Foxconn Inc., the world’s largest contract electronics manufacturer, is preparing an expanded collaboration with artificial intelligence (AI) pioneer OpenAI, Chairman Young Liu announced Friday at the company’s annual technology showcase.
Speaking to reporters in Taipei, Liu confirmed he met earlier in the day with OpenAI CEO Sam Altman. While specific terms remain under wraps, Liu stated an official announcement will follow once final details are locked down.
Any forthcoming deal will expand upon a November 2025 pact between the two tech giants to co-design and engineer data-center racks, cooling systems, power infrastructure, and specialized AI hardware in the United States.
Although the initial framework established design collaboration without binding financial obligations, it positioned OpenAI to evaluate and potentially deploy Foxconn-built hardware at scale.
The disclosure arrives as tech investors grapple with volatility over whether soaring capital expenditures on AI infrastructure can deliver long-term returns amid elevated borrowing costs. Liu dismissed fears of an impending AI bubble, projecting that Foxconn will sustain double-digit growth heading into 2027.
Formally known as Hon Hai Precision Industry, Foxconn has aggressively pivoted to AI infrastructure to secure its next wave of expansion, emerging as the dominant server manufacturing partner for semiconductor heavyweight NVIDIA Corp.
The strategic transition is already showing substantial returns. Foxconn’s second-quarter revenue surged 41% year over year to NT$2.53 trillion, driven by server shipments that are projected to more than double over the full year.
Additionally, Foxconn’s cloud and networking segment, which encompasses AI servers, generated 51% of overall sales in the second quarter — the first time the division represented most company revenues. Looking ahead, executive leadership anticipates AI infrastructure will represent between 60% and 65% of aggregate sales over the next three to five years.
To service surging global demand, Foxconn is scaling its physical presence across North America.
The company operates six factory facilities in Houston dedicated to assembling high-performance AI servers for NVIDIA, with Liu signaling that additional facility expansions are imminent.
Local initiatives are anchored by a $450 million investment pledge announced alongside the Greater Houston Partnership, an effort expected to add roughly 600 direct jobs, support more than 800 indirect positions, and deliver an estimated $920 million in local economic impact.
Foxconn is also using its Texas footprint as an operational testing ground for next-generation factory automation. Brand Cheng, chairman of subsidiary Foxconn Industrial Internet (FII), confirmed Friday that more than 10 humanoid robots have already been deployed for internal operations at the Houston facilities, with broader rollouts planned across international plants.
Utilizing software models built upon NVIDIA’s Project GR00T robotics foundation platform, the deployed machinery includes wheeled units tailored for heavy vertical lifting alongside articulated systems engineered to bend at the waist for precision assembly, loading, and material handling tasks.
