Alibaba reports continued profit squeeze from spending on AI and instant retail

Alibaba reported Q4 revenue of 243.38 billion yuan, missing the 247.22 billion yuan consensus, while adjusted EPS of 0.62 yuan fell far short of the 5.79 yuan estimate. Cloud revenue rose 38% to 41.63 billion yuan, but profitability was hit by heavy AI, cloud, and quick commerce investment, with adjusted EBITA down 84% and one-time adjusted net income down 99.7%. China e-commerce revenue of 122.22 billion yuan beat estimates, but the stock was down 2.3% premarket on the mixed-to-weak earnings print.

Analysis

The key read-through is not that Alibaba missed on profits; it is that the company is choosing to sacrifice near-term margin to buy relevance in AI-enabled commerce, and that makes the stock a battleground between optically weak earnings and a potential multi-year platform shift. The market is likely still underestimating how quickly AI can compress the “search-to-cart” funnel on Chinese e-commerce, which could lift conversion and take-rate mix over the next 12-24 months even if accounting profits remain noisy in the next few quarters.

For NVDA, the second-order effect is demand durability rather than immediate upside. A major China cloud player openly prioritizing AI capex, despite short-term margin pain, signals that enterprise AI spend in China is becoming strategic infrastructure rather than experimental spend; that supports a longer runway for accelerators, networking, and software attach rates. The risk is policy, not demand: any escalation in U.S.-China controls or China-specific supply substitution efforts could blunt the benefit with a lag of 1-3 quarters.

BABA’s quick-commerce investment is the clearest near-term overhang because it turns the equity into a subsidy war proxy. If management keeps defending share in instant retail, the earnings reset can persist for several quarters even if revenue stabilizes, and that likely keeps value investors sidelined. The contrarian angle is that the market may be over-anchored on near-term EPS misses and underpricing the optionality from AI monetization inside Taobao/Tmall, where modest improvements in conversion or ad yield could matter more than headline revenue growth.

The setup favors relative-value trades over outright direction. Into the next 1-2 quarters, the cleaner expression is to short BABA on rallies or own it only via options, while using NVDA pullbacks as a higher-quality way to stay exposed to China AI capex without owning the consumer subsidy risk. Watch for any signal that Alibaba is throttling quick-commerce spend or showing AI-driven monetization in cloud; that would be the inflection point that forces a re-rate.

Key Decisions for Investors

• Short BABA common or buy 1-2 quarter downside puts on strength; thesis is continued margin compression from AI and quick-commerce spend with earnings revisions likely negative for the next 2-3 prints.
• Long NVDA on weakness as a China-AI capex proxy, but size modestly given policy risk; best risk/reward is a 3-6 month horizon where accelerator demand can outlast headline volatility.
• Pair trade: long NVDA / short BABA to isolate AI infrastructure exposure versus consumer-commerce margin destruction; this is the cleaner expression if China tech sentiment stays mixed.
• For event-driven accounts, buy BABA call spreads only after evidence of AI monetization or quick-commerce spend moderation; avoid paying up before margin inflection because the burn rate can keep compressing multiples.
• Set a catalyst watch for Alibaba commentary on external AI revenue and cloud mix over the next 1-2 quarters; any evidence of conversion lift from AI shopping assistants could trigger a sharp short-covering rally.