Alibaba just landed the biggest Hong Kong stock sale in history to fund AI. Buried in the same earnings release: profit fell by three-quarters.
The Chinese tech giant said it plans to raise HK$80 billion, about $10.2 billion, through a new share placement, directing all of the proceeds toward what it calls full stack AI capabilities, chips, infrastructure and the AI models built on top of them.
The Largest Deal of Its Kind
The scale here is real. Alibaba says the placement would be the largest-ever primary follow-on offering by a Hong Kong-listed company and the biggest Regulation S equity offering on record.
That ranks it third-largest globally this year, behind only offerings from Alphabet and Intel.
The term sheet calls for selling 710 million shares at HK$112.70 each, a 3.6% discount to Alibaba's recent closing price, sold exclusively to non-U.S. investors.
"The Equity Placement is being undertaken to extend the Company's global AI leadership," Alibaba said in its announcement.
The Uncomfortable Number Buried in the Announcement
Alibaba's AI business is genuinely growing. Cloud segment revenue rose 45% year over year last quarter, and AI-related product revenue posted triple-digit growth for the twelfth consecutive quarter.
Net income told a different story, falling 75% year over year to roughly $1.54 billion, driven by lower operating income and weaker investment gains.
Capital expenditures jumped 75% in the same period, to nearly $10 billion for the quarter alone. Spending and losses moved in the same direction at the same time, which is exactly the pattern that should raise a flag rather than reassure anyone.
That combination, rising AI revenue and collapsing net income, is close to a textbook illustration of the margin structure Apollo's chief economist described in research MAIN covered this week: AI revenue growth at the model and application layer doesn't automatically translate into profit, even at a company as large and diversified as Alibaba.
The CEO's Own Timeline
Alibaba CEO Eddie Wu told investors on the earnings call that the company expects its AI computing investments to break even within three years, with a shot at two years if gross margins keep improving.
That's a specific, testable promise. Alibaba has already spent roughly $26 billion in AI-related capex through the end of June, part of a $53 billion commitment for AI investment between 2026 and 2029 announced last year.
Break-even promises are easy to make on an earnings call and harder to hold to three years later, especially when the current quarter's numbers are moving in the wrong direction rather than the right one. Wu's own timeline is the benchmark against which this raise should eventually get judged.
Alibaba's Qwen family of open-weight models already competes directly with Western labs on cost and adoption, a dynamic MAIN has covered in the context of enterprise AI buying decisions. This new capital is aimed at keeping that lead while absorbing the infrastructure costs behind it.
What This Means for Miami
This is the same dynamic playing out at Nscale, Broadcom and every other company MAIN has tracked raising capital against AI infrastructure buildouts this year, just at a scale large enough to make Hong Kong stock market history.
For Miami investors weighing AI exposure, Alibaba's own numbers this quarter are a useful real-time test case. A company posting genuinely strong AI revenue growth still saw net income fall by three-quarters in the same period, a reminder that top-line AI growth and actual profitability are currently two very different questions, even for companies with the scale and diversification to absorb the gap.