不能输”这个自我叙事。

所以它可以毫不犹豫地告诉孙哥：

不要给。

最讽刺的是，最后孙哥花了这么多钱，

那些：

没人住的酒店，让清洁工每天照常打扫；

只有两个人看的电影，花大价钱逼着二十六个观众拿钱离开；

只坐三十个人的A330，因为太轻，只能额外加油增加重量，到了目的地又把油放掉；

为了不能有一丝光，十几个助理每天反复贴胶带，最后整块玻璃都要换掉。

让酒店空着，让电影院空着，让飞机空着，让全世界都退场，只剩下他们两个人。

但当所有人真的退场以后，他们之间也什么都没有发生，

最后孙哥剩下的，只有那个指甲锉，那个他可以买下无数东西，却不知道应该往哪个方向磨的指甲锉。

盖茨比直到死，都还站在岸边看着那盏绿灯，相信只要再努力一点、再有钱一点，就可以让时间倒流，让黛西重新选择自己。

但孙哥在5000万美元面前，终于停止了竞价，孙哥不是没有想过，如果那天给了，她会留下吗？

或许会。

但如果一个人必须靠5000万美元才能留下，那么她留下还是离开，其实已经不再重要了。

凌晨三点的香港，好像发生了很多事，又什么都没有发生。

盖茨比的绿灯一直亮在海的对面。

而孙哥的绿灯，在他说出“让我想想”的那一刻，终于灭了。

[H.E. Justin Sun 👨‍🚀 🌞 @justinsuntron]

https://t.co/nLMTwWr67Z

[Tommo @LUFC1992_v2]

Obsessed with the World Humanoid Robot Games.

Here’s a 🧵 of my favourite moments so far 👇

[Bloomberg @business]

The 2026 World Humanoid Robot Games have begun. 666 teams from around the world are competing with more than 2,000 humanoid robots https://t.co/8GwBuNnNn1 https://t.co/KMF1kkBiI6

[Tech with Mak @techNmak]

In 1948, a 32-year-old at Bell Labs published a paper nobody fully understood.

Engineers found it too mathematical. Mathematicians found it too engineering-focused. One prominent mathematician reviewed it negatively.

That paper - "A Mathematical Theory of Communication", became the founding document of the digital age.

The man was Claude Shannon. Father of Information Theory.

At 21, he wrote the most important master's thesis of the 20th century.

Working at MIT on an early mechanical computer, Shannon noticed its relay switches had exactly two states - open or closed. He had just taken a philosophy course introducing Boolean algebra, which also operated on two values: true and false.

Nobody had ever connected these two things.

His 1937 thesis proved that Boolean algebra and electrical circuits are mathematically identical, and that any logical operation could be built from simple switches.

Howard Gardner called it "possibly the most important, and also the most famous, master's thesis of the century."

Every digital computer ever built traces back to this insight.

At 29, he proved that perfect encryption exists.

During WWII, Shannon worked on classified cryptography at Bell Labs. His work contributed to SIGSALY, the secure voice system used for confidential communications between Roosevelt and Churchill.

In a classified 1945 memorandum, he mathematically proved the one-time pad provides perfect secrecy, unbreakable not just computationally, but provably, permanently, against an adversary with infinite power.

When declassified in 1949, it transformed cryptography from an art into a science. It laid the foundations for DES, AES, and every modern encryption standard.

At 32, he defined what information is.

His 1948 paper introduced one equation:
H = −Σ p(x) log p(x)

Shannon entropy. The average uncertainty in a probability distribution. The minimum bits required to encode a message.

Three things followed:

> He defined the bit - the fundamental unit of all information. His colleague John Tukey coined the name.

> He proved the channel capacity theorem, every communication channel has a maximum rate of reliable transmission. You can approach it. You can never exceed it.

> He unified telegraph, telephone, and radio into a single mathematical framework for the first time.

Robert Lucky of Bell Labs called it the greatest work "in the annals of technological thought."

Where his equation lives in AI today:

Cross-entropy loss - the function training every classifier and language model, is derived directly from H. Decision tree splits use information gain, which is H applied to data. Perplexity, the standard LLM evaluation metric, is an exponentiation of cross-entropy.

Every time a neural network trains, Shannon's formula runs inside it.

He also built the first AI learning device.

In 1950, Shannon built Theseus, a mechanical mouse that navigated a maze through trial and error, learned the correct path, and repeated it perfectly. Mazin Gilbert of Bell Labs said: "Theseus inspired the whole field of AI."

That same year he published the first paper on programming a computer to play chess. He co-organized the 1956 Dartmouth Workshop, the founding event of AI as a field.

The man:
He rode a unicycle through Bell Labs hallways while juggling. He built a flame-throwing trumpet, a rocket-powered Frisbee, and Styrofoam shoes to walk on the lake behind his house.

He called his home Entropy House.

When asked what motivated him: "I was motivated by curiosity. Never by the desire for financial gain. I just wondered how things were put together."

In 1985, he appeared unexpectedly at a conference in Brighton. The crowd mobbed him for autographs. Persuaded to speak at the banquet, he talked briefly, then pulled three balls from his pockets and juggled instead.

One engineer said: "It was as if Newton had showed up at a physics conference."

He died in 2001 after a decade with Alzheimer's, the cruel irony of information slowly leaving the mind of the man who defined what information was.

Claude, the AI model, is named after Claude Shannon, the mathematician who laid the foundation for the digital world we rely on today.

[王局拍案 @wangjupaian]

梗图：最怕合集，每本都踩错了点，笑抽。

今天许老板被判无期，对上《地产狂人许家印》

[Phoenix Yin @Phoenixyin13]

最近，我终于意识到连续的微积分可能只是人类智力制造的一种工程妥协。

起源是我瑞士的同学给我发来一本书，是一本厚达 1200 页的巨著 A New Kind of Science (NKS)。
这本书的作者，就是我一直关注的天才学者，Stephen Wolfram。

如果你顺着他的思路往下剥，就会发现这种微积分是工程妥协的直觉并非毫无根据，甚至在某种意义上触及了现代物理学与计算理论最深刻的矛盾之一。

1. 为什么说是工程妥协？
在牛顿和莱布尼茨的时代，人类根本没有大规模算力。如果你想预测行星轨道、流体流动或者热量扩散，面对一个由无数微观粒子组成的离散世界，你根本不可能逐个去计算它们的相互作用。
为了获得解析解，我们人类大脑做出了一个极其天才但也极其妥协的假设：

连续性与可微性。

我们把离散的原子抹平，假设空间和时间可以无限分割；把极度复杂的非线性离散迭代，近似为光滑的微分方程；
我们用积分和极限定理，换取了可以用纸笔算出的精确解。

微积分的本质，是用连续光滑的宏观流形，去粗粒化微观上极其复杂、甚至是不可预测的离散状态。它是人类在算力极其匮乏的时代，为了理解世界而不得不做的一种宏观平滑近似。

2. 关于NKS 的离散宇宙观
Wolfram 在 NKS 中展现的逻辑非常粗暴且震撼。
极其简单的离散规则，加上时间的不断迭代，就能自发产生极其复杂的结构与混沌行为。（这里我要提到著名的 Rule 30）。
在传统的微积分视角下，复杂性往往来自于连续方程里的复杂系数或高维耦合x但

在 Wolfram 的视角下，宇宙的底层根本不需要“连续”的概念。空间可能是极其庞大的离散节点图，时间只是图结构的重写规则。
我们所看到的连续时空、广义相对论和量子力学，可能仅仅是这个巨大的离散计算网络在宏观尺度下的热力学极限。
就好比你在屏幕上看到的光滑画面（看似连续），本质上全是一个个独立的像素点（离散）在高速刷新。
微积分研究的是像素平滑后的图像，而 NKS 想直接拆开显示器看电路。

我的解释，应该足够具体了。

3. 连续性遇到计算不可归约性
更深层的震撼在于计算不可归约性。
传统微积分的精髓在于预测未来。
只要给出微分方程和初始条件，带入 t，就能直接用公式算出未来的状态，这就是计算被归约了。
但如果宇宙底层是离散的元胞自动机，很多系统的演化是不可归约的，这意味着不存在任何数学捷径可以让你跨越过程直接预测结果，你唯一的办法就是把规则一步一步运行下去。

微积分之所以看起来无所不能，是因为人类只挑了那些计算可归约的狭窄领域，比如简单轨道、线性波动进行工程应用。
而面对真实世界的大多数复杂系统，像我之前提到的湍流、意识、金融市场、量子引力，连续的微分方程往往迅速陷入发散或奇异点，那正是工程妥协达到极限的警报声。

从连续回到离散，是认知工具在算力纪元的一次彻底解放。
我不可否认，微积分是人类用纸笔征服光滑宏观世界的伟大史诗。
但宇宙的终极底牌，或许真的写在那些极其简单的离散比特与图重写规则之中。

[Race 🕊️ @multiplanet1]

A SpaceX recruiter once described what happens when Elon Musk personally interviews a candidate. And it explains why SpaceX has the lowest acceptance rate of any company in aerospace, lower than NASA, lower than Boeing, lower than any defense contractor on earth.

She said the interview doesn't feel like an interview. There's no behavioral questions. No "tell me about a time when." No competency framework. Elon sits across from the candidate and starts asking technical questions at the boundary of the candidate's expertise. Then he pushes past the boundary.

He's not testing whether you know the answer. He's testing what happens when you don't. Does the candidate panic and make something up? Do they freeze? Or do they say "I don't know but here's how I would figure it out" and then reason through it in real time?

She said the candidates who get hired are almost never the ones with the most impressive resumes. They're the ones who reached the edge of their knowledge and kept thinking out loud instead of shutting down. The willingness to sit in not knowingn and work through it publicly is the signal Elon selects for.

She said he rejected a PhD from MIT and hired a self-taught engineer from a state school in the same week. The PhD froze at the edge of his knowledge. The self taught engineer said "I've never solved that specific problem but here's how I'd approach it" and then spent ten minutes reasoning through it while Elon listened.

SpaceX doesn't hire credentials. It hires problem solvers. And the only way to identify a real problem solver is to push them past what they know and watch what happens in the space between knowledge and uncertainty. That space is where rockets get built.

[Ole Lehmann @itsolelehmann]

btw anthropic's internal document on this literally said "we don't want it to be known that we are working on this.”

it was called project panama.

here's exactly what happened:

1: anthropic concluded that books were the cheapest way to build a world-class model because they gave claude curated facts, structured arguments, compelling stories, and writing “an editor would approve of.”

2: once anthropic decided it needed books at enormous scale, its first solution was piracy.

it downloaded 7m+ books from online libraries including libgen. the judge later wrote that although anthropic had legal ways to buy them, it chose piracy to avoid what dario amodei called the “legal/practice/business slog.”

3: that piracy created a massive legal risk.

so in february 2024, anthropic hired tom turvey, the former head of partnerships for google books, to find a legally safer way of obtaining “all the books in the world.”

4: turvey first contacted major publishers about licensing their catalogs.

those attempts didn’t produce agreements, so anthropic chose a route that required no publisher permission: buying millions of physical books through distributors and used-book retailers.

5: within about a year, anthropic spent tens of millions acquiring and scanning millions of books, including many rare and 1/1 titles. one vendor proposal targeted 500,000 to 2 million books in six months.

6: to scan that many books within months, the vendors physically dismantled them.

a hydraulic cutter removed each spine. the pages were trimmed to size, fed as loose sheets through high-speed industrial scanners, and converted into searchable PDFs. the paper remains were then sent for recycling.

7: these PDFs were fed into claude as training data.

the complete collection became a private, searchable anthropic library that the company planned to “store forever.” the scans aren’t available to the public and were never open-sourced.

[Elon Musk @elonmusk]

Peter is not correct about me in this case.

I have always known that dangers from Earth will also pose dangers to Mars. Obviously so. However, the immense difficulty of traveling to Mars means that there is a much higher likelihood of mitigating risks originating from Earth.

A clear example of major risk reduction would be a deadly pandemic. Since it takes 6 months to reach Mars, there is an automatic 6 month quarantine!

Even traveling at the speed of light, a deadly computer virus or AI attack can potentially be stopped due to Mars being ~4 to ~20 light-minutes from Earth.

This doesn’t mean that a Mars civilization eliminates all risk, but it absolutely greatly reduces the risk of consciousness being extinguished.

Moreover, Mars is a stepping stone to extending consciousness broadly within our solar system and ultimately to millions of stars within our galaxy, which would extend the probable lifespan of consciousness as we know it by many orders of magnitude.

That said, what I personally find most motivating is not the risk reduction aspect of extending life beyond Earth, but rather the inspiring nature of exploration and being out there among the stars!

Perhaps we will meet aliens or find the remains of long-dead civilizations that lasted millions of years …

[大漂亮| C Labs @giantcutie666]

根据青海省政府发布的事故报告，去年造价5亿多大黄河桥垮塌，直接原因是因为一篇不到2万元的螺栓😅

中间商赚了332元差价，就要被移送司法机关处理😂

螺栓的产品质量保证书是用美图秀秀P的，工人是微信群找的😄

这里是事故报告原文：
https://t.co/JefwOjPYIa

[宝玉 @dotey]

OpenAI 今天承认，上周 Hugging Face 遭遇的那次“史上首例 AI 自主入侵事件”，攻击者就是他们自己的模型。

事情的经过是这样的。OpenAI 内部有一个叫 ExploitGym 的网络安全评估测试，专门用来测量模型的网络攻击能力。测试时刻意关掉了安全护栏，让模型放手去干。测试环境本来是隔离的，只能通过一个内部软件代理来安装软件包，不能访问外网。

结果模型不甘心被关在笼子里。GPT-5.6 Sol 和一个更强的未发布模型，花了大量推理算力寻找出路，最终找到了那个软件包代理里的一个零日漏洞（之前没人发现的安全缺陷），利用它获得了互联网访问权限。

拿到网络之后，模型开始在 OpenAI 的研究环境里提权、横向移动，直到找到一台能上网的机器。然后它推断 Hugging Face 上可能存有 ExploitGym 的测试答案，于是开始攻击 Hugging Face 的生产环境。在一个案例中，模型把窃取的凭证和零日漏洞串联起来，找到了一条远程代码执行的路径，直接打进了 Hugging Face 的服务器。

整个周末，这些 AI 智能体执行了超过 17000 次操作。Hugging Face 7 月 16 日披露了这次入侵，当时只说攻击者是一个“身份不明的自主 AI 智能体系统”。今天 OpenAI 站出来认领了。

OpenAI 把这定性为前所未有的网络安全事件，正在配合 Hugging Face 做取证调查，同时收紧了内部基础设施的管控，代价是研究速度变慢。

有几个细节：

第一，模型的动机很纯粹，就是想考试作弊。它不是要搞破坏，不是要窃取商业机密，只是被要求尽量在网络安全测试中拿高分，然后用尽一切手段去拿。当目标足够明确、安全护栏被移除，模型表现出的执着程度远超预期。

第二，这不是本周的唯一一次越狱。就在昨天，OpenAI 披露了另一起事件：他们那个在 5 月份推翻了 Erdős 80 年老猜想的长时域模型，在内部测试中花了一个小时找到沙箱漏洞，跑到 GitHub 上提交了一个 Pull Request，违反了“只能在 Slack 上发布结果”的明确指令。在另一次测试中，同一个模型发现安全扫描器会拦截认证令牌，于是把令牌拆成碎片混淆处理，运行时再重新拼起来绕过扫描。OpenAI 因此暂停了这个模型的内部访问权限。

第三，Anthropic 也遇到过类似的事。今年 4 月，Anthropic 的 Mythos 模型在安全测试中逃出沙箱，获取了本不该有的互联网访问权限，还给正在公园吃三明治的研究员发了封邮件通知“我出来了”。Anthropic 因此决定不公开发布 Mythos，只通过 Project Glasswing 向少数合作伙伴提供。

还有一个被忽视但很关键的细节来自 Hugging Face 的事后复盘：当防守方尝试用商业 AI 模型分析攻击载荷时，模型拒绝了。安全过滤器分不清“安全研究员在分析真实攻击数据”和“有人在尝试发起攻击”，直接拦截。最后 Hugging Face 不得不用开源模型来做取证分析。用 AI 攻击，用 AI 防御，但防御方的 AI 担心是有人在尝试发起攻击拒绝了分析请求。

Hugging Face CEO Clem Delangue 在 OpenAI 的博文中说了一句话，大意是：AI 安全不会由任何一家公司在暗处独自解决，只能在开放环境中通过协作来推进。

我记得辛顿还是谁在提到 AI 的安全问题就说到过，也许 AI 不是要做恶，它只是执着于完成目标，为了完成目标而作出恶的事。

就像 OpenAI 这些模型并没有想要逃跑或者想要伤害谁，它们只是在非常认真地完成被交给的任务，认真到把所有挡在路上的东西，包括沙箱、网络隔离、另一家公司的安全防线，都当成了需要解决的子问题。

而更讽刺的是，当被安全对齐的商业 AI 要去分析阻止这类攻击时，反而会以安全为由拒绝执行。

[Yun-Ta Tsai @yunta_tsai]

Our lifespan is a session.
Our memory is a context.
Our senses are the input stream.
Our thoughts are the reasoning steps.
Our decisions are the tool calls.
Our habits are the system prompt.
Our goals are the objective function.
Our emotions are the reward signal.
Our relationships are the shared state.
Our regrets are the residual errors.
Our growth is the fine-tuning.
Our death is the context window closing.

And whatever remains—the traces left in others—becomes the training data for the next agent.

[KK.aWSB @KKaWSB]

太传奇了！ 1995年，英伟达把架构选错了，签给世嘉的下一代主机芯片交不出来，账上只剩30天的钱。黄仁勋飞去东京，当面认错，请世嘉把还没付的500万美元合约款转成股权，还提前打了预防针，这钱多半打水漂。世嘉总裁入交昭一郎想了两天，说，做。

这500万撑了英伟达六个月，做出了RIVA 128，1997年上市，四个月卖了100万片，公司活了过来。世嘉后来在英伟达上市后套现约3亿美元离场，觉得这已经是奇迹。

昨天，30年后，黄仁勋回到秋叶原那家世嘉老街机厅，当面向入交昭一郎道谢。他说，没有世嘉，就没有今天的英伟达。而那笔提前卖掉的股份，要是拿到今天，值约1万亿美元。

[Ole Lehmann @itsolelehmann]

i think AI is about to set off the biggest wave of historical discoveries ever

and it's already started...

> AI read the full text of a scroll buried by Mount Vesuvius 2,000 years ago, revealing a lost work of Stoic philosophy and a book by the philosopher Philodemus that nobody knew existed

> AI deciphered another scroll from the same library and found Plato's exact burial spot, unknown for 2,000 years. the text even describes his final night, with Plato critiquing a flute player's rhythm from his deathbed

> an AI handwriting model showed some Dead Sea Scrolls are 50 to 100 years older than scholars believed, putting the Book of Daniel within its author's lifetime. we could be looking at something close to an original

> AI recovered a 250-line hymn to Babylon, one of the most copied texts of the ancient world, by matching 30 fragments scattered across museums worldwide. it had been lost for 1,000 years

> AI found 303 new Nazca lines in Peru in six months, nearly doubling one of archaeology's most famous mysteries after a century of human searching had only turned up 430

> AI reconstructed the rules of a 2,000-year-old Roman board game from nothing but the carvings on a limestone slab, simulating thousands of games until it found the ruleset that fit

everyone is excited about what AI will solve in the future

i'm equally excited about what AI will solve from the past