To The Moon

The market is currently navigating a "SaaS Reckoning" as of February 2026. The 22% YTD drop in the IGV ETF is not just a temporary sell-off; it signals a fundamental shift in how investors value the AI stack. We are moving away from "AI-at-any-price" toward a "Physical Bottleneck" phase.

$iShares Expanded Tech-Software Sector ETF(IGV)$

Will the rotation toward proven revenue continue?

Yes, but with a specific focus on "Tangibles." The rotation isn't just moving to "value" stocks; it’s moving toward companies that control the physical constraints of AI—specifically Power, Compute, and Infrastructure.

The Cannibalization Fear: High-valuation software (SaaS) is under pressure because "Agentic AI" (AI that performs tasks autonomously) is starting to cannibalize the "per-seat" licensing model. If an AI agent can do the work of 10 people, a company needs 9 fewer software seats, threatening the core revenue of legacy giants.

The Flight to Quality: Investors are punishing "AI losers"—companies with high R&D spend but no bottom-line lift—and rewarding those with massive, tangible cash flows. This has led to the unusual sight of Utilities and Consumer Staples outperforming tech during this correction.

$Utilities Select Sector SPDR Fund(XLU)$ $Consumer Staples Select Sector SPDR Fund(XLP)$

Who will lead the AI cycle in this "Cautious" environment?

The leadership has bifurcated into two distinct camps: the Hyperscale Infrastructure Spenders in the West and the Value-Driven Innovators in China.

A. The Western "Infrastructure Kings"

In the US, leadership is concentrated in the five companies capable of spending the $660–$700 billion projected for AI Capex in 2026.

Amazon (AWS): Projected to spend $200B in 2026. CEO Andy Jassy’s "monetizing as fast as we install" mantra has made AWS the anchor for infrastructure bulls. $Amazon.com(AMZN)$

Microsoft & Oracle: Microsoft’s $80B backlog in Azure orders (limited by power, not demand) and Oracle’s pivot to "industrial plumber" of AI clouds make them the "proven revenue" picks. $Microsoft(MSFT)$

Energy & Power Players: Because power is now the primary bottleneck (not just chips), companies like Vistra and Constellation Energy are being treated as de facto AI stocks.

B. The surging Chinese Tech Sector

Following the 2026 CNY, Chinese tech is seeing a "DeepSeek Moment." While the US focuses on proprietary, high-margin models, China is leading in cost-effective, open-source AI.

• The "Results-as-a-Service" (RaaS) Leaders: Companies like Bairong and Huawei are shifting the model from "buying software" to "paying for outcomes," which appeals to a cautious global market.

• Alibaba & Baidu: Their cloud divisions are primary beneficiaries of the "domestic substitution" policy, as China rapidly builds its own independent AI hardware and software stack.

• Hardware Surges: Domestic chipmakers and memory giants like YMTC and CXMT are entering 2026 with strong IPO momentum, filling the void left by evaporating Nvidia market share in China due to export controls.

Summary of Market Leaders

In the next section, we will look at the latest earnings reports from February 2026 reveal a staggering "Infrastructure Arms Race." The top five US hyperscalers are now projected to spend over $700 billion in Capex this year—a 60% increase over 2025.

However, the "Scaling Paradox" is emerging: while AI revenue is growing at triple digits, it is currently being outpaced by the sheer volume of capital being poured into the ground.

The US Hyperscalers: "Supply-Constrained, Not Demand-Constrained"

The narrative across Amazon, Microsoft, and Google is consistent: they cannot build data centers fast enough.

Amazon (AWS): * Capex: Projected $200 billion for 2026.

• AI Scaling: AWS sales surged 24% (its fastest in 13 quarters), reaching a $142 billion annualized run rate.

• The Proof: Their custom chips business (Trainium/Graviton) is now a $10 billion business, growing at triple digits. Amazon is betting that by building their own silicon, they can preserve margins that are otherwise being eaten by Nvidia.

Microsoft (Azure): * Capex: Hit $34.9 billion in a single quarter (Q1 FY26).

• AI Scaling: Azure revenue grew 40%, with AI services contributing significantly.

• The Bottleneck: CFO Amy Hood noted they are "behind" on supply and expect to be constrained until at least June 2026. This is a "good problem" for revenue but a "bad problem" for short-term free cash flow (FCF).

Google (Alphabet):

• Capex: The "shock" of the season, projecting $175–$185 billion for 2026.

• AI Scaling: Google Cloud revenue jumped 48%, and their cloud backlog doubled to $240 billion.

• The Yield: Google is successfully converting its "Gemini" ecosystem into enterprise contracts, but FCF dipped as Capex "ate the difference."

The China Factor: Efficiency over Brute Force

While the US is spending on "Brute Force" infrastructure, Chinese tech is leading a pivot toward Cost-Efficiency and Agentic AI, largely catalyzed by the "DeepSeek Shock" of late 2025.

• Alibaba Cloud: Revenue growth accelerated to 34% (as of late 2025/early 2026 reports). They are following a "high-utilization" model, where their Qwen open-source models have over 180,000 derivatives, creating a massive "top-of-funnel" for their cloud infrastructure.

• Tencent: Focusing on Internal Scaling. Their "Hunyuan" AI now drives over 900 internal scenarios, contributing to a 21% lift in marketing revenue. This proves AI can scale revenue by making existing businesses more efficient, rather than just selling "compute hours."

Summary: Scaling Efficiency vs. Capital Intensity

The following table summarizes the 2026 "Efficiency Ratio" (AI Revenue Growth vs. Capex Intensity).

The Bottom Line

The market is no longer rewarding "AI potential"; it is rewarding Backlog Conversion. Oracle and Google are winning on growth rates, but Amazon is winning on the "Self-Sufficiency" story by using its own chips to offset the massive Capex bill.

Appreciate if you could share your thoughts in the comment section whether you think it is time to rebalance our AI portfolio with some chinese names to navigate the capital intensive (high valuation) risk from U.S. names.

@TigerStars @Daily_Discussion @Tiger_Earnings @TigerWire @MillionaireTiger appreciate if you could feature this article so that fellow tiger would benefit from my investing and trading thoughts.

Disclaimer: The analysis and result presented does not recommend or suggest any investing in the said stock. This is purely for Analysis.