Is Alibaba a Buy on AI and Cloud Upside Potential?

Alibaba reported revenue up 3% to $35.3 billion, but adjusted EBITDA plunged 61% to $2.4 billion and adjusted EPS fell to $0.01 as the company ramps spending on quick commerce and AI. Cloud intelligence revenue rose 38% to $6 billion and AI-related product revenue posted triple-digit growth for the 11th straight quarter, with management highlighting custom AI chip capability as a structural advantage in China. The company expects quick-commerce unit economics to turn profitable by the end of 2027, but the article recommends staying on the sidelines until investment payoffs become clearer.

Analysis

The market is treating Alibaba less like a mature e-commerce platform and more like a capex option on China’s AI stack, but that valuation framing is fragile. The earnings deterioration matters because it reduces the company’s ability to self-fund multiple experiments at once; if management keeps pushing both quick commerce and AI, the next 2-4 quarters are likely to look worse before any operating leverage appears. That creates a classic “good narrative, bad numbers” setup where the stock can grind higher on AI headlines while the underlying multiple remains hostage to cash burn and margin compression.

The more important second-order effect is competitive: Alibaba’s custom chip effort is not just a product story, it is a cost-curve weapon against domestic cloud peers that must rent more of their infrastructure stack. If the company can internalize enough inference and training demand, it could compress cloud unit costs faster than rivals, which would matter more than near-term revenue growth. That said, the advantage is only meaningful if utilization stays high; underutilized in-house silicon can quickly turn from moat to fixed-cost overhang.

Quick commerce is the near-term drag with the clearest path to reversal, but the payoff window is long and uncertain. Management’s profitability target by 2027 implies at least several more quarters of depressed e-commerce margins, and any slowdown in consumer demand or promotional intensity would force either a retreat from share capture or a longer burn period. The consensus seems to be underestimating how long it can take for convenience retail to reach acceptable unit economics in a low-price, high-density market, especially if competitors respond with subsidies.

The contrarian view is that the stock may already be discounting the AI upside while still not fully pricing the duration of margin pain. In that case, the best risk/reward is not a naked long, but a structure that monetizes volatility and time decay while keeping upside exposure to an AI re-rating.

Key Decisions for Investors

• Avoid initiating a fresh outright long in BABA for the next 1-2 quarters; wait for either evidence of quick-commerce margin stabilization or a materially better cloud/AI monetization signal. Risk/reward is skewed by ongoing earnings downgrades.
• Consider a BABA call spread into the next 6-12 months only on weakness, sized for a narrative-driven rerating but capped for downside. Best use case is if AI disclosures improve before operating losses deepen further.
• Pair trade: long GOOGL vs short BABA on a 3-6 month horizon. Both have AI exposure, but GOOGL has a cleaner monetization path and less balance-sheet drag; BABA is carrying much more execution and margin risk.
• If you want China AI beta, prefer a basket/optionality approach over single-name BABA. The chip and cloud story is promising, but utilization risk and policy uncertainty make the distribution of outcomes too wide for a large outright position.
• Set a tactical trigger to reassess BABA if cloud growth accelerates while e-commerce margins stabilize for one full quarter; that is the point where the market can start capitalizing the AI stack rather than treating it as an offset to declining core profits.