Alibaba Raises $10.2 Billion in Record Hong Kong Share Sale — Every Dollar Earmarked for AI

Alibaba Group Holding made the biggest capital-raising move of its AI era on Sunday, launching a HK$80 billion (US$10.2 billion) share placement in Hong Kong — and pledging that every dollar of net proceeds will go directly into artificial intelligence. Coming just days after the Chinese tech giant reported a 75% collapse in quarterly net income driven by AI spending, the deal is an unambiguous signal: Alibaba is all-in on AI, and it is willing to dilute shareholders to fund the bet at unprecedented scale.

What Alibaba Announced

According to Reuters, Alibaba launched the placement on Sunday, August 23, with an aggregate consideration of HK$80 billion, equivalent to roughly $10.2 billion at current exchange rates ($1 = 7.8396 HKD). The South China Morning Post reports the company will issue HK$80 billion worth of new shares, with the entire proceeds directed at AI investment.

The mechanics, as reported by KFGO and other outlets tracking the deal documents: Alibaba plans to sell approximately 710 million ordinary shares at HK$112.70 per share — a 3.6% discount to the most recent closing price. The placement ranks among the largest equity raises in Hong Kong’s history and is by far the biggest share issuance Alibaba has attempted since its landmark 2019 secondary listing.

In its official announcement, carried by Morningstar and Business Wire, Alibaba stated it “intends to use 100% of the net proceeds from the Equity Placement to invest in its full stack AI capabilities, including to expand and extend the Company’s global AI” infrastructure footprint. That phrase — full-stack AI — matters. Alibaba is not funding a single model or product line; it is funding the entire vertical stack, from custom silicon and data centers through cloud infrastructure to the Qwen model family and consumer-facing applications.

The Backdrop: A 75% Profit Collapse That Management Chose

The placement lands on top of an extraordinary earnings week for Alibaba. On August 20, the company reported June-quarter results that stunned markets: revenue rose 9% to nearly 269 billion yuan (about $40 billion), beating analyst estimates, but net income plummeted roughly 75-76% to 10.54 billion yuan as AI infrastructure spending swallowed margins. Shares fell 5-7% on the news.

The cause was capital expenditure approaching $10 billion in a single quarter — a run-rate that has alarmed bulls and bears alike. Bloomberg’s analysis flagged rising concerns about “circular AI financing,” the pattern where AI investment flows between interconnected vendors, clouds, and model labs in ways that may inflate demand signals.

Yet the counter-argument is visible in the same earnings report. Alibaba’s AI cloud and computing revenue surged 45% year-over-year to 48.44 billion yuan (US$7.2 billion) in the quarter — the strongest growth in the segment’s history as a scaled business, with AI-related products posting triple-digit growth for the eleventh consecutive quarter. Management has repeatedly pointed to its three-year plan to invest at least 380 billion yuan (over $52 billion) in cloud and AI infrastructure, announced in February 2025, and analysts now expect the company to overshoot that target.

Why Raise Equity Now?

The choice of an equity placement rather than debt is strategically loaded. First, it removes any question of funding constraints: Alibaba’s AI capex ambitions now have a dedicated, unlevered pool of capital. Second, it signals duration — equity markets tolerate multi-year investment horizons better than bond markets watching quarterly interest coverage. Third, and perhaps most importantly, it is a statement of confidence aimed at Beijing, Washington, and global investors simultaneously: Chinese AI champions can still raise world-scale capital in Hong Kong, in size, at a modest 3.6% discount.

The timing also exploits a strong tape. Alibaba’s Hong Kong-listed shares (9988.HK) had been trading near HK$123 before the deal was announced, up substantially over the past year as investors rewarded Qwen’s momentum — the open-weight Qwen family has become the most downloaded model lineage on Hugging Face, and Alibaba recently unveiled its own AI inference chip to reduce reliance on Nvidia as Chinese firms race to secure domestic compute.

The Bigger Picture: The Global Capex Arms Race

Alibaba’s $10.2 billion raise is the newest entry in what has become a full-blown global capital arms race in AI infrastructure. Worldwide AI investment is estimated to approach $4 trillion over the coming years, with hyperscalers in the US — Microsoft, Amazon, Google, and Meta — collectively committing hundreds of billions annually. Nvidia’s recent $6 billion licensing-and-hiring deal with AI startup Poolside, and its earlier $20 billion Groq transaction, illustrate how value is accruing not just to chipmakers but to anyone holding critical AI technology or talent.

Within China, the placement escalates an already intense dynamic. Baidu, Tencent, ByteDance, and a wave of well-funded startups are all racing to build compute, while US export controls keep top-end Nvidia hardware out of reach — making capital even more critical for Chinese firms, who must fund domestic chip alternatives and less compute-efficient training runs. Alibaba’s equity raise effectively dares rivals to match its balance sheet commitment.

Risks and Open Questions

The bull case is straightforward: 45% AI cloud growth on a $7.2 billion quarterly revenue base, a world-leading open-source model family, and a captive domestic market being re-architected around AI. The bear case is equally clear. Profit is down 75%. The “circular financing” critique — AI vendors investing in AI customers who buy from those same vendors — has not been answered convincingly by anyone in the industry. And a 710-million-share issuance at a 3.6% discount implies existing holders absorbed immediate dilution of roughly 2.7% of shares outstanding.

The central question for the next eight quarters is whether Alibaba’s full-stack bet converts capex into durable returns before capital markets lose patience. Management has said AI spending could pay off within three years. The $10.2 billion raised on Sunday shortens the leash and raises the stakes: this is now one of the largest single-purpose AI war chests ever assembled by a listed company, and the world will be watching what it buys.