英特尔业绩暴跌拖累芯片股科技巨头扩资本开支显露AMD买点

浑水摸鱼清源

市场将英特尔内部经营溃败与半导体周期混为一谈属于非理性恐慌，云巨头激增的算力开支令深度回调后的AMD与亚马逊成为未来最具吸引力的核心资产。

AI-assisted summary of 【Financial Education】科技巨头财报引发抛售潮，半导体板块到底该不该抄底？ — original source. Check the original for full context.

Source published: 2026-10-31

Key ideas

纳斯达克100指数ETF从高点回撤10%进入技术性回调，英特尔暴跌超31%、AMD跌12.5%，半导体板块遭遇系统性非理性抛售。 (0:00)
本轮恐慌核心在于英特尔PC端指引疲软引发连锁误伤，但科技巨头（亚马逊、谷歌）反而宣布大幅提升2025/2026年AI资本开支，算力底座需求并未衰退。 (6:05)
半导体板块预计在11月至次年4月迎来强劲反弹，博主看好AMD在未来24至36个月实现股价翻倍甚至三倍，成长性显著优于英伟达与英特尔。 (10:00)
科技巨头中亚马逊被列为长期首选买入标的，AWS营业利润率维持在近40%且营收增速达39%，短期自由现金流承压主要系Q3资本开支前置所致。 (27:07)

Detailed notes

## Background 本期视频由UP主“浑水摸鱼清源”于2026年10月31日发布在哔哩哔哩平台。视频围绕科技巨头最新财报引发的纳指回调与半导体板块抛售潮展开，重点探讨科技公司大幅增加资本开支（CapEx）的长期影响，并深入分析英特尔暴雷对行业的传导以及亚马逊等龙头的投资价值。

## TL;DR 面对纳指遭遇10%的技术性回调以及芯片股大跌，创作者系统拆解了恐慌的根本成因：科技巨头激进上调AI资本开支引发短期自由现金流恶化的担忧，叠加英特尔疲软业绩指引带来的行业性非理性误杀。创作者强调英特尔面临的是十年级别的结构性份额流失，而不能代表整体半导体景气度，具备高算力需求的AMD与估值处于20-24倍合理区间的英伟达依然拥有强劲的成长跑道。在七大科技巨头中，创作者将亚马逊评为当前风险收益比最佳的核心标的，并结合分析师观点指出AWS接近40%的营业利润率与创纪录的经营现金流足以为未来的巨额投资托底，建议投资者保持五年维度的长线视角，切勿在恐慌中盲目抛售优质资产。

## Atmosphere & style 整期视频充满着资深投资者面对市场崩盘时的从容与笃定，创作者以一种“老兵安抚新兵”的镇定语气开场，将暴跌比作橄榄球赛中的常态碰撞，迅速稳定受众心态。在切入英特尔和华尔街的羊群效应时，语言风格急转直下，变得尖锐辛辣，将英特尔戏谑地比作日落西山的“Kmart”，对市场不分青红皂白盲目抛售优质芯片股的行为表达了强烈的荒谬感。随着视频结合CNBC直播访谈与财报细节进行逐层推演，语调重新归于严谨理性的长线逻辑，在滔天的市场恐慌中构筑起一份清醒、扎实且富有说服力的投资底气。

## Bulletin - [00:00](https://www.bilibili.com/video/BV1QZxC6kEWi?t=0s) 纳斯达克100指数ETF从高点回撤10%进入技术性回调，英特尔暴跌超31%、AMD跌12.5%，半导体板块遭遇系统性非理性抛售。 - [06:05](https://www.bilibili.com/video/BV1QZxC6kEWi?t=365s) 本轮恐慌核心在于英特尔PC端指引疲软引发连锁误伤，但科技巨头（亚马逊、谷歌）反而宣布大幅提升2025/2026年AI资本开支，算力底座需求并未衰退。 - [10:00](https://www.bilibili.com/video/BV1QZxC6kEWi?t=600s) 半导体板块预计在11月至次年4月迎来强劲反弹，博主看好AMD在未来24至36个月实现股价翻倍甚至三倍，成长性显著优于英伟达与英特尔。 - [27:07](https://www.bilibili.com/video/BV1QZxC6kEWi?t=1627s) 科技巨头中亚马逊被列为长期首选买入标的，AWS营业利润率维持在近40%且营收增速达39%，短期自由现金流承压主要系Q3资本开支前置所致。

## Key takeaways - [02:04](https://www.bilibili.com/video/BV1QZxC6kEWi?t=124s) 纳斯达克指数正式陷入10%回调区间，长线投资者应做好应对30%以上全面崩盘的心理建设，切勿妄图精准抄底。 - [06:05](https://www.bilibili.com/video/BV1QZxC6kEWi?t=365s) 博主评估未来12个月美股发生全面崩盘的概率约为25%，但市场不发生崩盘的概率依然高达75%。 - [06:05](https://www.bilibili.com/video/BV1QZxC6kEWi?t=365s) 谷歌与亚马逊宣布大幅追加明年资本支出，引发市场对其未来陷入严重负自由现金流、巨额举债及股东权益稀释的担忧。 - [09:07](https://www.bilibili.com/video/BV1QZxC6kEWi?t=547s) 英特尔沦为低质量受困标的，其在PC需求疲软和制程转型上的暴雷被市场错误解读为整个半导体周期的下行信号。 - [22:15](https://www.bilibili.com/video/BV1QZxC6kEWi?t=1335s) 将优质半导体企业与英特尔捆绑抛售毫无逻辑，好比因廉价折扣零售商Kmart业绩惨淡而去清仓好市多、沃尔玛和塔吉特。 - [10:00](https://www.bilibili.com/video/BV1QZxC6kEWi?t=600s) 历史规律显示11月至次年4月通常是芯片板块爆发窗口，优质半导体股票在未来四至六个月内有望出现大幅反弹。 - [12:35](https://www.bilibili.com/video/BV1QZxC6kEWi?t=755s) 半导体全面周期性降速至少还在两到三年之后，AMD坐拥3-4年明确增长期，内存厂商也有1-2年稳固需求。 - [12:35](https://www.bilibili.com/video/BV1QZxC6kEWi?t=755s) 英伟达目前预期市盈率在20至24倍左右，估值合理并未反映泡沫，但由于基数庞大，未来股价翻倍空间受限。 - [16:05](https://www.bilibili.com/video/BV1QZxC6kEWi?t=965s) 企业固定IT预算正遭到AI Token消费挤压，使得传统PC及移动硬件升级预算缩水，加剧了以PC业务为支柱的英特尔的困境。 - [25:35](https://www.bilibili.com/video/BV1QZxC6kEWi?t=1535s) 半导体标的配置上应坚决回避英特尔等低质标的，首选AMD，其向2027年迈进的营收与利润增速有望全面超越英伟达。 - [27:07](https://www.bilibili.com/video/BV1QZxC6kEWi?t=1627s) 美股七巨头最新评级排序中，亚马逊位列第一，Meta排第二，谷歌排第三，而特斯拉则是当前风险收益比最差的标的。 - [30:00](https://www.bilibili.com/video/BV1QZxC6kEWi?t=1800s) 亚马逊Q3出现创纪录的负自由现金流主要由于重资产开支被前置执行，全财年CapEx指引并未上调，Q4自由现金流有望回正。 - [31:50](https://www.bilibili.com/video/BV1QZxC6kEWi?t=1910s) 受制于当前疲软的宏观消费情绪，大型科技股需以5年长周期视角审视，短期内唯有AMD最具爆发潜力。 - [34:40](https://www.bilibili.com/video/BV1QZxC6kEWi?t=2080s) 亚马逊整体录得创纪录的营业利润率与经营现金流，随着资本开支逐步稳定，预计在2028至2029年重回巨额自由现金流造血状态。

## Detailed walkthrough ### [00:00](https://www.bilibili.com/video/BV1QZxC6kEWi?t=0s) 科技股遭遇血洗与纳斯达克回调

周五本不是创作者常规发布视频的时间，但在市场经历剧烈抛售的情况下，他破例录制了这期内容。从盘面表现来看，科技板块遭遇了全方位的重创：

- 英特尔（Intel）股价暴跌超 31%； - 闪迪（SanDisk）日内重挫约 16%； - AMD 跌幅达 12.5%； - CoreWeave 跌超 54%； - 纳斯达克 100 指数 ETF（QQQ）自高点累计下跌 10%，正式进入技术性回调区间（correction territory）。

市场多重利空交织，恐慌情绪蔓延。面对这种局面，创作者表示将在本期内容中系统梳理导致抛售的多重核心诱因、分析市场真正的风险与潜在投资机会，并重点分析亚马逊（Amazon）的财报表现，同时对分析师 Mark Mahaney 针对亚马逊的观点进行深入点评。

[02:04](https://www.bilibili.com/video/BV1QZxC6kEWi?t=124s) 长期投资者的心态管理与应对策略

每逢市场经历深度回调或暴跌，许多观众常将创作者视为心理咨询师。创作者结合自己打过 5 年美式橄榄球（tackle football）的经历分享了应对心态。在橄榄球这种高对抗性的运动中，即便自身队伍常年处于被看好的一方，也绝对不能低估任何对手；一旦将其视为胜券在握的轻松比赛，往往就会在遭遇强敌时手足无措。因此，必须在赛前做好应对最坏情况的心理建设，才能在比赛陷入僵局时避免恐慌。

创作者将这种心理机制直接迁移至股市投资：面对纳指 10% 的回调，成熟的投资者理应将其视为常态，甚至在心理上时刻做好应对全面崩盘（full on market crash）的准备。创作者强调，拥有 18 年投资经验的他见证过至少 3 次纳斯达克指数跌幅超 30% 的毁灭性暴跌、数次大型回调、大量类似于当前 10% 幅度的中小级别回调，以及数不胜数的日常回踩。

> “When you've been in the market 18 years, I've seen everything.”

对于新入市两三年的投资者而言，持仓个股自高点回撤 20% 可能会引发极度的恐惧或错过底部的焦虑（FOMO）；但对于资深长线跑者而言，跑 2 英里就像在公园散步一样轻松。在暴跌中建仓的关键原则是绝对不要幻想精准抄底。例如投资者当前计划以某个价位建仓 AMD，就必须在心理上做好它进一步深跌的准备，哪怕未来并不一定会发生，也要从容预设最坏情境，才能真正把握市场恐慌带来的建仓契机。

[06:05](https://www.bilibili.com/video/BV1QZxC6kEWi?t=365s) 科技巨头激进 CapEx 与现金流恶化担忧

对于未来 12 个月内美股发生全面崩盘的概率，创作者给出的评估是 25% 左右。这一概率虽然不容忽视，但意味着市场不发生全面崩盘的概率依然高达其三倍。当前科技股抛售的核心诱因，实际上源自两组相互矛盾的宏观逻辑博弈：以亚马逊和谷歌为代表的科技巨头，与以英特尔为代表的芯片制造商。

```mermaid flowchart TD accTitle: Tech Capex vs Semiconductor Dilemma accDescr: Divergence between Big Tech aggressive capex plans and Intel weak guidance causing semiconductor panic. subgraph BigTech["Big Tech (Amazon, Google)"] direction TB C1["Substantially increasing 2025 CapEx"] --> C2["Deep negative Free Cash Flow"] C2 --> C3["Financing pressure: Massive debt or dilution"] end subgraph MarketImpact["Market Interpretation"] direction TB C3 -->|"Divergence"| M1["Normally bullish for semiconductors"] M1 -.->|"Contradicted by"| I1 end subgraph SemiSpace["Intel & Chips"] direction TB I1["Intel weak PC guidance & demand drop"] --> I2["Sector-wide panic selling"] I2 --> I3["Quality peers mispriced: AMD, MU, etc."] end ```

谷歌与亚马逊在财报与电话会中释放了极为激进的信号：计划在明年大幅追加资本支出（CapEx）。这两家公司的自由现金流（Free Cash Flow）本已承受巨大下行压力，未来几个季度甚至可能陷入严重的负自由现金流状态，而管理层不仅没有放缓脚步，反而暗示明年的开支将跃升至全新台阶。这引发了华尔街对财务结构的强烈担忧： - **债务激增**：亚马逊可能不得不通过巨额发债来维持资本开支，导致利息费用高企，未来两到三年或沦为负债沉重的公司； - **股权稀释**：谷歌、Meta 等巨头若不再进行大规模股票回购，反而转为增发融资，将直接稀释现有股东权益，导致持股比例缩水。

通常情况下，科技巨头激进的算力投入对半导体板块（如 AMD、美光科技 Micron、闪迪等）属于重大利好，但这一逻辑却被英特尔的暴雷彻底打乱。

[09:07](https://www.bilibili.com/video/BV1QZxC6kEWi?t=547s) 英特尔业绩暴雷对半导体板块的负面传导

本轮半导体板块泥沙俱下的直接导火索是英特尔。创作者直言英特尔在行业内早已沦为一家低质量、缺乏竞争力的公司，其股价已自高点腰斩超过 55%。长期跟踪半导体产业链的投资者都清楚其产品力与运营水准的落后，但英特尔在财报中关于个人电脑（PC）市场需求放缓、整体业务表现疲软的悲观指引，直接引发了全行业的连锁恐慌。

> “Intel's a hunk of junk of company, okay? It's a real piece of junk.”

市场将英特尔自身在 PC 端的颓势和内部经营困境，错误地泛化为整个半导体行业即将迎来全面周期性走弱。这种非理性恐慌迅速向外蔓延，导致 AMD、美光科技等基本面完全不同的优质半导体标的被无差别抛售，形成了巨头巨额 CapEx 利好与英特尔疲软指引互相矛盾的混乱局面。

[10:00](https://www.bilibili.com/video/BV1QZxC6kEWi?t=600s) The Irrational Spillover from Intel and the Six-Month Semiconductor Outlook

The panic sweeping through Wall Street began when Intel reported its quarterly numbers and flagged softer guidance alongside memory pricing pressures. Investors immediately drew a line from Intel's troubles to the rest of the semiconductor landscape, assuming that weakness at Intel must mean AMD will come in light, which in turn means memory chipmakers will face shortfalls in revenue and earnings per share. The host finds this collective freakout completely backwards, pointing out the glaring contradiction in the tape: massive hyperscalers and big tech titans just confirmed they have to dramatically increase their capital expenditures next year, yet semiconductor stocks are crashing across the board simply because Intel stumbled. As the host flatly observed: “And yet all the semiconductor stocks are crashing right now because of Intel? Doesn't make sense.”

Looking ahead at the seasonal and cyclical setup over the next six months—roughly November through April—the host expects a sharp rebound rather than an extended decline. While historical patterns frequently show markets carving out a bottom right around Halloween, there remains the lingering risk of a late-year breakdown similar to late 2018, when equities steadily cratered through November and December before finally hitting a trough around Christmas. Nevertheless, the host views that prolonged decline as a low-probability scenario, arguing instead that semiconductor names are poised for a massive rally over the coming two quarters.

```mermaid flowchart TD accTitle: Intel Weakness Spillover vs Big Tech CapEx Reality accDescr: Diagram contrasting market panic caused by Intel with underlying demand from hyperscaler CapEx.

subgraph FearDriven["Market's Flawed Chain Reaction"] direction TB A["Intel warns on memory & guidance"] --> B["Market assumes AMD & memory stocks miss"] B --> C["Sector-wide panic selling across semis"] end

subgraph FundamentalReality["Underlying Industry Reality"] direction TB D["Big Tech increases 2025/2026 AI CapEx"] --> E["Explosive demand for GPUs, compute & memory"] E --> F["Projected 6-month 'ripper rally' (Nov–Apr)"] end

FearDriven -.->|"Disconnected from"| FundamentalReality ```

- **Market Bottom Window**: Historically, equities often form a seasonal floor around Halloween; the lower-probability downside analog is the fourth quarter of 2018, which did not bottom until Christmas week. - **The CapEx Disconnect**: Cloud hyperscalers are deploying record capital expenditure for AI infrastructure, yet semiconductor suppliers are selling off in sympathy with a legacy vendor. - **Six-Month Outlook**: The host predicts that four to six months out, high-quality semiconductor equities will trade substantially above current depressed levels.

---

[12:35](https://www.bilibili.com/video/BV1QZxC6kEWi?t=755s) Valuation Reality and Multi-Year Growth Runways Across Semis

While an eventual cyclical slowdown in the semiconductor trade will arrive one day, that cyclical ceiling is still two to three years in the future. For now, major tech platforms must allocate every spare dollar to compute, giving companies distinct visibility into multi-year expansion. AMD effectively possesses three to four years of virtually guaranteed growth, while memory chip manufacturers sit on a solid one to two years of high demand. Nvidia has at least another full year of massive expansion ahead.

Addressing concerns that Nvidia might face decelerating growth rates or margin pressure after another year, the host highlights the stock's modest valuation multiple. Rather than trading at an exorbitant valuation like 100 times forward earnings—which would make any future slowdown terrifying—Nvidia trades at a forward price-to-earnings multiple of roughly 20 to 24. A valuation at that level removes substantial downside risk because the market is not pricing in perfection. As the host explained: “If Nvidia was trading at a hundred forward P right now, it'd be scary to be like, oh crap. Nvidia has probably only got one more year of massive growth and then they're going to have, you know, slow growth rates after that... but when you're trading at a 24-P, it's not scary like that.”

- **AMD Growth Runway**: 3 to 4 years of massive, high-visibility expansion ahead as enterprise and datacenter share gains accelerate. - **Memory Chipmakers**: 1 to 2 years of guaranteed growth driven by server memory demand and AI clustering requirements. - **Nvidia Valuation**: Trading at approximately 20x to 24x forward earnings, keeping the stock grounded despite fears of eventual growth normalization. - **Cycle Timing**: The broader semiconductor supercycle slowdown remains an estimated 2 to 3 years away.

---

[13:58](https://www.bilibili.com/video/BV1QZxC6kEWi?t=838s) Portfolio Carnage, Tech Crashes, and Resilient Outliers

The immediate damage on the trading board is steep, with Intel dropping 31%, SanDisk sliding 14%, AMD falling 12%, CoreWeave shedding over 10%, and Micron down nearly 7%. In contrast to the bloodbath in chip and enterprise cloud stocks, specific consumer and defensive holdings in the host's portfolio held up remarkably well: Celsius Holdings, Nike, and Wynn Resorts traded green, while The Cheesecake Factory (CAKE) held near flat on an otherwise brutal day. Estée Lauder and Netflix barely registered losses, underscoring that the current liquidation is concentrated in tech rather than a broad-based macroeconomic unwind.

The host remains especially bullish on beaten-down small-cap turnaround plays like The Honest Company, reiterating high conviction that the stock will finish the year comfortably above $5 per share. Meanwhile, high-flying software names like ServiceNow took a severe beating—dropping hard two days in a row—whereas Salesforce weathered the storm far better, down only 1%. Cataloging drawdowns from all-time highs across social media, the host distinguished standard corrections from full-blown crashes: drops exceeding 30% represent genuine crashes, while declines surpassing 50% constitute "mega crashes."

- **Consumer Resiliency**: Outliers like Celsius, Nike, and Wynn remained positive, while The Cheesecake Factory held steady amidst broad market panic. - **Software Divergence**: ServiceNow suffered back-to-back plunges, whereas Salesforce displayed relative stability, dropping only about 1%. - **High-Conviction Microcap**: The Honest Company is projected by the host to exit the year trading well over $5.00 per share.

---

[16:05](https://www.bilibili.com/video/BV1QZxC6kEWi?t=965s) Vivek Arya on Intel's Downgrade and the AI Budget Squeeze

Turning to live television coverage, the host reacts to a CNBC interview featuring Bank of America Securities Senior Semiconductor Analyst Vivek Arya, who downgraded Intel from Buy to Neutral after the stock collapsed between 25% and 31%. Arya stated that the ongoing AI revolution is no longer lifting all boats equally. While segments tied directly to cloud infrastructure, token consumption, and hyperscaler CapEx are thriving, they are cannibalizing the corporate IT budgets that historically funded upgrades for traditional client hardware like PCs and mobile devices.

Although Intel's data center business grew nearly 60% year-over-year—its fastest pace in 15 years—it still lags the explosive growth delivered by AMD, Nvidia, and ARM-based suppliers. More critically, client PCs still represent roughly half of Intel's total business. Enterprise customers face a fixed overall IT budget: as cloud titans like Amazon, Google, and Microsoft post 40% to 100% cloud growth and report a 10x surge in AI token demand, corporate clients are reallocating their capital toward AI tokens rather than cyclical fleet PC replacements. Intel is forced to pour billions into manufacturing and foundries to court external customers down the road, while its core consumer revenue engine is starved for demand. Explaining the structural dilemma facing Intel, Vivek Arya noted: “So if they're having to invest more in AI, that means they have less money to invest in upgrading their PC infrastructure.”

- **Analyst Action**: Vivek Arya of BofA Securities lowered Intel from Buy to Neutral, advising investors to remain on the sidelines through a multi-quarter transition. - **Client PC Exposure**: Traditional PC hardware still represents approximately 50% of Intel's total business, leaving it overexposed to sluggish consumer and enterprise refresh cycles. - **Data Center Contrast**: Intel's data center revenue jumped nearly 60% year-over-year—its fastest growth in 15 years—yet it is still losing ground to Nvidia and AMD. - **Budget Cannibalization**: Enterprise IT budgets remain essentially fixed, meaning exponential increases in AI token consumption directly deplete capital previously earmarked for corporate PC refreshes.

[20:00](https://www.bilibili.com/video/BV1QZxC6kEWi?t=1200s) The AI Capex Dilemma and Intel's Lost Decade to AMD

The discussion opens with an interview clip exploring whether enterprise AI demand is genuinely non-cyclical or just undergoing much longer, multi-year cycles. The CNBC host references Gary Cohn's remarks questioning whether the industry is overbuilding infrastructure, asking whether a struggling incumbent like Intel should stay anchored in personal computing and consumer hardware or aggressively pour capital into chasing AI. The guest responds by drawing a clear boundary between real-world workload demand and current financial return cycles, noting: “I think, Becky, we have to differentiate between demand and near-term economics.” While demand for computing power is off the charts, companies must commit massive upfront capital expenditures to build out physical infrastructure for returns that might only materialize several years later.

```mermaid flowchart TD accTitle: AI Infrastructure Capex Mismatch accDescr: Front-loaded capital expenditures precede multi-year revenue realization, creating near-term financial pressure. subgraph Buildout["Front-Loaded Investment (Years 1-2)"] A["Massive Capex Commitments"] --> B["Data Center & Fab Buildouts"] end subgraph Lag["The Economic Gap"] B --> C["Depressed Free Cash Flow & Compressed Margins"] end subgraph Realization["Payoff Horizon (Years 3-5)"] C --> D["Scaled AI Enterprise Demand"] D --> E["High-Margin Revenue Realization"] end ```

Reacting to the interview, the host expresses utter exasperation with the market's sympathetic framing of Intel. Looking at the long-term fundamentals on `thousandxstocks.com`, the host points out that Intel’s collapse has nothing to do with temporary macro turbulence or healthy cyclical swings. As the host bluntly summarizes the competitive dynamics: “AMD has been eating Intel's lunch for years. Like, the last decade. They've been taking massive amounts of market share, right? AMD's been owning Intel.” Intel’s margins, earnings per share, and free cash flow trajectories all show persistent structural decay. In reality, the only reason Intel has experienced even modest top-line support recently is that the broader semiconductor boom created an environment where hyperscalers and enterprises were desperate for any available silicon on the market.

- Enterprise AI demand remains structurally high, but requires heavily front-loaded capital investment. - AMD has systematically captured server and PC market share from Intel across the past ten years. - Intel’s operational metrics—operating margins, EPS, and free cash flows—demonstrate structural deterioration rather than short-term cyclicality.

---

[22:15](https://www.bilibili.com/video/BV1QZxC6kEWi?t=1335s) The Kmart Analogy: Why Punishing the Entire Semiconductor Sector Is Irrational

The host turns to Intel's cash flow charts to underline how fundamentally broken the business has become, arguing that without the CHIPS Act capital infusions and the broader AI tide, Intel might already be facing bankruptcy. Despite this glaring internal divergence, Wall Street treated Intel's disastrous quarterly print as an indictment of the entire semiconductor space, sparking a broad-based panic. The host finds this guilt-by-association completely absurd, comparing it to retail investors dumping premier operators like Costco, Walmart, and Target simply because a failing discount retailer reported abysmal foot traffic.

To capture the irrationality of the sell-off, the host puts the market's logic into stark terms: “It's Kmart. Kmart is trash. We're going to sell off all the quality stocks because Kmart came in with bad numbers? What are we doing here?” Dumping industry-leading semiconductor companies with pristine balance sheets and soaring data center revenues over Intel’s self-inflicted woes reflects an emotional overreaction rather than rational capital allocation. The host laments the market's herd behavior: “It's amazing, man. The whole semiconductor stock's getting sold off today for Kmart.”

- Intel's financial survival has been heavily subsidized by government capital infusions and an overarching industry shortage. - Wall Street's indiscriminate dumping of top-tier chip equities is directly comparable to selling Target and Walmart because Kmart had a bad quarter. - High-quality semiconductor leaders are being dragged down by a low-quality laggard's inability to execute.

---

[24:14](https://www.bilibili.com/video/BV1QZxC6kEWi?t=1454s) Strategic Domestic Foundry vs. Wall Street Reality: The 3 to 5 Year Payoff Gap

The segment switches back to the television panel, where the defending analyst argues that Intel's core strategic roadmap is conceptually sound even if its near-term financials look dreadful. Intel continues to see demand in the data center, and its long-term bet hinges on heavy investments in advanced domestic fabrication facilities. However, advanced manufacturing is fundamentally an asymmetric bet: modern semiconductor fabs require three to five years of lead time and billions in front-loaded capex before yielding high-margin silicon at scale.

Beyond pure corporate earnings, the guest emphasizes that Intel serves a critical national security and geopolitical role as America's domestic manufacturing champion. Relying entirely on concentrated Asian supply chains carries severe geopolitical risks, making domestic diversification an absolute imperative. The guest makes the strategic case: “Intel is an incredibly important source of U.S. domestic manufacturing. We absolutely need diversification and chip supply. We cannot just be dependent on a handful of suppliers in Asia, as much as they are very critical and they are investing in the U.S., but Intel has to be successful as kind of the leading domestic supplier.” Yet for public equity investors, holding a company through years of capital drag while its competitors sprint ahead in software and architecture remains an agonizing proposition.

- Foundry manufacturing requires a 3 to 5 year lead time before heavy front-loaded capex yields meaningful free cash flow. - Intel remains strategically essential to U.S. domestic supply chain security and geographic diversification. - The fundamental tension lies between long-term national industrial policy and near-term equity performance.

---

[25:35](https://www.bilibili.com/video/BV1QZxC6kEWi?t=1535s) Semiconductor Quality Tiers: Betting on AMD Over Nvidia

Cutting back to the host, the conversation shifts to how investors should allocate capital within tech: portfolio positioning must prioritize top-tier quality and market share winners over troubled turnarounds. The host rejects low-tier plays outright, labeling Intel as an uninvestable "low-quality" (LC) stock and grouping it once again with Kmart. Even with memory plays, the host emphasizes sticking to premier operators, preferring Micron over secondary names like SanDisk. When looking at the elite AI chip landscape, the host narrows the field strictly down to two viable options: Nvidia and AMD.

```mermaid flowchart TD accTitle: Semiconductor Quality and Allocation Hierarchy accDescr: Hierarchy showing the host's preferences between top-tier winners, solid performers, and uninvestable laggards. subgraph Tier1["Top-Tier Core Holdings"] AMD["AMD: High 2027 growth, 2-3x upside potential"] NVDA["Nvidia: Dominant leader, maturing growth rates"] end subgraph Tier2["Solid Quality / Selective"] AVGO["Broadcom: Solid business, not personal pick"] MU["Micron: Preferred memory play over SanDisk"] end subgraph Tier3["Avoid / Low Quality"] INTC["Intel: Structural share loss, 'Kmart of Semis'"] end ```

While acknowledging Nvidia's dominance, the host explains the deliberate choice to concentrate capital on AMD instead of Nvidia over the past year. Nvidia’s stock had already climbed above $200 a year earlier and has essentially hovered around that level, whereas AMD has maintained explosive upside momentum. Looking out toward 2027, the host projects AMD's revenue and earnings growth rates to meaningfully outpace Nvidia's as AMD takes further market share. In the host's view, AMD possesses the underlying fundamentals to potentially double or triple in share price over the next 24 to 36 months, whereas expecting a similar multiple expansion from Nvidia's massive market capitalization over that same timeframe is far less realistic given its decelerating growth trajectory.

- Intel and lower-tier legacy memory makers are classified as low-quality traps to avoid entirely. - The host's semiconductor allocation is concentrated heavily in AMD rather than Nvidia or Broadcom. - AMD is projected to deliver faster growth rates leading into 2027, with the potential to double or triple within 24 to 36 months.

---

[27:07](https://www.bilibili.com/video/BV1QZxC6kEWi?t=1627s) Ranking the Magnificent Seven: Amazon, Google, and the Best Long-Term Buys

The host pivots to evaluate the broader Magnificent Seven tech giants, ranking their risk-to-reward profiles following the recent earnings pullbacks. Tesla is swiftly dismissed as the least attractive stock in the basket by a wide margin. Apple is described as uninspiring, while Microsoft is characterized as a dependable, low-volatility compounder likely to deliver annualized returns in the teens—between 13% and 19% per year over the next four to five years, easily beating the broader market. When identifying the single best opportunity for long-term investors, the host weighs Amazon, Meta, and Google:

> “I would say the most attractive risk-reward right now is... I've got to give it to Amazon.”

Amazon edges out the top spot due to AWS's re-accelerating growth at 39% year-over-year, with guidance pointing toward 40% or higher over the next two years. Meta comes in as a close second, though the host expresses caution regarding Mark Zuckerberg’s aggressively escalating capex when the monetization pipeline for speculative bets like Muse has yet to show concrete revenue returns. Google takes the third spot but provided the most positive quarterly surprise. While Wall Street reacted negatively to Alphabet’s earnings release, the host came away more enthusiastic about Google than in years, highlighted by Google Cloud’s explosive 103% growth rate. Across the mega-cap space, secular cloud expansion and tangible AI monetization make Amazon and Google the most compelling risk-adjusted buys.

[30:00](https://www.bilibili.com/video/BV1QZxC6kEWi?t=1800s) Wall Street Reaction to Amazon Earnings and CapEx Surge

Wall Street reacted sharply to Amazon’s earnings release after forecasts indicated capital expenditures reaching staggering levels, prompting a sell-off across the shares. Mark Mahaney, Head of Internet Research at Evercore ISI, joined the broadcast to analyze whether the market's response was warranted. While Meta and Alphabet (Google) also experienced negative market reactions following their prints, Mahaney pointed out that Amazon faced three distinct points of friction against street expectations. First, the market had priced in low-40s revenue growth for Amazon Web Services (AWS) of 40% or more, but received 39%. Even though AWS growth accelerated for the sixth consecutive quarter, it fell slightly short of whisper numbers. Second, advertising revenue grew at 12%, missing expectations that sat closer to 18% or 19%, largely due to tough year-over-year comparison issues that the company expects to normalize back into the high teens or near 20% in the following quarter. Third, Amazon reported record negative free cash flow, which was driven by CapEx being heavily pulled forward into the third quarter rather than an increase in full-year spend.

```mermaid flowchart TD accTitle: Amazon Earnings Expectations vs Realities accDescr: Diagram illustrating the three areas where Amazon results diverged from Street expectations and the long-term outlook.

subgraph Divergence["Q3 Expectations vs. Realities"] A["AWS Revenue Growth"] -->|"Wanted: 40%+"|"A1"["Delivered: 39% (6th straight quarter of acceleration)"] B["Ad Revenue Growth"] -->|"Wanted: 18-19%"|"B1"["Delivered: 12% (Comp issues normalizing to high teens)"] C["Free Cash Flow"] -->|"Negative Record"|"C1"["Heavy Q3 CapEx pull-forward; Q4 expected positive"] end

subgraph Outlook["Long-Term Trajectory"] D1["Capital Cycle"] --> D2["AWS 40%+ Growth Target"] D2 --> D3["Substantial FCF Recovery by 2028-2029"] end

Divergence --> Outlook ```

Mahaney emphasized that this sell-off is fundamentally an expectations correction rather than an operational breakdown, maintaining Amazon as his top long-term pick despite the near-term pain. When asked where the stock could trade twelve months out from its current $226 level, Mahaney argued that the company can return to trading around 25 times earnings, provided investors receive clarity on the duration of its capital investment cycle and when the free cash flow trough—which could see a negative $50 billion hole next year—begins to reverse.

- Wall Street projected total capital expenditures reaching roughly $220 billion across the big tech investment cycle. - Amazon advertising growth came in at 12% against street targets of 18% to 19%, but management signaled high-teens to near-20% growth in subsequent quarters. - Full-year CapEx guidance was not raised; the negative free cash flow was caused by front-loaded Q3 outlays, setting up Q4 for positive free cash flow. - Mark Mahaney maintained his bullish conviction on the equity: “Yeah, it's a number one pick, and it's my number one buy, even here off the correction.”

[31:50](https://www.bilibili.com/video/BV1QZxC6kEWi?t=1910s) The Long-Term 5-Year Horizon vs. Short-Term Consumer Headwinds

Stepping back from Wall Street's commentary, the host cautions viewers against falling into the trap of expecting megacap tech stocks to stage an immediate, explosive rally over the next 6 to 12 months. While it remains theoretically possible for Amazon to push past $300 or Meta to cross $900 in the coming year, broader macroeconomic headwinds are heavily weighing on corporate growth. The host highlights that business spending around enterprise AI infrastructure remains durable, but any company tied to consumer spending is dealing with an increasingly strained backdrop reflected in consumer sentiment metrics and retail company earnings. Because of this drag, treating these companies as short-term momentum trades risks significant disappointment.

Instead, the host urges investors to evaluate big tech positions strictly through a five-year investment lens, looking beyond current cash-flow burn to major commercial milestones. For Amazon, an upcoming catalyst is Project Kuiper (referenced as Project Leo), which is scheduled to transition into commercial operations before the end of the year. Shifting that project into active service will allow Amazon to capitalize satellite launch and manufacturing costs, relieving pressure on the income statement. Among the major tech names experiencing multiple compression, the host singled out AMD as the solitary name positioned with high potential to blast substantially higher over a shorter window, whereas the rest of mega-cap tech requires patient, multi-year positioning.

- Consumer sentiment and broader consumer-facing spending data remain exceptionally soft, posing ongoing headwinds to tech retail and digital ad volume. - Commercialization of Amazon’s satellite broadband initiative before the end of the year will allow rocket manufacturing and deployment expenses to be capitalized. - The host explicitly contrasted short-term hype with realistic portfolio management: “You really need to be thinking about these sorts of companies in a five-year span.” - Addressing short-term potential across tech names, the host noted: “The only one I would say I have my hopes up on in that I think could blast higher in the shorter term is AMD.”

[34:40](https://www.bilibili.com/video/BV1QZxC6kEWi?t=2080s) AWS Profitability Path and Long-Term Operating Margins

The discussion returned to the live CNBC exchange where anchor Becky Quick questioned Mark Mahaney on whether Amazon’s aggressive spending possesses a viable path to bottom-line profitability, citing CEO Andy Jassy’s conference call remarks that the leadership team sees a clear line of sight to returns. Mahaney confirmed that the underlying unit economics support Jassy's claims. AWS continues to post operating margins near 40%, and maintaining those margins while sustaining 40% top-line expansion across the next two years provides exceptional operating leverage. Backed by expanding commercial deal backlogs, Amazon is positioned to return to substantial positive free cash flow as early as 2028, and almost certainly by 2029.

Beyond the cloud division, Amazon’s consolidated results demonstrated enterprise-wide operational strength, printing record overall operating margins and record operating cash flow. The underlying business engines are generating unprecedented internal cash; the apparent financial strain stems purely from the timing of heavy infrastructure investments rather than declining profitability. Closing out the analysis, the host embraced the mantra that Amazon fundamentally remains an essential core holding, before concluding the broadcast with an announcement regarding a limited-time research platform membership discount.

- AWS maintains operating margins hovering around 40%, providing the foundation for outsized cash generation once CapEx spend stabilizes. - Amazon printed record overall operating margins and record operating cash flow on a consolidated basis. - Substantial positive free cash flow is projected to normalize between 2028 and 2029 as CapEx investments begin compounding. - Highlighting CEO Andy Jassy's conviction during the earnings call, Becky Quick asked: “Andy Jassy said that he sees a clear line of sight to profitability for all of this. Do you?” - Mark Mahaney defended the structural durability of the cloud segment: “I mean, the margins at AWS are close to 40 percent.”

## Notable quotes > [20:00](https://www.bilibili.com/video/BV1QZxC6kEWi?t=1200s) AMD has been eating Intel's lunch for years. — 浑水摸鱼清源

> [10:00](https://www.bilibili.com/video/BV1QZxC6kEWi?t=600s) all the semiconductor stocks are crashing right now because of Intel? Doesn't make sense. — 浑水摸鱼清源

> [27:07](https://www.bilibili.com/video/BV1QZxC6kEWi?t=1627s) I would say the most attractive risk-reward right now is... I've got to give it to Amazon. — 浑水摸鱼清源

## Mentioned 英特尔 (Intel), AMD, 英伟达 (Nvidia), 亚马逊 (Amazon / AMZN), AWS, 谷歌 (Google / Alphabet), Meta Platforms, 美光科技 (Micron), 闪迪 (SanDisk), CoreWeave, 纳斯达克 100 指数 ETF (QQQ), 微软 (Microsoft), 苹果 (Apple), 特斯拉 (Tesla), 博通 (Broadcom), ServiceNow, Salesforce, The Honest Company, Celsius Holdings, Nike, Wynn Resorts, The Cheesecake Factory, Estée Lauder, Netflix, Mark Mahaney, Vivek Arya, Andy Jassy, Becky Quick, Gary Cohn, Project Kuiper

## Video description 【Financial Education】科技巨头财报引发抛售潮，半导体板块到底该不该抄底？ crash

｜半导体板块的非理性恐慌近期科技股遭遇剧烈回调，纳指跌幅一度达到10%。英特尔股价跌幅超过31%，美光科技跌超30%，AMD跌幅也达到25%。市场情绪极度脆弱，主要源于英特尔发布了疲软的业绩指引，导致投资者对整个半导体行业产生恐慌。然而，将英特尔的经营困境与英伟达、AMD等优质半导体公司混为一谈并不合理。英特尔在行业内属于低质量公司，其基本面与行业龙头的增长动能完全不同。

｜巨额资本开支的逻辑博弈亚马逊、谷歌等科技巨头近期宣布将大幅增加明年的人工智能资本开支，这引发了市场对自由现金流恶化的担忧。亚马逊在财报中披露，第三季度自由现金流出现大幅负增长，导致股价承压。但从长期看，亚马逊云业务AWS营收增速保持在39%，且订单积压规模增加了800亿美元。这种资本开支是为未来两到三年的高增长打基础，而非单纯的财务恶化，市场对短期现金流的过度反应提供了长线布局的机会。

｜半导体行业的季节性动能从历史规律来看，每年11月至次年4月通常是半导体板块表现较好的窗口期。尽管短期内市场可能在年底前继续震荡，但半导体行业整体增长逻辑依然稳固。英伟达目前的预期市盈率约为20倍，估值并不算高。对于具备高成长性且质量过硬的标的，当前的回调并非基本面崩塌，而是市场在消化短期预期差。

｜长期投资者的核心关注点面对科技巨头的波动，投资者应将视野拉长至5年。亚马逊、谷歌和Meta等公司在人工智能领域的布局已初见成效，虽然短期内面临消费者需求疲软和高额投入的压力，但其核心业务的护城河依然深厚。相比之下，英特尔等依赖传统PC市场的公司面临转型阵痛，不建议作为科技板块配置的首选。投资者应重点观察巨头自由现金流何时触底回升，这才是股价重回上升通道的关键信号。

接下来建议的操作指引： 1. 亚马逊（AMZN）当前估值已具备吸引力，建议以5年为周期进行长线配置，重点观察未来两个季度自由现金流是否如预期般回正。 2. AMD（AMD）作为半导体领域的高成长标的，在当前回调区间可保持关注，重点观察其在2027年前的市场份额扩张速度。 3. 避开英特尔（INTC）等处于转型困境的低质量标的，即便股价已腰斩，其基本面修复仍需数年时间，不建议在当前阶段进行抄底。

[0:02] 科技股出现大面积抛售，纳指进入回调区间 [2:04] 强调应对市场波动的心理准备与长期投资心态 [6:05] 分析科技巨头增加资本开支对现金流的影响 [9:07] 拆解英特尔业绩疲软对半导体板块的负面传导 [11:09] 展望11月至次年4月的半导体行业上涨动能 [13:13] 评估英伟达等头部公司当前的估值水平 [16:15] 探讨英特尔在行业竞争中的弱势地位 [26:24] 筛选科技巨头中的优质标的与投资优先级 [29:27] 讨论亚马逊财报后的市场预期差与长期价值 [34:34] 总结亚马逊云业务的盈利路径与未来增长点

[Source](https://www.bilibili.com/video/BV1QZxC6kEWi)

中文简报

英特尔业绩暴跌拖累芯片股科技巨头扩资本开支显露AMD买点市场将英特尔内部经营溃败与半导体周期混为一谈属于非理性恐慌，云巨头激增的算力开支令深度回调后的AMD与亚马逊成为未来最具吸引力的核心资产。面对纳指遭遇10%的技术性回调以及芯片股大跌，创作者系统拆解了恐慌的根本成因：科技巨头激进上调AI资本开支引发短期自由现金流恶化的担忧，叠加英特尔疲软业绩指引带来的行业性非理性误杀。创作者强调英特尔面临的是十年级别的结构性份额流失，而不能代表整体半导体景气度，具备高算力需求的AMD与估值处于20至24倍合理区间的英伟达依然拥有强劲的成长跑道。在七大科技巨头中，创作者将亚马逊评为当前风险收益比最佳的核心标的，并结合分析师观点指出AWS接近40%的营业利润率与创纪录的经营现金流足以为未来的巨额投资托底，建议投资者保持五年维度的长线视角，切勿在恐慌中盲目抛售优质资产。纳斯达克100指数ETF从高点回撤10%进入技术性回调，英特尔暴跌超31%、AMD跌12.5%，半导体板块遭遇系统性非理性抛售。 (0:00) 本轮恐慌核心在于英特尔PC端指引疲软引发连锁误伤，但科技巨头（亚马逊、谷歌）反而宣布大幅提升2025/2026年AI资本开支，算力底座需求并未衰退。 (6:05) 半导体板块预计在11月至次年4月迎来强劲反弹，博主看好AMD在未来24至36个月实现股价翻倍甚至三倍，成长性显著优于英伟达与英特尔。 (10:00) 科技巨头中亚马逊被列为长期首选买入标的，AWS营业利润率维持在近40%且营收增速达39%，短期自由现金流承压主要系Q3资本开支前置所致。 (27:07)

Reading summary

创作者认为，近期科技巨头财报后的抛售主要由科技股激进资本开支担忧与英特尔业绩暴雷诱发的错误恐慌所致；半导体基本面与云巨头算力需求依然强劲，当前优质芯片股与科技龙头的深度回调呈现出显著的长期不对称投资机会。市场恐慌根源：巨额CapEx与英特尔暴雷 (6:05) 本轮科技板块剧烈回调呈现出两组相互矛盾的宏观逻辑。一方面，亚马逊与谷歌在最新财报中均释放了明年大幅追加资本支出（CapEx）的信号，引发华尔街对科技巨头自由现金流恶化、高额举债或股权稀释的深切担忧。但正常情况下，超大规模数据中心的算力与基础设施投入本应利好半导体供应链。然而，英特尔发布的疲软PC端业务指引及其长达十年的竞争退化打乱了这一逻辑。美银证券分析师Vivek Arya指出，由于企业整体IT预算相对固定，激增的AI Token消费正在挤占传统PC和客户端硬件的更新换代支出。市场错误地将英特尔的自身经营败局（类似零售业破产边缘的Kmart）泛化为整个芯片周期的全面萎缩，导致AMD、美光等基本面优秀的企业遭遇非理性抛售。半导体板块分化与高胜率标的选择 (10:00) 面对半导体板块调整，创作者预计未来4至6个月（约11月至次年4月）板块有望迎来强劲反弹，真正的行业周期性拐点距离现在仍有2至3年。英伟达虽然增速将随基数抬升而逐步常态化，但其远期市盈率仅为20至24倍左右，并未透支过多估值泡沫。在具体的标的选择上，创作者明确主张避开英特尔等处于结构性衰退的低质量资产，并将核心资金集中于高确定性赛道。存储芯片领域更倾向于美光而非闪迪；在先进制程与算力芯片方面，创作者更偏好AMD而非英伟达。创作者判断AMD在数据中心与企业级市场将持续侵蚀英特尔份额，到2027年有望实现更高的营收与盈利复合增速，具备在未来24至36个月内实现股价翻倍甚至两倍增长的潜力。七大科技巨头评估：亚马逊与谷歌的配置价值 (27:07) 在美股“科技七巨头”中，创作者将特斯拉列为风险回报比最差的一档，苹果缺乏增长催化剂，微软则被视为年化回报约13%–19%的稳健复合增长标的。综合当前估值与增长前景，亚马逊被评为风险回报比最佳的首选标的，Meta次之，谷歌位列第三。谷歌云录得103%的强劲增长，展现出远超华尔街悲观预期的韧性。而针对亚马逊，Evercore ISI互联网研究主管Mark Mahaney指出，其财报引发的下跌本质是预期落差而非经营崩溃：AWS营收增长39%（连续第六个季度加速），广告业务增长12%受高基数扰动但后续有望回升，负自由现金流则主要源于三季度资本开支的前置投入。随着AWS维持近40%的高营业利润率以及整体经营现金流创新高，其巨额投资有望在2028至2029年形成强劲的正向自由现金流回收。心态管理：以五年周期应对短期消费逆风 (2:04) 创作者结合18年市场经验与美式橄榄球经历指出，成熟投资者应将纳斯达克10%的技术性回调视为常态，并对市场全面崩盘（预估未来12个月概率为25%）做好充分的底线心理准备，建仓时切忌幻想精准抄底。当前科技股出现极端分化，消费与防御类资产表现出较强韧性，说明并非全面的宏观流动性踩踏。考虑到终端消费疲软等宏观逆风正在拖累零售与广告增长，创作者提醒投资者不要对大型科技股抱有6至12个月内快速暴涨的幻想，而应以5年投资视角去审视其商业落地节奏。除AMD具备短期迅速反弹的爆发力外，大型科技股更需要耐心持有以穿越资本支出的密集投入周期。当前市场因资本开支激增与英特尔的颓势而过度惩罚了半导体与云巨头，但云厂商对计算集群的刚性投入仍将直接兑现为AMD等核心供应商的业绩支撑；投资者的核心分歧依然在于巨额AI资本开支转化为实际自由现金流的确切时间表，以及消费疲软背景下企业长期投资回报率的兑现节奏。

精炼阅读摘要

节目创作者认为，近期科技巨头财报引发的市场抛售与英特尔暴雷直接相关，但将个别企业的经营困境错误放大为整个半导体周期的见顶是严重误判；随着云服务巨头持续追加资本支出，优质算力与存储芯片标的在未来六个月及更长周期内具备极佳的风险收益比。英特尔业绩暴雷诱发的非理性蔓延 (9:07) 英特尔股价暴跌超 31%，并给出了疲软的 PC 市场需求指引与存储价格压力预警，导致美股半导体板块遭遇全面抛售，AMD、美光科技等标的跟跌。美银证券分析师 Vivek Arya 指出，企业固定的 IT 预算正被激增的 AI 算力与代币消耗挤压，导致传统 PC 等客户硬件更新周期放缓，而传统 PC 硬件仍占英特尔约一半业务，其自身数据中心业务增速也显著落后于竞争对手。创作者强调，英特尔过去十年在数据中心与 PC 市场持续被 AMD 抢占份额，其毛利率、每股收益与自由现金流恶化属于自身长期执行不力。市场将英特尔的挫折当作整体半导体行业恶化的信号并无差别抛售优质公司，其逻辑就像因为某家经营失败的廉价折扣零售商客流惨淡而抛售开市客和沃尔玛一样荒谬。云巨头资本开支与芯片基本面的背离 (10:00) 当前市场的恐慌情绪与底层产业的基本面形成了强烈反差。亚马逊、谷歌等云厂商在财报中明确释放出 2025 年及 2026 年将大幅追加资本支出（CapEx）以扩张 AI 算力基础设施的信号。华尔街虽因巨头自由现金流承压、甚至可能短期陷入严重负现金流而对巨额债务及潜在股权稀释感到担忧，但这些支出直接转化为对 GPU、服务器算力与高性能存储芯片的刚性需求。从产业周期来看，创作者预计整个半导体超级周期的顶峰仍在两到三年之后。AMD 拥有未来三至四年的高能见度业务扩张期，存储芯片厂商具备一至两年的强劲需求，英伟达亦有一整年的大规模扩张空间且其远期市盈率仅在 20 至 24 倍之间。创作者判断，随着 11 月至次年 4 月季节性与基本面发酵，优质半导体标的有望在未来四到六个月内迎来大幅反弹。半导体分层配置与七巨头投资顺位 (25:35) 在资产配置层面，创作者主张严格区分标的质量，彻底规避英特尔这类结构性失势公司以及二线存储厂商，在芯片核心主线上集中于优质资产。相比市值巨大、业绩增速基数变高的英伟达，创作者过去一年更集中配置 AMD，并预测随着 AMD 在 2027 年前持续获取市场份额，其股价在未来 24 至 36 个月内具有实现两到三倍上行的潜能。在美股七巨头中，创作者将亚马逊列为首选买入标的，其次为 Meta 与谷歌，而将特斯拉列为性价比最差标的。尽管谷歌三季度财报发布后遭遇市场抛售，其谷歌云业务仍实现了 103% 的高增长；微软则被视为未来四到五年有望实现 13% 至 19% 年化复合收益的低波动稳健标的，苹果则缺乏明确的增长催化剂。亚马逊资本支出周期的长期回报路径 (30:00) 针对亚马逊财报后的下跌，Evercore ISI 互联网研究主管 Mark Mahaney 指出，本轮回调本质上是预期层面的修正：AWS 营收增速虽连续六个季度加速至 39%，但略低于市场低买保底的 40% 预期；广告业务受高基数影响增速放缓至 12%；自由现金流承压则是由于资本开支在前三季度提前集中支出，全年总体指引并未上调，第四季度现金流预计转正。创作者与 Mahaney 均认为，投资者需拉长至五年视角审视大型科技股，不可因短期消费端环境疲软而对短期爆发性反弹抱有过高预期。AWS 维持在近 40% 的高营业利润率，加之集团整体创纪录的营业利润与经营现金流，奠定了强劲的内部造血能力；随着 Project Kuiper 卫星宽带项目在年底前转入商业化并允许支出资本化，巨额基建投资预计将在 2028 至 2029 年转化为显著的自由现金流复苏。本轮科技股重挫是由英特尔暴雷外溢、巨头巨额 CapEx 引发的现金流担忧以及消费端宏观逆风共同催化的情绪性出清。然而，超大规模云服务商向 AI 基建转移资本的趋势并未动摇，半导体优质龙头与 AWS 支撑下的亚马逊基本面依然坚挺；投资者应保持对回撤的平常心并拉长投资期限，避免在恐慌中盲目抛售核心资产。

Source quotes

[20:00] 浑水摸鱼清源: AMD has been eating Intel's lunch for years. [10:00] 浑水摸鱼清源: all the semiconductor stocks are crashing right now because of Intel? Doesn't make sense. [27:07] 浑水摸鱼清源: I would say the most attractive risk-reward right now is... I've got to give it to Amazon.

Original transcript

[0:00] For me to make a video on a Friday, you know Flapjack's got to be falling straight through the floor. And that's what we have going on out there, ladies and gentlemen. Look at the damage. Intel stock down 31% plus. sand disc down 16% here day a d down 12 and a half percent uh core weaves down 54 like it's just obliteration out there in the tech world right uh officially queues just hit a uh correction territory down 10 so we're gonna have a lot to go through in this video here today ladies and gentlemen okay uh yeah busy days for a friday so we'll talk about you know, stocks. I want to talk about some of the best opportunities out there in the market, in my personal opinion, things like that. I also want to explain why this is happening, right? Why is this occurring? Because we have multiple things going on right now that's sending all these stocks down. So I want to explain that in detail. So I want to react to this video here. I think that's going to be a really important one to react to. And then I also want to react to this one regarding Amazon and Mark Mahaney's view on Amazon. So I want to react to that one. Obviously, in my opinion, is an incredible opportunity out there for long-term investors. So lots to go through in this video here today. One thing, one thing I only need from you guys, just smash that like button if you haven't already done so. Hit that little thumbs up icon, it lets me know you enjoy me covering a Friday video for you guys. I almost never do a Friday YouTube video, it's extremely rare. Usually on Fridays, I kind of do a half day essentially. I'll post for my Patreon, I post a video for my private group and call it a day usually, Monday through Thursday, some days I'll work 12, 14 hours. So a lot going on. Okay, so a few things to... Let me just say this because I think whenever you're in one of these sorts of markets... Uh, people are looking for like, uh, me to be a therapist a little bit. I always see the comments like, you know, and this has always been like this, like, you know, when, wherever we're going through a correction or a crash in the market, people are like, you're my, you're my therapist, right? Um, let me tell you my view on this market. Let me tell you my views on a crash in those sorts of things. Um, and just kind of like where my head's at, right? Um, and, So, Let me say this. When I played football, I played tackle football for five years. So tackle football, obviously a very physical sport, and it's a sport in which you can never underestimate your enemy. You can never underestimate your opponent. You underestimate your opponent. That's how you could create a game where, you know, my team, we were generally like almost always like the favored team to win, essentially. go into a game and think like this is an easy game or this is going to be a lock in the park for us or you know we're going to get to play all the backups in this game we always had to take it like like this is the hardest opponent ever and the reason you have to do that mentally is so you prep for the worst case scenario so then you know if it ends up actually being a tough game you're not caught off guard and you don't panic so the same thing i've you know applied in football is same thing I apply to the stock market nowadays, I always kind of prep for the worst case scenario. So when I'm greeted with, you know, a NASDAQ down 10%, to me, that's to be expected, right? Like that's to be expected mentally. I'm also preparing for much worse. That does not mean I'm expecting much worse, but I'm preparing for worst case scenario. I'm preparing for a full on market crash. right? Mentally preparing for that. I'm always, that way I'm never caught off guard. I'm never in like a panic situation, right? I've been in the market 18 years now. And I know everybody always laughs because they're like, you know, you always bring it up how long you've been in the market and those sorts of things. But I say that because I am of the understanding that, you know, if you're in your first few years of investing and you're going through a Nasdaq down 10%, that could be a big deal for you. You could be scared. You could be like, oh my gosh, you know, having FOMO about like this is the greatest deals ever and those sorts of things, right? When you've been in the market 18 years, I've seen everything. I've seen at least three crashes in the NASDAQ, down 30% plus in my 18 years of being in the market, right? Seen several large corrections. I've seen a ton of small corrections, like what we're going through right now, a 10%-ish down for the NASDAQ. And I've seen a million pullbacks in the 18 years I've been in the market, right? So for me, as an experienced investor, this is pretty easy to get through. This isn't even like, like this is like you know let's say you're a great long distance runner and you got you know you're used to running 15 miles and you know you're running like two miles you're like this is a walk in the park that's what this is to me but I'm of the understanding this is not a walk in the park for everybody because if you've never been [5:00] through this. And all of a sudden, you know, this stock you hold is down 20% from its all time high. You're like, oh my gosh, right? The moral of the story is here, whenever you get these sorts of opportunities in the market, you've got to take full advantage. You got to take full advantage and don't think you're buying the bottom. Some people think they're buying the bottom. Don't assume you're buying the bottom. Do not do that. Assume it might even go lower. If you're going out there and you're buying... And let's say you want to buy AMD stock, okay? And AMD stock's trading at whatever it's trading at today. AMD's at, you know, $494. So you're like, okay, I want to go buy AMD. Assume you're going to buy it at $394 also. You might not, and you probably won't, but just like be mentally prepared for what if that happens, right? Next up here, let's talk about why is this occurring? What is going on here in the market, right? Okay. Let me explain this, okay, the best possible way I can, okay. By the way, let me let you guys know about something before we talk about if a market crash is coming, all that good stuff. The 1,000X sale is now ongoing, the Halloween sale. I got the link early, so I'll put that as the pinned comment down there here today for all you guys that want access to 1,000X in detail. We'll send you your steel membership cards over once you join us in there, all that good stuff. That'll be pinned comment. sale is ongoing. Okay. Um, literally just started. So, um, You know, what are my views about a crash in the market? I would call it lower probability. If I had to put odds on a crash in the next 12 months, let's say we already started, I'd put it at like 25%. Which you might view as a high, but actually that's not a very high probability, right? There's still three times higher probability we don't have a crash than we do, right? You know, why is this occurring? Intel. Intel is a huge component of why this is occurring. Amazon and Google is another huge reason. And let me explain both these differently because they're actually both contradicting each other, which is actually fascinating what's going on right now in the market. Okay. So. Google and Amazon said something very scary. They're talking about... substantially increasing capex next year. in a major way. That's scary. These companies are already going huge negative free cash flow. and their numbers are going to pick up over the next several quarters. And now they're talking about next year. They're taking it to another level. And, you know, not like there's going to be a modest CapEx increase. They're talking about, like, substantially, like, you know, like – basically like guiding to like, we're going to open it in a big way again next year. Right. That's very scary because you're like, how negative free cash flow these companies are going to go? How much... are they going to have to dilute shareholders? to make this happen. How much debt are they going to need to take out to make this happen? In Amazon's case, they might have to take out fortunes of debt, which has to then you got huge interest expense and that debt needs to be paid back at some point in time. Right. So now you can be looking at Amazon as a massively debt laden company within the next two to three years. Right. With a company like Google or maybe even a matter of some of these other companies, they're going to dilute shareholder value. So now we're looking at these companies that used to take shares off the market and be massive. you know buyback companies now we're looking at these companies and they're turning into companies that are now going to be diluting shareholders right and issuing more shares and so your ownership percentage ends up shrinking instead of increasing when they do buybacks, when they dilute your, your ownership stake is shrinking actually. Right. And so, That's a negative dynamic out there. But you would usually look at that and you say, "Okay, well, that's great news for semiconductors. That sucks for all those big tech companies that are going to spend crazy amounts next year." But hey, semiconductors, We're good, baby. You own AMD, you own Micron, you own Sandisk, any of these semiconductors are good, right? The issue happened with Intel. So, Intel. First off, let me say this, okay? Intel's a hunk of junk of company, okay? It's a real piece of junk. It's down 55% from its highs, by the way. But, you know, anybody that's been tracking the semiconductor industry over the last several years... You know Intel is not a good company. It's not a quality company. But what happened... is Intel just scared everybody in the semiconductor space. You know what happened? They said PC demand is going to be slower. And they're saying basically that business is going to be light. Right? [10:00] in. That scared everybody. Because now people are assuming, oh boy, Intel, they cited memory prices being an issue. Intel's looking light. Oh boy, that means AMD is going to be light. which then that means the memory chip stocks are going to be light on revenue and maybe even EPS, right? And so then you start running to all these conclusions... Because of intel? Because of Intel? But that's what's going on right now. So you've got a very weird dynamic where these massive tech companies Just hold you. They're going to have to up CapEx another major way next year. And yet all the semiconductor stocks are crashing right now because of Intel? Thank you. Doesn't make sense. So-- The thing I would say, like, do I think semiconductors... are gonna crash more or do I think they're going to boom? for the next six months. Let's just call it November through April. My belief is They're going to go through a ripper rally the next six months. Now, I'm not going to play short-term call option bets on these companies. I'm not going to buy extra shares of any of these companies. But I would say the November through April set up, and keep in mind, a lot of times, The market bottoms right around Halloween. Now, also I want to bring up 2018 because we could be in a 2018 dynamic where everything just keeps tanking through November into December and we don't bottom until Christmas. Okay? Okay. or around Christmas. That's always possible, but that's a lower probability event. So I would say the higher probability is semiconductors go through a reper rally in November, through April, and it's party times. And you look at these stocks, you know, four to six months from now in all their stock prices are substantially higher than where they're at right now. So, The sell-off in semiconductors doesn't make sense. I'll say that, right? And by the way, all these stocks, there's going to be a day when these stocks all do sell off, and it does make sense. Because it's like, ooh, the semi-trade is going to start slowing in a major way. Numbers are going to get tough for these companies. That's still two to three years away. We're not even close. I'm not even close. So right now we're still in these companies have to up CapEx in a major way and they have to spend every dollar coming in. Right. AMD's got basically three to four years of massive growth ahead. Almost like guaranteed. The membership stocks, they've got one to two years of massive growth ahead, basically like guaranteed. Right. Nvidia has at least one more year of massive growth ahead right so all these semiconductor stocks by the way are gonna be at different stages and The good thing you could say, well, if video only has one more year of massive growth ahead, that's scary. Uh, not really. You know why? I'll explain why. Nvidia doesn't trade at a rich valuation. Nvidia's forward P is probably like 20. Thank you. There's like 20 people. Like a 24 PM NVIDIA. So it's not scary. If NVIDIA was trading at a hundred forward P right now, it'd be scary to be like, oh crap. NVIDIA has probably only got one more year of massive growth and then they're going to have, you know, slow growth rates after that. Ooh, boy, that's scary. And maybe margins going down after that. That would be scary, but when you're trading at a 24-P, it's not scary like that. That doesn't mean the stock can't go down a year from now or six months from now or something like that, but it takes a lot of the scariness out of it because the valuation isn't up there. So something to keep in mind there. So you're having obviously substantial damage out there. Intel down 31% here. Sandisk down 14%. AMD down 12%. Core Weave down over 10%. there. And you down almost 7% here today. I mean, yeah, there's a lot of damage out there. Certainly, right? Especially with big tech stocks. If I go up and pull my personal stocks up here, right? I mean, you know, I got some greens today. Celsius is green. Celsius the wealthiest. Nike's green. Wynn's green. Cake's almost green. That's crazy, man. Cake is just extraordinary. Look at that stock. Almost green. on a day when stocks are getting absolutely obliterated, right? Um... Estes, louder, say it louder for the people in Vax, hardly down. Netflix is hardly down, right? But it's really those semiconductor-related stocks. Honestly, it's getting hammered today, too. That doesn't make any sense. Honest. Honest, I told you guys, that stock exits this year $5 plus. I continue to believe way over $5 is where we exit the year. So on honest... So that one remains a big buy. RH having a little pullback here today but that one's been [15:00] Red hot. Service now, that's brutal, man. I mean, that stock got hammered yesterday and then to be down another 5% here today. That's no fun. Right? Salesforce is not getting hit nearly as hard as ServiceNow. It got hit yesterday, but not even remotely close to as much as ServiceNow. And then not really getting hit today either, 1%. So maybe people know what I know. I don't know. We'll see. We'll see what happens there. Right. As far as the housing trade, that is... Nothing going on too crazy. We got tech-specific problems. Tech-specific problems is what we have going on out there in the marketplace. I posted this on my X page. I'm not sure if you guys follow me on X. That's always linked in the description area down there if you use that. CoreWeave is now down 71% from its highs. Intel 55% down from its highs. SanDisk 52% down from its highs. Google McDougal 31% down from highs. Micron MU down 30% from highs. AMD 25% down from highs. Metas 21% down from highs. So both of those are... or in You know, those are large corrections. Basically, these 30 percent plus moves, those ones are all in crash. Well, these are in a crash. Micro and Google, SanDisk, Intel and CoreWeave, those are in what I call a mega crash. When you talk about 50 percent plus, that's a mega crash going on in those stocks. Amazon 16% move and the Q's is at 10% there. Okay, let's go ahead and react to this Intel video. We'll talk more about this and what this means for the market, and then we'll get into amazing zone. Right now is B of A Securities Senior Semiconductor Analyst Vivek Arya. Vivek, let's take a look at this. The stock is off 25% this morning. Obviously still up over. Oh, hey, listen. It went down more, 31% actually. your rating on this company to a neutral from a buy. I guess that means that even with this 25% drop, you wouldn't tell people to wade into the water at this point. What happened? Yeah, thank you, Becky. Good morning. So I think the AI is not helping all boats rise at the same time. You know, there are parts of the market that are doing extremely well that have to do with AI token demand, with AI CapEx, right, with cloud service growth. But what that is doing is that is creating a squeeze on the supply that is required for other parts of the market, such as consumer PCs. and smartphones and i think that is what intel is having to grapple with that its data center business is doing extremely well, but parts of its business that have to deal with consumer and the PCs is not doing well. And I think that is just a tough mix for the company, which has to invest in its factories to help attract external demand over the next few years. So I think it's a multi-quarter issue, which is why we thought the best thing is to wait on the sidelines and see how it makes progress consistently. in getting external customers or improving its data center business. Yeah, it's a big deal for the PCs. I mean, that's still 50% of their business, isn't it? Yeah, you know, PCs is kind of the core of the company. It is an important part of their production. You know, there's a lot of IP that they share between the PC and the data center market. So, you know, it's not all, you know, being wasted. But the problem is that the exposure to the PC market is still larger than their exposure to the data center market. And within the data center market, the good news is that the demand is off the charts. I mean, their data center business grew the fastest in the last 15 years. It was up almost 60% year on year. It's still not growing as fast as what we see from AMD and Nvidia and a lot of the ARM. based suppliers. But it is still growing pretty fast. The problem, of course, as you mentioned, is that it is not able to offset the weakness and the sudden drop in demand, you know, because IT budgets are not growing as quickly. So if they're having to invest more in AI, that means they have less money to invest in upgrading their PC infrastructure. You mean their corporate customers are no longer upgrading in the same sort of cycle and fashion that they did because they're taking so much of their budget and pushing it towards AI, the new technology that's out there? Absolutely. You know, there is kind of a fixed buy that customers need to spend. So a lot more is going into spending on AI tokens. As an example, which is why we see off-the-charts growth in AI. We recently heard from Amazon, Google, Microsoft, Cloud business is up between 40% and 100%. Token demand is up by a factor of 10x. CapEx intentions are very strong. So I think that part of the business is doing very strong. The problem is Intel is not as exposed to that part of the economy. It is a lot more exposed to the part of the economy that is sort of on the other side, which is getting squeezed. And I think that's what we are seeing in their numbers and the stock performance right now. Vivek, we've always looked at these businesses as cyclical. [20:00] You're going to see ups and downs, not just with what you're talking about with PCs, but also with enterprise demand for AI and other things, too. There's this question right now whether the AI part is either no longer cyclical or just deals with much longer cycles. Do you think it would be right for Intel to chase the AI part of this, or do you think that eventually runs the risk of coming down where we've overbuilt? Yesterday, Gary Cohn talking about how we build out everything. Maybe we build this out too far too. should they stay with one foot firmly planted in the the PC business, the consumer business, the rest of this? Do you want to see them chase AI? What's your... hope as somebody who follows and recommends this company. Don't... I think, Becky, we have to differentiate between demand and near-term economics. And from that perspective, the story of what we are seeing in the broader market in AI is very similar to what we are seeing at Intel. So if you look at the broader market... Demand for AI, as I mentioned, is off the charts. The problem, though, is that we have to spend to create this infrastructure for growth that might come over the next few years. So there is kind of a mismatch between demand. How do I say it as simple as this? Listen. A-M-D. has been eating Intel's lunch for years. Like, the last decade. They've been taking massive amounts of market share, right? AMD's been owning Intel. And it remains like that to this day. And so Intel is just, you know, they're just a very weak company. Very weak company. They've been a weak company, right? Like, look at... the charts here on thousandxstocks.com for Intel, right? Does this look like a strong company to you? Look at this. Look at this. Right. Look at the margins. Does this look like a strong company? Look at the earnings per share of this company. Right. And this is what's supposed to be coming. Right. But look at this. Right. And let's be clear. Like the only reason Intel's even getting any lift in their business right now is just because you happen to be in a huge semiconductor boom where these companies need whatever you got. Right. If it wasn't for that, it would be even a lot uglier for this company. Right. You know, look at the free cash flows company over time. Does this look like a high quality company to you? going up up up up up everything you look at with this company is it's bad right it's just a bad company like there's no other way of putting it right um and so you know the fact that the market is freaking out about this and selling everything in the semiconductor space for a bad company a bad quality company It is ridiculous. It would be like, I don't know. Let's say... We're talking retailers. Okay. And let's say, I think Kmart went bankrupt a long time ago, but let's say Kmart was still a public company and like Kmart came in and they reported bad results. And so then Target stock got sold. Everybody sold their Target stock and everybody sold their Walmart stock and everybody sold their Costco stock. because Kmart came in with bad results. We all, we all be like, It's Kmart. Kmart is trash. We're going to sell off all the quality stocks because Kmart came in with bad numbers? What are we doing here? infusion of money into this company and then this huge semiconductor boom this company might be bankrupt right Like seriously, they might be bankrupt. They got saved. Let's be very crystal clear about that. It's amazing, man. The whole semiconductor stock's getting sold off today for Kmart. as Intel Kmart and then economics. I think it's the same situation or a similar situation that's happening at Intel. Their strategy is absolutely right, right? They are growing well in the data center. You know, they are investing in their manufacturing facilities. The problem is that investing in manufacturing facilities and getting the results from that is a three, four, five-year process. Whereas a lot of capex has to be front and loaded. So in some ways, I think there is kind of that mismatch between demand and economics. So I think Intel's strategy is right. It says that it's going to take time to pay out. But by the time they catch up, is the demand still going to be there for AI? This AI demand is here to stay, and it's not going to change over the next two, three years as Intel chases it. Absolutely. I think it's more than just AI demand. I think Intel is an incredibly important source of U.S. domestic manufacturing. We absolutely need diversification and chip supply. We cannot just be dependent on a handful of suppliers in Asia, as much as they are very critical and they are investing in the U.S., but Intel has to be successful as kind of the leading domestic supplier. the night and very [25:00] Early stages of that. So I think the strategy is right. I think the demand will absolutely be there because there is demand for leading edge technology manufacturing, but it's getting harder for them to match the strength they are seeing today with all the investments that have to come before they realize a lot of that demand. You know, I really want to be in top-tier companies if I'm going to be playing... You know, in different spaces. And so what do I mean by that? Like a company like Intel, why would I want to be in Intel? If I'm going to be in something semiconductor related in the space, I want to be to be in NVIDIA or AMD and that's it, right? And so obviously you guys know I put my chips on AMD instead of NVIDIA, right? Which has been a great decision because NVIDIA has kind of been, you know, just... stock let's call it like this stock was over 200 you know around this time last year and you still look at the stock right about 200 bucks right meanwhile amd has been flying and amd's growth rates are going to be much faster in 2027 as a whole than than video right but with that being said if you told me you know i have to buy something in this space like those are two i want to be in broadcom's good company right I don't want to be in Broadcom. Intel, trash company. I don't want to be in Intel. Let's say we wanted to play something memory related. I'd much rather be in a Micron versus a SanDisk or something like that. I'm always going to try to look for a higher quality. than a lower quality in these sorts of environments, right? I'm going to look for the more growth, but a high quality company at the same time. And so Intel is just not high quality. It's very, very low quality. We call it LC. We call it Kmart. Okay. Let's talk about some amazing Zon. And then you know what I'll do? I'll tell you what my favorite MAG7 stocks are to buy right now. I'm least interested in. - It's aggressive AI spending plan, so that has not changed. The company continuing to project-- - No, no, no, let's do it right now. Let's do it right now, before we get into this video. What's my... What's the best buy in MAG7 right now for a long-term investor? What's the worst buy? Worse by... You know, Tesla, least interesting of the bunch by a large margin. Best, I would say... . It's tough. is really tough right now because valuations have gotten more compelling I would say the most attractive risk-reward right now... is... I've got to give it to... Amazon. It's tough though. It's very tough because there's a, the close second rate behind it is meta. And the close third behind it is Google. Like those companies are like, All three of them I think are really interesting buys. Google really got my attention with those latest quarterly results. Wall Street walked away from that very negative. I walked away from those results for Google feeling the best I felt about Google in quite some time. Obviously, it's a smaller position for me in the public account. I walked away. I walked away from those results feeling even better about Google than Meta and Amazon. Right. And that doesn't mean I feel bad about I mean, meta, I kind of feel a little bad about that one's just like the spends out of control and they just don't have the business lined up to really generate these crazy growth rates. Right. We're hoping like Muse can be something big someday, but we don't have any definitive proof. With AWS, at least you've got the growth rates, right? Like they're putting up the growth rates, 39%. They're talking about 40% plus next year and the year after. Like you got something there, right? Something real. Google, the growth rates are insane for Google Clouds. 103% latest quarter, it's got the growth, right? So... Yeah, you know, outside of that, Apple, it's not super interesting. I don't think it's horrible. I think Microsoft's actually a pretty interesting buy. We ran valuations on that. It probably... I would say it likely produces a teens. a teen's return per year for the next four or five years. So let's call something between like 13% and maybe 19% returner. So that's good, solid. It should beat the market over the next four years. Um... And then who else? NVIDIA is a decent buy, but I don't love it here. You know, I would still much rather be an AMD. Like AMD has real potential to double and triple in the next 24 months to 36 months. AMD double and triple in the next 36 months. Very unrealistic in my opinion, because we're going to have those much slower growth rates. starting at the end of next year and then moving into future years. So, okie dokie. Let's [30:00] Go ahead and react to Wall Street. That the capital expenditures have 220 billion dollars for the year that's a forecast that sent the shares lower and joining us right now to break down some of those numbers mark mahaney is evercore head of internet research Good morning to you. I wasn't that surprised by their numbers, Mark, but I'm surprised that the market's taking it the way it has, given maybe the way... They reacted to some of the Google numbers, but the Google store is kind of different. What do you think is going on here? I think you're right. There's a big difference between Google results and then meta results and these results, but all three of them had pretty strong negative market reactions. With Amazon, it's three things that... The market wanted 40% plus low 40s AWS growth, and it only got 39%. It accelerated, but not by as much as the market wanted. The market got 12% ad revenue growth, and the market had wanted something more like 18%, 19%. So that was a disappointment. expectations correction not a fundamentals correction aws revenue growth accelerated for the sixth straight quarter uh in a row and then they laid out these goals for the next two years of 40 plus aws growth at that scale i think that's really impressive and the margins were good at aws advertising just had some comp issues and they went out of their way to kind of normalize the ad revenue growth rate and pretty strongly suggest is going to get back closer to 20 high teens in the next quarter. And then the negative record free cash flow, well, that's because the CapEx got pulled forward. They didn't change their CapEx for the full year. It's just that it was a lot heavier in Q3 than the street had expected, which actually means Q4 is going to have positive free cash flow. So, you know, look, this company's still going through a major investment cycle, but I look at this as an overreaction. I like Amazon's stock. Yeah, it's a number one pick, and it's my number one buy, even here off the correction. I don't like corrections on my top picks, but it's happened, and I understand why. I just think there are answers to all three of those major concerns. Okay, so 226 bucks. If we're sitting here together a year from now, what do you think it's worth? Well, I think this thing gets back to doing like 25 times earnings. I think what the market needs to see are two or three things. We need to see the light at the end of the tunnel in terms of this free cash flow hole that the company is going to go through next year. I think it's going to be about $50 billion in negative free cash. This is where I think everybody is making a big mistake. Big mistake with these big tech companies, okay? Big mistake, listen. So don't assume like these companies are going to like just their stocks and go crazy over the next year, right? It's possible, but I wouldn't be super bullish on them necessarily in the very short term, meaning the next 6 to 12 months. Is it possible Amazon, you know, over the next 12 months goes to 300 plus and Meta goes to 900 and blah, blah, blah? It's possible, but I'm just like these companies do have a lot of headwinds, right? They got the economy, which the consumer is in a weaker spot. We know that. Look at the consumer sentiment. Look at the company numbers. We're just seeing a lot of weak numbers across the board. For anything that's really dealing with the consumer, the AI-related stuff, like business spending is fine. But anything consumer related looks weak right now, very, very weak, right? So that's something to kind of keep in mind there and and so I uh... You know, I would say... don't get your hopes up that these stocks are going to go blasting higher in the short term. Like, I'm just like, you know, like, You really need to be thinking about these sorts of companies in a five-year span. If you're thinking about it just over the next 12 months, you could be set up for disappointment. I'm not getting my hopes up around any of these stocks, Amazon, Meta, Google, any of these big techs. The only one I would say I have my hopes up on in that I think could blast higher in the shorter term is AMD. Outside of that, no. There's one coming, it's launching, and that's Project Leo, which is going to go commercial before the end of this year. That'll also actually be very beneficial for the P&L because they'll start capitalizing some of those rockets. What? that manufacturing expenses. So those are the two things I'm really looking for. But it's most important for I think for investors is show me when can I get confidence that the worst of the free cash flow losses is behind me. Yeah, and Mark, really, it's the street wants to know that the capex that's being spent right now. does have a way to profitability. True, Becky Quick. Last night on the earnings. [35:00] Andy Jassy said that he sees a clear line of sight to profitability for all of this. Do you? Yeah, and I think there's some evidence right in the numbers. I mean, the margins at AWS are close to 40 percent. And if they can do, they can keep margins relatively close to where they are today and grow 40 percent top line the next two years. I mean, you're going to get you're going to get back to substantial free cash flow, perhaps as early as 28 and very likely in 29. So, yeah, I think so. set of new deals that were signed up. So I just think you have a lot of visibility into demand and the track record that they're showing with their cloud business, the most important one is showing rising margins. By the way, the company as a whole gave you record operating margins last night. And I think those margins are going to continue to move higher in the next couple of years. You're at record operating cash flow. So it's really a matter of when the capex spend. Amazon stands for always a buy baby okay speaking about always a buy we got that thousand xcl ongoing right now that's for the entire Research platform. That's what I use on a day in day out basis. You get access to everything. And so make sure you guys take advantage of that. It's the last sale for any product I have. for 2026. And the sale will be over shortly. So I'll pin the comment down there to get access to that. We'll send you your steel membership card to your house once you join us in there. and uh yeah enjoy it guys all right much love and have a great day

英特尔业绩暴跌拖累芯片股 科技巨头扩资本开支显露AMD买点