Foxconn expects third quarter to outperform market as AI server demand rewrites its business

Foxconn has spent decades assembling iPhones in vast factory complexes across Asia. Now the company that built its empire on consumer gadgets is quietly becoming one of the most important picks-and-shovels players in the global AI buildout, and its financial results are starting to show it.

The world’s largest contract electronics manufacturer announced on September 5 that it expects third-quarter performance to exceed market estimates, citing persistent AI demand and the seasonal strength that typically lifts information and communication technology products in the second half of the year.

The numbers behind the confidence

August revenue hit T$921.8 billion, a 51.98% jump compared to the same month a year earlier, roughly equivalent to $29.15B at current exchange rates.

Year-to-date consolidated sales climbed to T$6.51 trillion, up 39.73% from the same period last year.

Second-quarter net profit rose 35% year-over-year, clearing analyst projections by a meaningful margin. The cloud and networking segment, which houses Foxconn’s AI server operations, now accounts for more than half of total company revenue. A year ago, consumer electronics held that title.

Foxconn is Nvidia’s principal server manufacturing partner, which puts it at the center of every data center buildout tied to large language models, generative AI applications, and the GPU clusters that power them.

A structural bet, not a seasonal one

Management has been careful to frame the AI server surge as something durable rather than a passing wave. Foxconn’s leadership is backing the conviction with capital: the company projects a 30% increase in capital expenditure for 2026, directed specifically at expanding AI server production capacity.

Management’s longer-range view suggests the AI industry could eventually scale to approximately $1 trillion, which is the kind of projection that explains why a contract manufacturer is willing to commit that kind of capital now.

Foxconn’s pivot toward AI infrastructure mirrors a broader pattern playing out across the tech supply chain. Foxconn still assembles a significant share of the world’s iPhones, and Apple remains a major customer. But the center of gravity inside the company has shifted.

The cloud and networking segment outpacing consumer electronics in multiple consecutive quarters reflects deliberate capacity allocation decisions, deeper partnerships with chipmakers, and a willingness to take on the engineering complexity that AI server manufacturing demands. These are not commodity machines. They require precise thermal management, high-speed interconnects, and integration of components that arrive from a handful of specialized suppliers.

What to watch from here

Leadership explicitly flagged the need to stay alert to fluctuating global political and economic conditions. That phrasing covers a lot of ground, from US-China trade tensions that could complicate supply chains, to currency moves that affect the translation of Taiwan-dollar revenues into other reporting currencies, to any slowdown in hyperscaler capex that could ripple through Foxconn’s order flow.

The concentration of AI server revenue in a segment tied to a relatively small number of large customers is a feature when those customers are spending aggressively. Nvidia GPU availability, regulatory approvals for data center construction, and power grid constraints at potential data center sites are all variables outside Foxconn’s direct control but very much inside its earnings exposure.

Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy.