Linxiang Zhao

AI researcher

[Linxiang Zhao @Nat87a]

@Duduook 是的，我甚至把小红书哔哩哔哩还有炒股软件都删了🫠之前不经意就占用太多时间

[Linxiang Zhao @Nat87a]

@deanwball Beijing is old (and rich) as well. However few people know that many downtown areas in the 2-tier cities of China are quite new and modern

[Linxiang Zhao @Nat87a]

lmao

[Milo Smith @mil000]

THE GREAT OPENCLAW MAC MINI SELLOFF HAS STARTED!!

[Eric Yeung 👍🚀🌕 @KingKong9888]

🔥China physical Gold buying streak:

PBOC has added gold for 18–19 consecutive months as of May 2026 (streak began ~November 2024 after a pause in mid-2024). This is confirmed across multiple outlets.

⭕️ Monthly additions (verified):

• Jan 2026: +40,000 oz → ~74.19
million oz total. Consistent with reports.

• Feb 2026: +30,000 oz → 74.22 million oz total. Confirmed (Reuters, Bloomberg, etc.).

• Mar 2026: +160,000 oz → 74.38 million oz total. Confirmed.

• Apr 2026: +260,000 oz (~8 tonnes) → 74.64 million oz / ~2,322 tonnes total. Confirmed by WGC, Reuters, and others (highest since Dec 2024; 18th straight month).

• May 2026: +320,000 oz (~9.95 tonnes) → ~74.96 million oz / ~2,331.5 tonnes total. This matches the post’s “BREAKING” claim and fits the pattern of larger buys during price dips (post notes May price range ~$4,480–4,730/oz).

🔶As of June 7, 2026, this aligns with the typical early-month PBOC/SAFE data release timing; no contradictions found in available sources.

[Linxiang Zhao @Nat87a]

抄底抄底抄底

[Bloomberg @business]

China’s central bank extended its gold-buying streak in May, adding to holdings as prices of the precious metal remained under pressure https://t.co/szMR1a5Kxa

[Christopher Kratovil @chris_kratovil]

It’s deeply odd to me that America is a far less 24/7 hour society today than it was 10, 20, or even 30 years ago. I vividly remember friends from the UK back in 1996 marveling at the fact that in the mid-sized Indiana town where I went college it was possible to buy groceries, clothing, a lawn mower, a snow blower, Lego sets, and bow hunting gear at 3 AM on any given Tuesday of the year. That was peak American Empire, and it’s long gone.

[0x鸣人 @LuBtc888]

黄仁勋：中国普通人的一生充满了痛苦和磨难，他们把这个叫做毅力和坚持。

一语落下，周围满堂哄笑...

[Xiuyu Li @sheriyuo]

AI 领域工业界对比学术界确实有断档领先，很多学术界可以水很多篇的 idea，可能一两年以前就在工业界是共识了，只是工业界根本不屑于发出来 or 内部保密不让发而已

paper 对于工业界来说，几乎等价于彰显自身水平的 tech report，而不是对学术界无数 funding / title / graduation 挂钩的硬指标，没听说过有很多厂把发 paper 数量当成是 kpi 吧（虽然真的有）

现在的 AI / CV 三大会也早就不是以前那个三大会了，论文质量和投稿量的变化大家也感受得到，只能说顶会依然在这样的灌水 decay / saturation 下相比于非顶会更有含金量罢了，paper pipeline 不就跟新东方厨师学校炒菜一样吗

对于高质量 paper 来说，最核心的肯定是算力和人类劳动的部分。而算力和人类劳动并没有变多，所以每篇 paper 平均下来的价值一定会变少，而那些坚持古法手作匠心传承的组怎么生存呢，被劣币驱逐？

大家读 ms / phd 总得毕业，毕业就需要 paper，现在扩招大环境只会有越来越多的 AI ms / phd，而不是 AI 方向的那些人自然也得被迫往 AI 转型(硬蹭)，那 paper 必然会越来越多

私以为大包并不会永远越来越多、越来越大，而 paper 注定会越来越多，有大包拿就别水 paper 了吧，给各位还要毕业的学生们一条出路

[Linxiang Zhao @Nat87a]

Once the bubble pops, Anthropic and OpenAI will also become Fannie Mae and Freddie Mac

[Mo @atmoio]

Once the bubble pops, Anthropic and OpenAI will become the Coinbase and Block of the AI world. Mundane companies that ship narrative wrappers on mundane bytes.

That the bubble will pop isn’t some apocalyptic doomsday prophecy. It’s not that complicated: AI is freakishly expensive to serve. If the returns on the other end are not justified, the bubble pops. And thus begins the decades long buildout to actually economically justifiable AI.

It’s amusing how resistant reality is to our fictions and fantasies. In the peak of the crypto bubble we thought reality was going to be transformed into financial liberty and democratization for all, and network states and decentralized reserve currencies. Coinbase stood to be a multi-trillion dollar company and is now just a mundane tech startup.

Today we spin similar narratives about the intellectual upheaval of AI, about the new democratization of intelligence and how everything will soon begin to orbit this new technology.

At the end, Anthropic and OpenAI will be mundane IT providers with an insanely grim research outlook to make AI economically sensible and useful, no different from Google’s position in trying to make quantum commercially viable.

Reality is, fortunately, pretty hardened against our delusions.

[Zephyr @zephyr_z9]

So Meta, Google, Anthropic, SpaceX, and OpenAI will raise around $350B-$400B from the public markets in the next 9-12 months
At this rate, Amazon & Microsoft will join the party too
We might even see $550B-$600B raised from the public markets

[Hereafter @idyllicmusing]

Soil fertility alone explains 34% of the differences in national IQ. Countries on the best soils (Mollisols, Andisols) average 10–15 IQ points higher than those on the wort soils (Oxisols, Ultisols), even before accounting for education or income.
This geographic pattern is visible in real populations. Japan and Taiwan sit on volcanic Andisols and consistently rank among the highest in global intelligence metrics due to their nutrient‑rich, high‑CEC soils. Meanwhile, much of equatorial Africa rests on Oxisols, soils so weathered and nutrient‑stripped that they produce chronically low micronutrient availability.
The further a soil’s pH drifts from 6.5, the more national IQ (MNIQ) declines. If your soil is too acidic (like much of the tropics), zinc and iron become unavailable. If your soil is too alkaline (like the Middle East), micronutrients get locked up too. The correlation between soil fertility and national IQ is r = 0.58, meaning soil alone accounts for 34% of the variation. Explaining 34% of a complex human trait is extremely high.

[Xiao Ma @maxiaoalex]

The real engine? A triangular dynamic between central constraints, local government ambition, and private capital's agility. Hundreds of cities—shut out of lucrative SOE joint ventures—formed underground alliances with private automakers to build their own industries.

[海伦子Hellen @peng_hellen]

我见过所有富有的人，他们的财富一开始都是挣来的，或者家族传承的，没有谁是光靠投资得来的，投资只是锦上添花滚雪球而已。

投资首先要有本金，没有稳定现金流的人基本都会手抖，生活压力都在投资上，市场一波动就容易有生理性恐惧，根本拿不住。

[Linxiang Zhao @Nat87a]

20260606
比特币本质上属于一个美元体系的资产，大多数人购买比特币之前都要先拿到USDT/USDC，这是美元稳定币，相当于数字美元。
和美股美债相似，一个人若想买到美股、美债，ta要先拿到美元，用别的货币是买不到美股、美债的。比特币现在也基本成这样了，要先拿到美元才能大规模参与这个市场。(2/n)

[Linxiang Zhao @Nat87a]

20260605 开个thread准备随意记录一下对比特币以及其他各个资产的思考

这样以后能更清晰看到自己思路的进展

现在BTC 60K

占个位

(1/n)

[Linxiang Zhao @Nat87a]

🤔感觉是这样的

[马东锡 NLP @dongxi_nlp]

很多推友私信问我，为啥最近不咋分享 AI论文了。

其实很明显，Auto Research + Coding Agent 之后，很多论文从idea到实验和写作，基本看不见碳基的影子了。而 MAI 的论文基本说明白，模型迭代非常的管道化和工业化了，何况一般实验室的论文制造。

最重要的，作为碳基生命，去分析硅基idea，这件事开始觉得奇怪。当然，如果遇到论文中有碳基感很强的idea，还是会大力分享的。

[Linxiang Zhao @Nat87a]

@Linahuaa Based

[Linxiang Zhao @Nat87a]

卧槽涨停了???

[Serenity @aleabitoreddit]

专门写给我的中文读者：

绿的谐波（LeaderDrive，688017，577.3亿人民币）是我在布局人形机器人赛道时最青睐的中国上市标的。

他们的业务涵盖：

谐波减速器（据称占有超过60%的国内市场份额，以及1800多家全球客户）

人形机器人旋转关节减速器

直线执行器

电机/关节，以及许多其他核心零部件。

并且他们正在进军行星滚柱丝杠（planetary roller screws）领域。

目前占据着绝对的主导地位：超过60%的国内市场份额以及1800多家全球客户。

主要客户包括优傲机器人（Universal Robots）、优必选（UBTech）、智元机器人（Agibot），而特斯拉（Tesla）、Figure 以及绝大多数其他人形机器人研发商都很可能是他们的潜在客户。

如果你看一下这些零部件的 BOM（物料清单），粗略估计它们占到了每台研发出的人形机器人的 4-15%，随着他们进军更多细分领域，这个比例可能还会更高。

我认为没有人能对这类公司进行精确的财务建模，其前景更多取决于未来几年内会有多少人形机器人实现量产。

这其实是对一家能够在每台生产出的人形机器人中占据可观份额的公司，进行的一种方向性做多（directional long）。

在可扩展的规模化量产方面，中国显然是该领域的领导者，许多西方玩家无法将成本降到绿的谐波那样的水平。

随着我们向实体人工智能（physical AI）的规模化迈进，我极其看好机器人这个赛道。

[Linxiang Zhao @Nat87a]

@Duduook 你值得被爱被喜欢的

[Nawab @WDLD0712]

Though much of real estate asset appreciation globally is undeniably due to hyperfinancialisation and unproductive money printing, it is entirely possible to have real estate asset appreciation without hyperfinancialisation.

When an underdeveloped area undergoes infrastructural and manufacturing investment, the appreciation in real estate prices captures the future potential of economic activity in that area; it is directly linked to productivity.

Unlocking this real estate appreciation mechanism is actually how many East Asian countries financed the building and operation of high-speed rail and metro projects, which are otherwise money-pits from a purely accounting POV.