Foxconn’s Q3 revenue tops estimates as AI server demand surges

Foxconn’s Q3 revenue tops estimates as AI server demand surges

Hon Hai posted NT$3.03 trillion in third-quarter revenue, up 47% from a year earlier, as cloud providers kept spending on AI infrastructure

The company best known for assembling iPhones is increasingly a company that builds AI.

Hon Hai Precision Industry Co., better known as Foxconn, reported third-quarter 2026 revenue of NT$3.03 trillion (approximately US$95.4 billion). That is a 47% jump from the same quarter last year. It also cleared the market’s estimate of NT$2.83 trillion by a comfortable margin.

The reason is no mystery. Cloud service providers kept raising their capital expenditures on AI infrastructure, and Foxconn is one of the main places that money lands.

The numbers behind the beat

In the second quarter of 2026, Foxconn reported revenue of NT$2.53 trillion, up 41% year-on-year. So growth did not just continue. It accelerated, from 41% to 47%.

Advertisement

The second quarter was already a strong showing on the bottom line. Net profit hit a record NT$60 billion (approximately US$1.86 billion), up 35% from a year earlier. That figure topped the NT$58 billion analysts had expected.

Cumulative revenue from January through August 2026 reached a record NT$6.51 trillion, nearly 40% higher than the same stretch a year earlier.

AI servers now carry the business

AI server-related products accounted for more than half of Foxconn’s total revenue in the second quarter. In the third quarter, the cloud and networking segment, driven primarily by AI servers, remained the main engine of growth.

Foxconn assembles products for both Apple and Nvidia. It is also described as Nvidia’s primary server manufacturing partner. Rather than shipping individual parts, it delivers complete AI server racks and the advanced networking gear that links them together.

Foxconn expects AI server rack shipments to more than double in full-year 2026. It also forecast that revenue from 800G+ networking switches will double.

From contract manufacturer to AI supplier

Foxconn has long held the title of the world’s largest electronics contract manufacturer. The company has continued investing in global production capacity, particularly in the United States.

What this means for investors and the AI supply chain

For investors, Foxconn’s results work as a real-time read on AI infrastructure spending. The company sits downstream of chip designers and upstream of the data centers.

With AI servers making up more than half of revenue, Foxconn has effectively bet its near-term trajectory on a single trend. Revenue growth of 47% is striking, but the second-quarter profit increase of 35% trailed revenue growth of 41%.

The next data points to watch are whether rack shipments actually double by year-end, whether the 800G+ switch forecast holds, and whether cloud providers keep their capital spending plans intact into the coming year.

Disclosure: This article was edited by Diego Almada Lopez. For more information on how we create and review content, see our Editorial Policy.
Foxconn’s Q3 revenue tops estimates as AI server demand surges
Foxconn’s Q3 revenue tops estimates as AI server demand surges

Hon Hai posted NT$3.03 trillion in third-quarter revenue, up 47% from a year earlier, as cloud providers kept spending on AI infrastructure

The company best known for assembling iPhones is increasingly a company that builds AI.

Hon Hai Precision Industry Co., better known as Foxconn, reported third-quarter 2026 revenue of NT$3.03 trillion (approximately US$95.4 billion). That is a 47% jump from the same quarter last year. It also cleared the market’s estimate of NT$2.83 trillion by a comfortable margin.

The reason is no mystery. Cloud service providers kept raising their capital expenditures on AI infrastructure, and Foxconn is one of the main places that money lands.

The numbers behind the beat

In the second quarter of 2026, Foxconn reported revenue of NT$2.53 trillion, up 41% year-on-year. So growth did not just continue. It accelerated, from 41% to 47%.

Advertisement

The second quarter was already a strong showing on the bottom line. Net profit hit a record NT$60 billion (approximately US$1.86 billion), up 35% from a year earlier. That figure topped the NT$58 billion analysts had expected.

Cumulative revenue from January through August 2026 reached a record NT$6.51 trillion, nearly 40% higher than the same stretch a year earlier.

AI servers now carry the business

AI server-related products accounted for more than half of Foxconn’s total revenue in the second quarter. In the third quarter, the cloud and networking segment, driven primarily by AI servers, remained the main engine of growth.

Foxconn assembles products for both Apple and Nvidia. It is also described as Nvidia’s primary server manufacturing partner. Rather than shipping individual parts, it delivers complete AI server racks and the advanced networking gear that links them together.

Foxconn expects AI server rack shipments to more than double in full-year 2026. It also forecast that revenue from 800G+ networking switches will double.

From contract manufacturer to AI supplier

Foxconn has long held the title of the world’s largest electronics contract manufacturer. The company has continued investing in global production capacity, particularly in the United States.

What this means for investors and the AI supply chain

For investors, Foxconn’s results work as a real-time read on AI infrastructure spending. The company sits downstream of chip designers and upstream of the data centers.

With AI servers making up more than half of revenue, Foxconn has effectively bet its near-term trajectory on a single trend. Revenue growth of 47% is striking, but the second-quarter profit increase of 35% trailed revenue growth of 41%.

The next data points to watch are whether rack shipments actually double by year-end, whether the 800G+ switch forecast holds, and whether cloud providers keep their capital spending plans intact into the coming year.

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