Amazon Bought the Dip. SpaceX Has to Prove the Cash Flow.
The market: Amazon led as QQQ rebounded faster than SPY
$Invesco QQQ(QQQ)$ rose 1.21 per cent to US$718.36 on Monday, $SPDR S&P 500 ETF Trust(SPY)$ gained 0.42 per cent to US$766.48, and the $S&P 500(.SPX)$ added 0.41 per cent to 7,690.73. Amazon rose 4.98 per cent, lifting technology and the data-centre supply chain and giving the Nasdaq-100 its best performance in almost a month.
The advance went beyond one large-cap rebound. Money spread from Amazon into memory, semiconductors and networking equipment, showing that Friday's selling had cooled. The questions around rates and investment returns remain. Results from SpaceX, Meta and Tesla will test whether AI spending is becoming revenue and cash flow. Monday put positions back on; earnings decide whether investors keep them.
SpaceX reports first tonight, followed by Meta and Tesla tomorrow
$SpaceX(SPCX)$ rose 1.32 per cent to US$147.96 on Monday, $Tesla Motors(TSLA)$ gained 3.76 per cent to US$368.06 and $Meta Platforms, Inc.(META)$ fell 0.55 per cent to US$705.44. The three platforms diverged as results approached. SpaceX reports on Tuesday, followed by Meta and Tesla after the close on Wednesday.
Consensus for SpaceX is earnings per share of US$0.1361 on revenue of US$12.97 billion. Meta is expected to earn US$6.0031 per share on US$64.69 billion, while Tesla is expected at US$0.4358 per share on US$28.44 billion. SpaceX must address AI compute contracts, Starlink growth and cash flow. Meta must show that advertising can cover capital expenditure. Tesla needs margins and autonomous-driving progress to support its growth expectations. Scarcity is already in the valuations; profit now has to catch up.
Memory: SanDisk led the rebound as Micron prepared to discuss its technology roadmap
$Micron Technology(MU)$ rose 2.97 per cent to US$904.90 on Monday, $SanDisk Corp.(SNDK)$ gained 5.53 per cent to US$1,371.12 and $SK hynix(SKHY)$ added 4.67 per cent to US$142.87. All three rebounded from consecutive selling and substantially outpaced the broader market, as investors reconsidered whether supply and cycle concern had been priced too aggressively.
Micron holds a technology webinar at 8am on 21 October. Chief technology and product officer Scott DeBoer will cover DRAM, HBM and NAND roadmaps and the bandwidth, capacity and data-transfer requirements of next-generation AI infrastructure. A roadmap and joint customer design work can reinforce long-term demand expectations, but an event cannot replace shipment and pricing data. HBM share, NAND pricing and customer qualification must improve together before this becomes more than a short-term rebound.
GTC Berlin opened as the chip rebound waited for a new source of demand
$NVIDIA(NVDA)$ rose 1.38 per cent to US$217.09 on Monday, $Marvell Technology(MRVL)$ gained 4.06 per cent to US$237.09, $Advanced Micro Devices(AMD)$ added 1.76 per cent and $Broadcom(AVGO)$ rose 1.78 per cent. The sector recovered broadly, but the difference in gains showed a greater willingness to buy back higher-beta networking and custom chips than to chase every semiconductor equally.
GTC Berlin runs from 20 to 22 October. Nvidia chief executive Jensen Huang speaks at 17:00 Beijing time on 21 October, covering accelerated computing, AI infrastructure, open models, AI agents and physical AI. Quantifiable progress in inference costs, enterprise deployment or robotics would create another source of compute demand. Product vision without orders and customer budgets would leave the rebound constrained by the same capital-return question.
CoreWeave ignored an upgrade while Nebius rebounded almost 3 per cent
$CoreWeave, Inc.(CRWV)$ fell 0.44 per cent to US$61.12 on Monday, while Nebius rose 2.88 per cent to US$170.24. Goldman Sachs upgraded CoreWeave to Buy with a US$106 target, but the shares did not follow, showing that investors still question the effect of leverage, depreciation and continued financing on cash flow.
The institution argues that funding concern is excessive: asset-backed finance does not move one-for-one with corporate bond yields, and higher revenue per megawatt can improve investment returns. Delivery, utilisation and contracted revenue must prove that case. Nebius's rebound shows investors are willing to distinguish between neoclouds. CoreWeave's muted response to good news shows that one rating change has not resolved the credit question.
This is personal analysis and not investment advice.
💬 【Talking Point】
Can one strong platform result turn an AI short-covering rebound into a durable trend?
💰 【Bounty】
Drop your view and reasoning in the comments and there are Tiger Coins in it for you! 🎁
🔔 Better shared than saved — tag a friend and split the coins!
Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.
我会重点关注三个方面：
1）AI巨头的业绩能否兑现预期？
Meta需要证明AI投资能够提高广告收入，特斯拉需要改善利润率，其他AI企业也需要证明巨额资本开支能带来实际回报。业绩好不代表股价一定上涨，如果市场早已提前反映利好，股价仍可能下跌。
2）芯片需求是否真正增长？
英伟达、美光和博通等公司的长期机会，取决于AI需求、芯片出货量、产品价格及客户订单，而不是单靠一场发布会或股价反弹。
3）AI公司的现金流是否健康？
像CoreWeave这类需要大量资金建设数据中心的企业，我会特别留意债务、利息支出、折旧和融资需求。收入增长很快，不代表最终利润和现金流一定理想。
我的看法： 一次反弹可能来自空头回补，但要形成持续上涨趋势，最终仍需要盈利和现金流支持。
作为投资新手，我不会因为市场反弹就急着追高，而是先看财报、估值和风险，再决定是否买入。
大家认为，接下来最能决定AI行情的，是英伟达的技术突破、科技巨头的盈利表现，还是AI基础设施的投资回报？