2026.40: Dots and Question Marks
2026.40: Dots and Question Marks The best Stratechery content from the week of September 28, 2026, including Meta's focus, what OpenAI is doing, and Mao and NBA Media Day. https://t.co/4gi1glx7RT
精选
X（推特）上的财经、科技与宏观账号。按时间排，不论是否入选精选。
X·财经博主AI 评分2020 X·财经博主AI 评分1515 I think you are translating the SV experience - employees, management, etc.
I think you are translating the SV experience - employees, management, etc. Meet many leaders from a variety of cap sizes that do not really play around with AI at all despite communicating that their company is heavily investing into it Average person I know has never vibe 引用 @krishnanrohit：Eve
X·财经博主AI 评分2828 OpenAI and Anthropic are the two largest customers of compute and rumors are they both run 30-40%…
OpenAI and Anthropic are the two largest customers of compute and rumors are they both run 30-40% inference so training definitionally has to be bigger than inference today? 引用 @jessiedong_：inference will be the biggest market. it's already bigger than training and I bet this is going to change how
X·财经博主AI 评分2121 Reason people aren’t controlling the full stack is the capital intensity
Reason people aren’t controlling the full stack is the capital intensity Hypers will and can underwire the power, shells, chips, etc Others like neos and managed inference provides can barely execute the IT before trying to move upstream This is a capital problem vs lack of 引用 @kellyjgreer：lot
X·财经博主AI 评分1818 I fully agree here but I think one thing that’s under appreciated is people talking about the…
I fully agree here but I think one thing that’s under appreciated is people talking about the duration mismatch Lot of forward credit for TAM penetration x steady state economics when in reality extremely early in usage and adoption That means compute demand going up but it 引用 @nicbstme：This is m
X·财经博主AI 评分3232 Brother, what will your ARR do
Brother, what will your ARR do That answers your capex question 引用 @firesidealpha：Anthropic's Sholto Douglas says AI capex could rise from $1 trillion this year to $4 trillion in 2028 and begin doubling humanity’s GDP by the early 2030s “Over the course of the last four or five years, we’ve been
X·财经博主AI 评分99 You literally could open this app and see 100 of these
You literally could open this app and see 100 of these “The market is asleep at the wheel. I’ve found the next HUGE AI bottleneck. Follow me for more posts like this” *stock up 5,000% in the last 6 months already* **half the time it was literally just pitching nvidia again** 引用 @ParadisLabs：I'm p
X·财经博主AI 评分1717 Citadel is impressive man, consistently putting up #s
Citadel is impressive man, consistently putting up #s 引用 @nishantkumar07：We have some initial hedge fund returns for September. Looks like machines scored a win over humans. Traders faced a volatility month as inflation concerns pushed bond yields to multidecade highs and hit fixed-income wagers. h
X·财经博主AI 评分2828 One notable trend that goes against my priors is seeing a lot of long time veterans from the hypers…
One notable trend that goes against my priors is seeing a lot of long time veterans from the hypers move to Neo clouds That doesn’t mean it’s bulletproof or anything but people with many years and in senior positions of the companies with the most flexibility are moving to
X·财经博主AI 评分66 Markets Weekly October 3, 2026
Markets Weekly October 3, 2026 Bonds behave badly Le spread Anthropic IPO https://t.co/z4A7UCjuDI
X·财经博主AI 评分33 One fun thing about reading old finance books is that they’ll talk about how shocking some number…
One fun thing about reading old finance books is that they’ll talk about how shocking some number was when we’ve since blown past it. For example, here’s a book that came out on Tuesday: https://t.co/1xy8FDlqtf
X·宏观AI 评分3131 This ho-hum labor market report doesn't really change the story for the Fed, whose senior officials…
This ho-hum labor market report doesn't really change the story for the Fed, whose senior officials had gone out of their way this week to signal that an October rate hike probably wasn't their base-case. The biggest development was what it didn't show: few signs from wages or https://t.co/6GY9Ud2Q
X·宏观AI 评分1313 Adam Tooze's concept of a profit dollar works for sovereign investors too:
Adam Tooze's concept of a profit dollar works for sovereign investors too: "Mr Setser estimates that nearly all the new funds and banks are more exposed to America than their countries’ central banks are" 2/3 https://t.co/kuLDaYydUT
X·宏观AI 评分1111 In a shocking turn of events, this time I completely agree with David and tweeted the math broadly.
In a shocking turn of events, this time I completely agree with David and tweeted the math broadly. 引用 @DavidZervosDC：In the @ftopinion article below, @martinwolf_ claims that the average maturity of US Treasury issuance decreased from 6.1 years in 2024 to 2.5 years in 2026. He should check his mat
X·宏观AI 评分1818 Our latest #OnInvesting podcast episode has dropped on which and I chat about bond yields, the…
Our latest #OnInvesting podcast episode has dropped on which @CollinMartinCS and I chat about bond yields, the yield curve, and weak stock market breadth; and I sit down with the always-insightful Eric Liu from @vandaresearch, who discusses Vanda’s latest upgrade to their view
X·宏观AI 评分22 FYI: Some of the links to this morning's conversation on Yahoo Finance.
FYI: Some of the links to this morning's conversation on Yahoo Finance. https://t.co/x6VvRYRyAc https://t.co/JCcakw6t1G https://t.co/IUtlA0AtlF #economy #markets #jobs #federalreserve #employment
X·宏观AI 评分11 They could be compatible if it turned out that the best/fastest way to build a successful business…
They could be compatible if it turned out that the best/fastest way to build a successful business product is to assemble a bunch of people who share the same idiosyncratic or interests. 引用 @yudhanjaya：I genuinely don't get Anthropic. It seems like every other day they're loudly puzzling over the e
X·宏观AI 评分2222 I very very very strongly disagree and think that AI adoption in the enterprise beyond coding is…
I very very very strongly disagree and think that AI adoption in the enterprise beyond coding is basically at the starting line 引用 @krishnanrohit：Everyone uses AI by now. Senior leaders are vibecoding apps. They've adopted claude. They're finding workflows to automate. Folks have their own claws. A
X·宏观AI 评分2222 So the AK capital growth loop is the thing that makes this work. However it ignores physics limits…
So the AK capital growth loop is the thing that makes this work. However it ignores physics limits on energy and materials and Baumols Cost disease. But perhaps most importantly it ignores the sources and constraints on capital due to cash flow timing deficits Minsky squeeze 引用 @firesidealpha：
X·宏观AI 评分3636 A manufacturing surplus somewhere north of 4 trillion, a customs surplus north of $2.5 trillion --
A manufacturing surplus somewhere north of 4 trillion, a customs surplus north of $2.5 trillion -- Basically, China's current trajectory, with sustained 10% growth in export volumes and a more modest but still pronounced increase in the surplus in manufactures https://t.co/jtfRvKoQCG 引用 @Balding
X·财经博主AI 评分2222 Bill Ackman’s post about inflation took me back to the late 1970s, when similar concerns were…
Bill Ackman’s post about inflation took me back to the late 1970s, when similar concerns were everywhere. What followed was stronger growth and falling inflation, helped by the PC and software revolution. We believe today’s innovation platforms could have an even greater impact. 引用 @ARKInvest：“It's
X·财经博主AI 评分1616 蠢货天天再谈加息影响股市。
蠢货天天再谈加息影响股市。 我的暴论：你加到10%，股市还是创新高。 美国经济的动力不再是消费而是AI基建。现在利润搞到比挖金子还高，你加息怎么压制它？
X·财经博主AI 评分1616 我对评论区一天涨跌就喊某某不行的行为非常嗤之以鼻，炒股从来不是每天猜涨跌，而是一个长期的过程，并且上涨路上永远是伴随下跌的，你看的短，跌了就嘲笑某某垃圾，只能暴露你的水平还处在🐸🐸🐸水平…今天跌了，过…
我对评论区一天涨跌就喊某某不行的行为非常嗤之以鼻，炒股从来不是每天猜涨跌，而是一个长期的过程，并且上涨路上永远是伴随下跌的，你看的短，跌了就嘲笑某某垃圾，只能暴露你的水平还处在🐸🐸🐸水平…今天跌了，过两天涨了，那到底是垃圾还是不垃圾呢？以这种思路评价股票的话，那你会发现每一只股票
X·财经博主AI 评分1717 最近总有人回我说存储不行，涨不过 cpu，我说实话真的挺无语的，事实的结果可能非常反直觉。mu 和 intc 目前距离新高都只有 16%，如果你觉得英特尔更猛，可结果两人的位置一摸一样。
最近总有人回我说存储不行，涨不过 cpu，我说实话真的挺无语的，事实的结果可能非常反直觉。mu 和 intc 目前距离新高都只有 16%，如果你觉得英特尔更猛，可结果两人的位置一摸一样。 也许你又会说，那 intc 从底部反弹的多啊？可事实是，intc 从 7 月 29 日反弹了 49%，美光则是
X·财经博主AI 评分1717 我五月底关闭订阅，预言科技股会大跌，我的预言准不准，当时通知订阅学员买入债券、BIL避险，准不准？我们的学员幸不幸福？
我五月底关闭订阅，预言科技股会大跌，我的预言准不准，当时通知订阅学员买入债券、BIL避险，准不准？我们的学员幸不幸福？ 引用 @cnfinancewatch：我看空六月以来一直 看空科技，预言龙头跌50%，杂毛跌两个50%， 只有在八月看多黄金（已经清仓）， 现在不悲观：老登新登同步砸盘，最后一跌就在眼前。刚刚588000破新低了。恐慌盘出来，最后一砸，挖坑就会完毕。目前大资金还有8000亿需要出掉。 可能机会就在这两天，大家订阅吧：
X·财经博主AI 评分99 达利雷生我吃了3次能安稳的睡8h以上后面换了缓释褪黑素，现在不用任何药物能做到7h左右睡眠现在缓释褪黑素也不用了除非倒时差。
达利雷生我吃了3次能安稳的睡8h以上后面换了缓释褪黑素，现在不用任何药物能做到7h左右睡眠现在缓释褪黑素也不用了除非倒时差。 我现在随身携带口腔喷雾褪黑素，旅游或者换酒店不习惯喷，中途醒来睡不着了喷，能睡着比啥都强。 如果是因为仓位原因，单子太大压力或者浮盈太高兴奋这个大家都会出现。 https://t.co/vEklVvoa1i 引用 @WallStreet0Name：你们有没有吃过达利雷生和达卫可这两个安眠药？ 我晚上可以很容易入睡，长时间半夜四五点经常有大行情，一醒就睡不着了，我看别人说这两个安眠药主要是维持睡眠，中间就不会醒了，你们谁吃过啊？
X·财经博主AI 评分1212 如何让民营企业家纳入体制管理，成为体制的支持者？
如何让民营企业家纳入体制管理，成为体制的支持者？ 这是最近十来年管理层非常成功的地方。其他，都很次要。 工商联——政协——人大…… 制度安排是非常有效的。
X·财经博主AI 评分44 商务局：能带来合作机会的，要去。
商务局：能带来合作机会的，要去。
X·财经博主AI 评分33 和解局：冤家宜解不宜结，要去。
和解局：冤家宜解不宜结，要去。
X·财经博主AI 评分1818 【A股盘前】 2026年10月3日
【A股盘前】 2026年10月3日 ======================================== 数据窗口：9月28日~9月30日（近3个交易日） 一、今日结论 环境分：38分，只观察，不给买点。
X·财经博主AI 评分88 A股大跌四个月，我们都学员账户大涨四个月。
A股大跌四个月，我们都学员账户大涨四个月。 订阅跟踪检阅：https://t.co/0hbYvJoY65 引用 @cnfinancewatch：$### 收盘于历史新高。 祝福我的订阅者。订阅我，我就是财富密码，我们下周可能动员上杠杆： https://t.co/0hbYvJoY65
X·财经博主AI 评分1515 14亿中国人，仅有90-95万境外证券账户。2000-2400亿。
14亿中国人，仅有90-95万境外证券账户。2000-2400亿。 中信数据！ 惊掉下巴，这点儿鸡肋也要？
X·财经博主AI 评分2121 8/8 Why push back on debt levels as a driver of interest rates?
8/8 Why push back on debt levels as a driver of interest rates? Because way too many think debt is the MAIN driver of higher yields, interest costs are the burden, and the solution is financial repression such as yield curve control. This would be disastrous. If yields are 引用 @biancoresearch：1
X·财经博主AI 评分2020 1/8 Is Too Much Debt the Reason Bond Yields are Heading Higher?
1/8 Is Too Much Debt the Reason Bond Yields are Heading Higher? -- My argument for why this case is overstated. (It's a problem, but a secondary one.) 🧵 https://t.co/zBzpq4Mada
X·财经博主AI 评分99 对中国政治地理懂的人不多。
对中国政治地理懂的人不多。 现代中国，政治地理，东南，有自己的出海通道，离心是比较容易的，但是，因为人种原因，有北向的精神需求，再加上那个岛，岛变成了它的外围锁，反而外向没了出路，反之岛连不上珠三角的势和力，没有什么离心能力。
X·财经博主AI 评分1111 目前重点： （AI算力&数据中心）：IREN、CIFR、WULF、CORZ, NBIS, CRWV, SMCI, DELL
目前重点： （AI算力&数据中心）：IREN、CIFR、WULF、CORZ, NBIS, CRWV, SMCI, DELL （电源/电网/温控）：VRT、ETN、GEV、CEG, SMR, OKLO, VST
X·财经博主AI 评分88 说个暴论： 分析国内经济和企业，多数人看不懂。
说个暴论： 分析国内经济和企业，多数人看不懂。 那些企业看起来是制造四轮盒子的，其实是制造特种的，拼起来就是特种。你看得出来？
X·财经博主AI 评分1717 $MBX（MBX Biosciences）Guggenheim由88.0升至150.0上调70.5%，距现价58.51空间156.4%；
$MBX（MBX Biosciences）Guggenheim由88.0升至150.0上调70.5%，距现价58.51空间156.4%；
X·财经博主AI 评分1717 三只股票 $MU：评分86：RSI14 59.3且MACD柱8.115金叉，支撑902.6压力1108.72，目标价上调至1596.52美元为重要加分。
三只股票 $MU：评分86：RSI14 59.3且MACD柱8.115金叉，支撑902.6压力1108.72，目标价上调至1596.52美元为重要加分。 $ORCL：评分78：RSI14 48.3且MACD柱-1.738死叉，支撑131.58压力170.7，内部人买入3483750美元与云订单积压支撑。 $AMZN：评分74：RSI14
X·财经博主AI 评分77 港股只要买汇丰就行了。 其他不用碰！
港股只要买汇丰就行了。 其他不用碰！ https://t.co/lZ9u9moPFg