Alibaba: An AI Loser So Far, But Don't Dismiss The Earnings Story

Alibaba is still viewed as a buy despite a 40% drawdown, but the stock faces headwinds from high capex, negative free cash flow, and AI uncertainty. Cloud revenue accelerated to 38% YoY, and management is targeting 40%+ intermediate-term growth with AI expected to drive future external revenue. The intrinsic value target was cut to $144 per share due to lower EPS estimates and a reduced P/E multiple amid continued profit downgrades.

Analysis

The market is still treating BABA like a value trap, but the more interesting setup is a re-rating of the cloud/AI option value versus the core commerce multiple. Heavy capex and negative FCF are not just a margin issue; they are a strategic moat-building spend that can widen the gap between Alibaba and domestic internet peers that lack the balance-sheet capacity to fund compute, model training, and cloud infrastructure at scale. That should eventually force smaller competitors into customer-acquisition-led growth with weaker economics, while hyperscale capacity concentration at Alibaba creates a path to take share in higher-margin enterprise workloads.

The main risk is timing: AI-driven revenue is a multi-quarter to multi-year catalyst, while earnings downgrades hit the stock immediately. In the next 1-2 quarters, the stock can still underperform if the market remains focused on near-term free cash flow dilution and if management is forced to keep funding capex to stay competitive. The reversal trigger is not just faster cloud growth, but evidence that AI usage is converting into paid workloads, higher attach rates, and improving incremental margins — that would shift the debate from “spend” to “payback.”

The contrarian miss is that the setup may be more asymmetric than the headline multiple implies: when a company is investing through the cycle, the market often underestimates how quickly operating leverage can reappear once capex normalizes. If cloud growth sustains in the high-30s and management credibly holds 40%+ growth guidance, the valuation floor should rise even without near-term EPS upgrades, because the cloud segment becomes the embedded call option on China enterprise IT demand. The stock can still be cheap on a sum-of-the-parts basis even if the consolidated P/E looks optically expensive versus a depressed earnings trough.

What would invalidate the thesis is a second leg of capex with no cloud monetization inflection, which would turn the story into permanent capital intensity rather than temporary investment. The key tell is whether external revenue begins to decouple from general macro softness over the next 2-3 reporting cycles; if not, the market will keep discounting the AI narrative as a cost center rather than a growth engine.

Key Decisions for Investors

• Go long BABA on weakness over the next 2-6 weeks, but size for volatility: the setup favors a 6-12 month hold if cloud growth remains >35% and management keeps guidance intact.
• Buy BABA Jan-2027 call spreads to express the AI/cloud re-rating with defined downside; this is better than outright stock if earnings downgrades continue to pressure the multiple in the near term.
• Pair trade: long BABA / short a basket of China internet names with weaker cloud or AI monetization pathways, to isolate the value of infrastructure-scale balance sheet and compute capacity.
• If BABA gaps higher on any commentary confirming paid AI workloads, use strength to add via call spreads rather than common, since the near-term risk remains capex-driven FCF volatility.
• Set a hard risk trigger: reduce exposure if cloud growth falls back below the mid-30s or if management signals another step-up in capex without a clear monetization runway.