Cloud Growth Wasn't Enough. Cash Flow Split Microsoft and Alphabet.
The Federal Reserve held rates as technology shares edged higher
The $Invesco QQQ(QQQ)$ gained 0.60 per cent on Wednesday, while the $S&P 500(.SPX)$ rose 0.28 per cent. The Federal Reserve voted 10–2 to keep its target range at 3.75 per cent to 4.00 per cent. The two dissenters preferred a 25-basis-point increase. The statement described economic activity as strong and inflation as still elevated, offering no clear signal of imminent easing.
The indices moved little during the regular session. The more important change came after the close. Microsoft and Alphabet both reported strong cloud growth, but Microsoft gained about 4 per cent in extended trading while Alphabet fell nearly 10 per cent. Investors did not compare cloud growth alone; they assessed company-wide earnings, capital expenditure and free cash flow together.
Microsoft beat expectations and disclosed quarterly Azure revenue
$Microsoft(MSFT)$ reported first-quarter financial-year 2027 revenue of $92.3 billion, up 19 per cent, with adjusted earnings of $4.89 a share. Both exceeded market expectations. Azure revenue reached $33.17 billion and rose 48 per cent, the first time Microsoft had directly disclosed Azure's quarterly revenue.
Capital expenditure reached $45.2 billion, including data-centre assets acquired through finance leases. Cash purchases of property and equipment were about $39.4 billion, while operating cash flow was $58.0 billion, leaving $18.6 billion of free cash flow. Investment continued to rise, but cloud revenue and cash generation were not completely overtaken by the build-out. That combination supported the positive share-price reaction.
Alphabet doubled cloud revenue as free cash flow turned negative
$Alphabet(GOOG)$ reported third-quarter revenue of $127.6 billion, up 25 per cent but slightly below expectations. Earnings of $2.12 a share also missed. Google Cloud revenue rose 103 per cent to $30.7 billion, while cloud operating income reached $11.1 billion. The cloud business itself delivered strong growth and profitability.
The problem appeared in company-wide cash generation. Quarterly capital expenditure rose to $51.4 billion, while operating cash flow was $42.3 billion, producing negative free cash flow of $9.07 billion. Alphabet also recorded an $8.7 billion unrealised loss on equity securities, which reduced net income and earnings per share. Doubling cloud revenue did not close the cash gap, so investors focused first on spending pressure and the earnings shortfall.
The difference was not the strength of AI demand
Azure and Google Cloud both accelerated, showing that corporate demand for AI computing, data and cloud services remains strong. The divergence does not mean that one company has demand and the other does not. It reflects whether revenue growth can cover construction costs and whether cloud progress can lift company-wide profit and cash generation.
Microsoft still produced $18.6 billion of free cash flow and exceeded expectations for revenue and earnings. Alphabet delivered faster cloud growth, but total revenue fell slightly short, free cash flow remained under pressure and the investment loss widened the earnings gap. Investors were not rewarding lower capital expenditure; they were rewarding the ability to retain cash after spending increased.
SK hynix surged as hardware demand reached reported results
$SK hynix(SKHY)$ rose 13.30 per cent on Wednesday, Micron gained 5.86 per cent and Sandisk advanced 7.66 per cent. SK hynix reported third-quarter revenue of Won104.6 trillion and operating profit of Won83.3 trillion, with HBM4 and server-memory sales helping it set quarterly records.
The storage rally shows that data-centre construction is reaching chip and storage suppliers' revenue. It also raises spending for cloud platforms: sales recognised by hardware suppliers often become capital expenditure for buyers such as Microsoft, Alphabet and Amazon. Stronger hardware demand therefore increases the need for platforms to prove that additional computing capacity can generate faster revenue and cash returns.
Amazon, Intel and Sandisk provide the next tests on Thursday
$Amazon.com(AMZN)$, Intel and Sandisk are due to report after Thursday's close. Amazon must show whether AWS growth can keep pace with data-centre investment. Intel needs to explain how server demand and manufacturing investment are affecting profit. Sandisk must update investors on NAND, enterprise SSD demand, inventory and margins.
Microsoft showed that high investment can coexist with positive free cash flow. Alphabet showed that even a doubling of cloud revenue may not produce cash immediately. The next reports will be judged on more than revenue beats; investors will examine the profit and cash remaining after expansion. AI demand remains strong, but the test for returns has become more specific.
This is personal analysis and not investment advice.
💬 【Talking Point】
What evidence would convince you that today's AI infrastructure spending can earn an adequate return?
💰 【Bounty】
Share your view and reasoning in the comments for a chance to earn Tiger Coins! 🎁
🔔 Better shared than saved — tag a friend and join the discussion!
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.

① 自由现金流持续增加
AI公司投入几百亿美元建设数据中心后,仍能留下更多现金,说明投资可能开始产生回报。
② AI收入增长快过成本
例如微软Azure增长48%,是好现象;但还要看未来收入能否持续增长,利润率是否改善。
③ 客户愿意持续付钱
如果企业不断购买AI服务,云业务订单增加,说明AI不只是炒作,而是真正有人愿意付费。
我的投资判断
微软(MSFT):收入、Azure和自由现金流表现较均衡。
谷歌(GOOG):云业务强劲,但要继续观察整体现金流。
存储芯片股:受惠于AI需求,但要小心价格上涨不可持续。
记住:AI需求强,不代表所有AI股票都值得买。
我会等待至少连续几个季度的证据:收入增长、利润率改善、自由现金流增加。 这比单季财报超预期更有说服力。
这里有一个值得警惕的现象:英伟达、美光、SK海力士等上游企业的收入增长,很大程度上来自微软、谷歌、亚马逊等云厂商不断增加的资本开支。
上游公司的收入是真实的,但不代表整条AI产业链已经实现了最终盈利闭环。 如果云厂商持续采购GPU、HBM和服务器,却无法通过AI订阅、API调用及企业服务收回投资,未来资本开支就可能放缓,反过来影响上游订单。
我会重点观察三个证据。
第一,AI收入增长能否持续超过相关运营成本增长。 不仅要看云业务总收入,还要关注AI服务的实际付费需求、单位推理成本和利润率。
第二,新增资本回报率能否改善。 假设一家公司新增投入100亿美元,每年只能产生3亿美元增量税后经营利润,即使收入增长很快,3%的简单增量回报率也未必能覆盖资本成本。
第三,自由现金流能否在扩张过程中逐步改善。 单季度现金流为负并不可怕,真正危险的是资本开支不断增加,而新增经营现金流长期无法跟上。
此外,还必须考虑GPU更新周期和折旧风险。如果AI芯片技术迭代过快,部分数据中心设备可能在充分收回成本前就面临升级压力,实际投资回报会进一步下降。
微软与谷歌的财报分化,恰好提醒投资者:AI需求强劲和AI投资划算,是两个完全不同的命题。
我的判断是,未来两年市场可能逐渐从奖励AI资本开支规模,转向奖励AI资本使用效率。
真正能够证明AI投资成功的,不是今天建成了多少数据中心,而是几年之后,这些数据中心还能为股东持续创造多少自由现金流。🐯