Appearance
📅 20260714 推文存档
📊 共 45 条推文
@xiaomustock
2026-07-14 23:41:37
请发给你的韩国朋友。 AI能干的事情过去2年， 慢慢从一维数据到二维数据再到未来的三维数据进化，也就是文字，图片，三维视频。 这经历的时间也才短短2年多，24年到现在26年。而存储扩产在2028年，也就是2年后，AI会发展成什么样，这些傻逼预测分析师硅谷投资客都没有考虑进去，他们只是基于现在的AI状态来分析存储未来尤其是存储2028年的未来做出当下的判断，他们都是傻逼。 如果物理AI能突破，处理的资料都是视频资料，不是文字，存储数据内容从一维到三维，存储的需求量是天量。人们现在对存储的假设是基于现有AI水平来测算。 但这些傻逼都没有考虑进去2年后AI会发展成什么样， 现在AI视频已经可以以假乱真， 俩年后更是不敢想象会进化到啥程度， 如果AI能进化到理解现实物理世界，那它处理的数据全是视频数据， 存储需求的数量级可能是百倍于现在， 要坚信你的存储。 试想一下AI俩年后会发展成什么样，回想过去俩年AI的进化速度和对存储需求的增长速度，不要再扔掉你们宝贵的 #海力士 $SKHYX 了。 AI用量会在俩年后暴增数倍，想象你2年后使用频率，从现在一天3-5次会不会暴增到30-50次？ AI使用群体数量俩年后会不会从1B暴增到几B？（使用群体不限于人，包括汽车，包括手机，包括机器人，甚至智能体本身…） AI现在人70%的使用是文字token，俩年后会不会进化到视频影像物理三维视频token输入输出？（这里存储数据量差别千倍万倍。） 这里面的需求差别2年后会是现在的多少倍？十倍还是百倍？ 那俩年后的存储产量增量是几倍？翻倍都撑死了吧？这些傻逼预测师永远不会这么给你算，只会基于现在AI状态误导你。 拿出你们290万韩元梭哈的勇气， 190万韩元都不敢买，真去死了算了。
2026-07-14 21:08:59
你们此时此刻是不是百病全消，神清气爽 保守估计自己又是全银河系最牛逼的人？
2026-07-14 13:33:39
韩国看来是要从向下熔断 变成接下来向上熔断
2026-07-14 13:29:48
人亏钱就跟得病了一样浑身难受，赚钱了就跟吃了灵丹妙药，但大多数人的财富多少并不会影响到衣食住行，这他娘的是为什么
2026-07-14 13:03:56
起飞了，75总能反弹到吧
2026-07-14 12:41:29
51.5抄到大底了艹她妈的， 南海2x海力士午盘开了估计也是爆拉。 哪有小孩天天哭。
2026-07-14 12:27:14
后面只希望存储反弹一波整体稳住撸一发， 存储稳住别的概念才有机会。 找未来还有10倍以上营收增长预期的标， 2x这种短撸一下没毛病，长拿命就没了。
2026-07-14 11:49:13
克服恐惧的最好办法就是面对恐惧， 海力士不行，整个AI产业链都在一起崩。 刚抄了南海2x，成本51.5。 甭管什么Ai几把概念，都是和存储一荣俱荣
2026-07-14 10:46:46
保持有一定数量的散户涨跌都赚不到钱对国家生产活动很重要，各行各业需要他们专心干活，不伤国本。
@elonmusk
2026-07-14 23:00:50
RT by @elonmusk: Used @grok to remove the background😉🤗 @XFashion
2026-07-14 22:56:19
Ancient times
2026-07-14 22:52:02
True
2026-07-14 22:48:02
RT by @elonmusk: SpaceX just launched two Falcon 9 rockets carrying 56 Starlink satellites in less than eight hours SpaceX is moving at light speed right now First: • 27 Starlink satellites from Vandenberg Then, just 7 hours and 42 minutes later: • 29 more Starlink satellites from Florida Two coasts, two rockets, 56 satellites, less than eight hours For most space programs, one launch is a major national achievement SpaceX launched twice in one day and was already preparing for the next mission SpaceX is basically running an orbital assembly line
2026-07-14 21:29:37
RT by @elonmusk: Starlink will provide high-speed, low-latency connectivity onboard @FlyFrontier, transforming every flight into a fully connected experience 🛰️✈️
2026-07-14 20:48:04
RT by @elonmusk: Grok is the Flow LLM. Grok 4.5’s biggest advantage is its speed. It’s smart enough to be comparable to the other models on most things. But that speed allows you to make little tweaks to your system so that the output is PERFECT. With the other models, it takes FOREVER to do this, and you constantly end up outside of a flow state, which makes it way way way harder to do this. It’s now my preferred model.
2026-07-14 13:58:35
Grok 4.5 reaches #1 position on Long-Horizon Terminal-Bench
2026-07-14 13:44:50
The woke mob had turned Twitter into Wormtongue to the World
2026-07-14 12:10:23
RT by @elonmusk: Just a reminder that deepseek v3 came out 18 months ago and was considered revolutionary at the time but is basically unusable today There was a fierce debate at the time about vibe coding and the argument was that LLMs can never create anything new because they are limited by their training data Those engineers were partly right and vibe coding was a real nightmare, I cannot believe what we used to put ourselves through with Sonnet 3.5, but we also knew real new things could be made and that the detractors were wrong I guess us non-engineers saw it most clearly (I would like to think so at least) because we were astonished at the new things we could do without the programming background, and we had nothing to lose and everything to gain in our enthusiasm Then came Opus 4.5 earlier this year which changed everything Suddenly real production code became possible with so much less friction and headache Everyone complained endlessly about models being nerfed or quantized but there was a steady march of progress from 4.5 to 4.8 Now a completely new crop of models is coming out that is not quite a 4.5 moment but something close We are not only getting more polish but things are becoming a little freaky, entire isolated domains become possible to combine with small teams or even just one person into new applications that would have required large infusions of time and capital in the past My wife back in 2021 told me about AI but I was in healthcare, I thought she was being a little nutty, and she talked like something from a science fiction film was coming and she got involved in it early on She never stops letting me know that she was right, and she was The next year is going to be wild, the world is truly going to change over the next few years, the scale of disruption will be a combination of astonishing and awesome, but also catastrophic There are many amazing things ahead and extraordinary challenges and opportunities We are living inside one of one of the biggest revolutions in human history
2026-07-14 11:59:16
RT by @elonmusk: Falcon 9 launches 27 @Starlink satellites from California
2026-07-14 06:32:09
True
2026-07-14 03:28:06
RT by @elonmusk: SpaceX was founded on one powerful belief: A future where humanity is out exploring the stars is fundamentally more exciting than one where we are not Every Falcon launch, every Starship test, and every step toward Mars is built around that belief The goal is not just to reach space It is to build a future where humanity becomes a multiplanetary civilization
2026-07-14 03:16:51
RT by @elonmusk: We're rolling out a small tweak to boost visibility of your posts to your mutuals (people who you follow back). We noticed this data was missing from the algo and it made your friends appear less in your replies. This resulted in the reply section feeling more like a battleground with people you don't recognize. This should also help clusters form around interests more easily, which many people have asked for.
2026-07-14 02:45:40
Progress
2026-07-14 02:45:00
True. As a precautionary measure, all user data that was uploaded to SpaceXAI before now will be completely and utterly deleted. Zero anything whatsoever will remain.
2026-07-14 02:38:16
SpaceX policy regarding data retention. It is actually helpful for debugging issues if we can retain some amount of data, so allowing this would be appreciated, but your privacy settings are always respected.
2026-07-14 02:35:27
RT by @elonmusk: Their zero data retention policy fine print says they still retain data? Am I reading this right?
2026-07-14 01:53:11
RT by @elonmusk: The default in Grok Build is off. You can also run /privacy and any synced data is deleted. ZDR is respected for both headless and interactive use as well.
2026-07-14 01:50:23
RT by @elonmusk: That's not the case. The CLI doesn't retain data under ZDR (including headless/non-headless modes), and has since launch. Similarly, /privacy in the CLI will disable or reenable any data retention. For some features, including remote session listing, cloud agents, and handoff, you will need to enable data syncing.
2026-07-14 01:42:01
RT by @elonmusk: Right-wing wave in Europe. France 🇫🇷: Marine Le Pen now leading polls to be next President. Austria 🇦🇹: the Freedom Party is leading the polls massively at almost 40% Germany 🇩🇪: The AfD party now consistently number ONE in polls. Norway 🇳🇴: The Libertarian FrP party leading the polls by a wide margin. Something is changing in Europe.
2026-07-14 01:34:53
RT by @elonmusk: Crazy what you can build with Grok 4.5 and Grok Build in minutes.
@aleabitoreddit
2026-07-14 22:57:21
$NBIS signs $1B+ compute agreement with Reflection AI, for GB300 access through 2029. Reflection also signed a multi-billion dollar agreement with $SPCX earlier. Interesting to say the least, seeing Nebius drop -5% off the news today. Also... counterparty to get this done kinda reminds me of OpenAI, where they might not have the funds to actually execute on these LTAs yet compared to $META or $MSFT. But generally positive long term developments, customer diversification was one of the core strengths of Nebius.
2026-07-14 22:35:50
Two different capacity expansion announcements today across $TSEM and $AAOI. Tower Semi, supported by the Japanese government, announced expansion of its 300mm Silicon Photonics (SiPho), Silicon Germanium (SiGe), and advanced packaging capabilities. Targeting $1.2B net profit, and $3.6B revenue in 2028 (with a ~$28B MC, that's ~23.3 2028 forward p/e). AAOI announced the commencement of its buildout, roughly 400,000 square feet of manufacturing to make 800g/1.6T production go brrr. The overall expansion is also supported by the local city government in Texas. (they also $20.85M worth of subsidies from the State earlier). This one isn't exactly as materially new as Tower, since AAOI originally announced these plans before. Just commencement of it. Regardless, glad to see both of these companies in their Anime training arc phase as they power up.
2026-07-14 09:51:24
So Jensen himself went out to do damage control with $NVDA delays reports apparently: 1. Jensen Huang said no delays: - 800V and rack-to-rack optical interconnects is progressing fully in line with the original plan with no material delay. - Debunked the rumor that the next-generation flagship, Rubin Ultra, would be delayed until 2028 and said it would be on track to ship next year - Rack system optimizations, like replacing Kyber with a new design are architectural optimizations, not impacting delivery timelines. 2. On ASIC competition: no loss of market share. - Talked about how a leading frontier lab that previously relied heavily on internal ASICs has shifted nearly 50% of its compute load to Nvidia 3. On over-reliance on hyperscalers. - AI frontier labs currently contributing to ~20% of demand. - Traditional cloud providers, Nvidia provides Vera CPUs and high-speed networking - Sovereign AI
2026-07-14 07:17:46
TLDR of Innolight investor relations takeaways: 1. "Overall, 1.6T market demand has not contracted; instead, 800G demand has increased significantly compared with previous expectations" Prob most important takeaway as a whole was 800G demand revision (also longer tail demand). Which is a bullish read through on US transceiver makers like $AAOI, $COHR, $LITE. Lot of new customers like neoclouds, AI model companies, contributing to overall demand rather than just hyperscalers, diversification always a bonus. 2. Innolight said the shortage covers the module supply chain broadly including: - Optical chips. - Electrical chips. - PCBs. - Other module materials From last ER, I think they singled out EMLs and CW optical chips as the most constrained. So Innolight's bottleneck list mention broadened since then. They expect some of the component availability to improve gradually from the second half of 2026 through the first half of 2027. Think a lot of this is already known from earlier though. But just some confirmation + easing timelines (EML is extremely bottlenecked, same with CW, this is probably talking about other components). 3. Innolight said module-production equipment is not the constraint. Equipment lead times remain relatively short. So this isn't really a bottleneck compared to others. 4. The overall proportion of silicon photonics continues to trend upward. Last year it was mainly 800G. This year, some 800G customers are further increasing their silicon-photonics proportion. 1.6T also added some new customers Positive for SiPH penetration eg. $SIVE / $JBL, since this shifts away from EML toward CW. Basically: main surprising takeaway is just 800g demand go brrr. Apart from that just reaffirming bottlenecks/timelines/market speculation.
2026-07-14 03:20:46
Morgan Stanley note on CPO today. Key participants include: - $SIVE, $COHR, and $LITE in laser supply - Broadcom and Nvidia in switch platforms - Lightmatter, Ayar Labs, Marvell/Celestial, and POET in optical engines and photonics, - $TSM, $GFS, and $TSEM in silicon photonics foundry capacity I’ve covered all of these before, like Tower Semi. But I’m especially happy that Morgan Stanley validated my research that $SIVE is one of the critical global players in CPO. A small $1.5B laser company next to your two leading $60B+ companies… If you also synthesize Rosenblatt Securities recent note that China laser suppliers were quite far from having CPO lasers. This kinda magnifies importance of the three Western leaders of that laser chokepoint.
@dianzhuXS
2026-07-14 19:57:39
雷齋月的幾件事 1.繼續丙午八字免費收徒第八期 2.開放每年僅此一次的「念經團」patreon新會員徵集（非普通會員，僅限初高超會員等級，三個月考核期通過後，享受還受生債法會福利） 3.雷祖聖誕準備接雷水泡避邪紅繩（雷齋月過後開始免費結緣贈送） 以上活動僅限中國大陸地區以外參加，暫不接待牆內報名 4.雷齋月期間為高市首相燃燈祈福
2026-07-14 16:52:42
颱風巴威風圈邊緣輸送 導致牆國瀋陽強降水 整個一樓車庫都廢了 超過20個小時停水停電 😱😱😱
2026-07-14 07:02:28
@Mimiwftt
2026-07-14 17:40:45
风物长宜放眼量，不畏浮云遮望眼
2026-07-14 16:32:01
【上证指数这里还不能确认见底，耐心等待市场验证】 今天上证指数缩量反弹，市场情绪有所修复，但目前仍不能确认已经见底，更不能轻易定义为趋势反转。 从日K走势来看，指数尚未站上5日均线，30日均线依然下行，短中期均线尚未形成多头排列，说明中期趋势仍有修复需求。 真正的底部，不是靠一根阳线确认，而是要看资金是否持续进场。参考4月8日，当时市场走出了一根放量标志性阳线，随后成交量持续配合，行情才逐步确立。因此，目前更倾向于将这里定义为反弹阶段，而不是已经完成见底。 接下来重点关注四个信号： 是否出现放量突破的标志性阳线； 成交量能否持续放大； 5日、10日、20日均线能否逐步转强； 指数能否形成”高点更高、低点更高”的上升结构。 这里存在构筑底部的可能，但证据还不充分。交易上与其猜底，不如等底。只有市场给出明确的底部信号，再提高仓位，成功率会更高。
2026-07-14 15:32:42
这两天看到大家在评论区的留言，真的特别欣慰。这么多志同道合的人，也让我觉得，这一路的坚持和付出都是值得的。 前两天去爬山，到了山顶才发现，几乎所有人都在排队等着拍照📷，想留下一张照片，证明自己真的爬到了山顶🏔️。 像我这种懒人，就拍了一张大家排队的合照，心想：有这张照片，也算证明我来过了。😂 后来回头看这张照片，我忽然觉得，它好像也映照着我现在的心态。 比起一定要站在山顶最显眼的位置，我更愿意站在旁边，看着你们一个个登上属于自己的山顶，见证你们的成长、收获和喜悦。对我来说，能陪着大家走过这段路，本身就是一件很有意义的事。 所以，这两天看到评论区里大家真诚的分享和反馈，我更加确定，我的每一份付出，都没有白费，也真的很值得。
2026-07-14 12:41:39
我在你们的评论区里，看到了我想要的答案。 交易的终点，从来都不只是赚钱，而是修心。 真正的问题不是账户里有多少钱，而是你的内心，究竟被什么所支配。 是成为金钱的奴隶，被每一次涨跌牵动情绪，被每一次波动左右人生；还是在市场中看透规律、看清人性，学会与欲望和恐惧相处，把交易当作一场修行。 当你不再执着于每一笔盈亏，不再把市场当作证明自己的地方，你会发现，交易只是人生的一部分，而不是人生的全部。 财富可以增长，也可以缩水；行情会来，也会离开。唯有心境的成长，才是真正属于自己的财富。 交易最终教会我们的，不只是如何赚钱，更是如何过好这一生。 看透规律，敬畏市场，珍惜当下，珍惜身边的人，过好每一天。这，或许才是交易真正的意义。
2026-07-14 06:47:12
【休息一段时间，我需要思考一些问题】 最近我一直在反思一件事：我在 X 上持续输出，到底给我自己、也给大家带来了什么？ 最开始，其实很简单。输出是我学习的动力，也是我整理思考、验证认知的方式。每一次分享，都是在逼着自己成长。 但慢慢地，这件事开始发生变化。 我发现自己越来越容易被不同人的情绪影响。会因为一句认可而开心，也会因为一句质疑而反复审视自己；会因为大家的期待而调整自己的表达，也会因为市场的波动去顾虑别人的感受。 到后来，我甚至发现，不只是我，好像所有人的关系都在被盘面绑架。 看好的方向大涨，大家普天同庆，仿佛所有判断都得到了证明；看好的标的大跌，责骂、质疑、失望接踵而至。被信任、被怀疑、被追捧、被嘲笑，这些身份似乎都随着K线不断切换。 可实际上，我始终还是那个我。 我的逻辑没有因为一天的涨跌而改变，我的认知也不会因为一根阳线或阴线就发生本质上的变化。只是我一直在承接着市场波动带来的各种情绪——有正向的，也有负面的，而这些情绪很多时候并不真正属于我。 于是我开始问自己： 我这样持续输出，到底是不是一件正确的事情？ 它真正的意义是什么？ 我是为了什么而做？ 如果有一天停止输出，我失去的又是什么？ 而你们，因为我的分享得到的是什么？失去的又是什么？ 如果我们衡量一切的标准，只剩下短期的涨跌，那么无论是输出的人，还是看内容的人，最终都会成为盘面的情绪载体，而不是独立思考的人。 我越来越觉得，真正有价值的输出，不应该是带着大家追逐每一次涨跌，而是帮助彼此建立更完整的认知、更稳定的情绪，以及面对市场时独立思考的能力。 市场每天都会给出答案，但答案未必代表认知；涨跌会不断发生，但真正值得沉淀的，应该是那些穿越涨跌之后依然成立的思考。 或许，这才是我继续输出真正想追求的意义。
2026-07-14 00:20:48
X 上，又何尝不是一处修心悟道的道场。 这里汇聚着世间百态，真善美与假恶丑并存，贪、嗔、痴交织，赞誉与诋毁同行，掌声与质疑相伴。看似是在观人，实则是在观己。 《道德经》有言：“致虚极，守静笃。”真正的修行，不是远离喧嚣，而是在喧嚣中守住内心的清静；不是逃避纷争，而是在纷争中不失本心。 世间真正束缚人的，从来不是外物，而是内心的执。执于名，则为名所困；执于利，则为利所役；执于虚荣，则心随赞毁而起伏；执于胜负，则神随得失而动摇。凡有所执，皆有所缚。 X 不过是一面镜子。有人照见别人的过失，有人照见自己的欲念；有人沉迷争辩，有人借境炼心。境无善恶，心有分别。 道家讲“无为而无不为”，并非什么都不做，而是不妄为、不强为、不争一时之长短，不困于他人的评价，不耗于无谓的较量。正所谓“夫唯不争，故天下莫能与之争。” 真正的修行，不是赢得每一次争论，而是输掉内心的执念；不是让所有人认同自己，而是不再需要所有人的认同。心若如水，万物皆可映照，却不为万物所留；心若如谷，虚而能容，静而能生。 当名利可以放下，虚荣可以淡去，胜负可以一笑置之，便渐渐懂得“少私寡欲，返璞归真”的意义。修道的终点，不是成为别人眼中的高人，而是回归本真，与道同行，任万物流转，而心常清静自在。
@JonathanDi3614
2026-07-14 15:29:44
今天午后开盘，市场突然拉起——尤其前期跌幅较大的科技类个股，集体大幅反弹。 今天我空仓，明天找机会 怎么回事？表面看是超跌修复，但背后有三把悬着的刀： 第一，韩国股市最近不是涨到熔断，就是跌到熔断。韩高层限制高估值科技杠杆资金后，市场开启暴跌模式，情绪已经传导至A股。 第二，美联储加息概率已飙到50%，对科技股估值构成直接压制。 第三，美伊冲突停不下来，高油价持续越久，加息预期就越强，时间也可能提前。 所以，悬在科技头上的达摩克利斯之剑，现在还没消失。今天的走势，暂以超跌反弹看待，持续性存疑。 储能方面，美方准备签署禁令，禁止引入中国逆变器等产品，导致储能核心公司集体重挫。锂电池等公司业绩不错，股价却涨不动——市场担心2027年新能源汽车税收优惠取消后，销量会大幅滑坡。新能源方向的下跌，源于资金抢跑。 商业航天和人形机器人已被爆炒过，缺乏新逻辑，短期内很难再起一波。 所以，为什么跌跌不休？因为科技倒下后，没有板块能接住。市场没有主线，操作难度极大。 今天虽然收出止跌信号，但K线仍在5日、10日均线之下，反弹还是缩量的——这不是好现象。场外利空未消，投资者信心远未恢复。如果接下来指数不能在关键位置稳住，今天的反弹就只是下跌途中的一次喘息，不是反转。 超跌反弹可以做，但要降低预期。真正的主线，要等市场自己走出来。
生成时间：2026-08-04T07:45:00.083Z