Alibaba Cuts Buyback by 80% to Invest in AI Growth

Alibaba’s board just sent a message to shareholders, and it wasn’t written in an earnings press release. It was written in a single line item: share buybacks. In the quarter ending June 2026, the company spent just $162 million to repurchase its own shares. A year earlier, in the same fiscal quarter, that figure was $815 million. That’s an 80% cut in the pace of repurchases at a company that still had a staggering $19.3 billion of authorization left on the books. You don’t need a degree in finance to read that signal. Capital is being deployed elsewhere, with urgency.

Walking the floor of the New York Stock Exchange, you hear a lot of theories about capital allocation. The chatter this week is all about this pivot. The money isn’t vanishing; it’s being poured into silicon and concrete. Capital expenditures surged 75% year-over-year to RMB 67.7 billion, almost entirely for AI infrastructure. This spending hemorrhage turned what was already negative free cash flow into a deeper RMB 44.7 billion deficit. A newly broken-out segment, AI Labs and Applications, posted a steep adjusted EBITA loss of RMB 13.9 billion, more than quadruple the loss from a year ago. Layer on a EUR 550 million European Commission fine and a nearly RMB 4.5 billion goodwill impairment, and the quarter paints a picture of a company in the thick of a costly transformation.

I’ve covered enough earnings cycles to know when a narrative is being tested. The internal cash flow clearly isn’t covering this buildout. To bridge the gap, Alibaba raised approximately $3.2 billion in convertible notes and HK$12 billion in exchangeable bonds during fiscal 2026, largely to fund its cloud and international commerce ambitions. The consequence is on the balance sheet: total debt to adjusted EBITDA roughly doubled to 2.29x. The full-year buyback tally of $1.046 billion is a faint echo of prior years’ programs. This isn’t subtle portfolio management; it’s a strategic bet with the balance sheet on the line.

On the August 20 earnings call, CEO Eddie Wu didn’t shy away from the scale of the shift. “AI has become Alibaba’s most certain growth engine,” he stated plainly. CFO Toby Xu followed with the financial rationale, arguing the investments have “a clear path to attractive ROIC.” The thesis, as they laid it out, rests on a constrained supply of AI computing power. Management believes the industry-wide crunch for advanced chips and data center capacity will last until at least 2030. Their math suggests the AI hardware itself can reach break-even within three years on a standard five-year useful life. It’s a gamble on scarcity creating value.

The early returns offer some validation. Revenue from AI Cloud and Compute Services grew 45%. AI-related product revenue notched its twelfth straight quarter of triple-digit growth. Cloud external revenue growth hit its highest rate in 22 quarters. Perhaps most telling, the annual run rate for its Model-as-a-Service (MaaS) offering surpassed RMB 16 billion as of August, putting it on track to exceed a year-end target of RMB 30 billion. These are the green shoots management is banking on to justify the redirected cash.

For investors watching the tape, the setup is now remarkably clear, if binary. The stock closed at $130.53 on August 20, up for the day and the week but still down for the year. If the monetization of models like Qwen and the deployment of its proprietary Zhenwu chips accelerate as guided, today’s buyback cut will be remembered as a masterclass in disciplined reinvestment. If however the losses in AI Labs continue to widen without a corresponding expansion in cloud profitability, the company’s substantial net cash pile of $46.5 billion becomes the shock absorber. As Wu succinctly put it, “It’s only possible to monetize when you have that compute capacity in place.” Alibaba is buying the capacity first, and asking shareholders to wait for the monetization later. In the high-stakes game of AI, that’s the only move they believe is on the board.

• Share buybacks have decreased by 80%
• Capital expenditures surged 75% year-over-year
• AI Labs posted a steep adjusted EBITA loss of RMB 13.9 billion
• Alibaba raised $3.2 billion in convertible notes
• Total debt to adjusted EBITDA roughly doubled to 2.29x
• AI Cloud revenue grew 45%