Foxconn will also produce critical components for U.S.-based AI data centers, including cabling, networking equipment, cooling infrastructure, and power systems.
OpenAI has partnered with Hon Hai Technology Group (Foxconn) to strengthen U.S. manufacturing readiness for next-generation AI infrastructure hardware, the companies announced on November 20.
While the agreement does not include any purchase commitments, OpenAI will gain early access to test Foxconn’s new systems, with the option to procure them later. The partnership is focused on meeting the surging demand for advanced physical infrastructure needed to train and deploy next-generation AI models.
Under the initiative, the companies will co-design and develop multiple generations of AI data centre racks simultaneously, aligning OpenAI’s infrastructure roadmap with Foxconn’s engineering and manufacturing strengths.
Sam Altman, CEO of OpenAI, said the collaboration marks a generational opportunity to reindustrialise the United States. “This partnership moves us closer to ensuring that the foundational technologies of the AI era are built domestically. We believe this work will reinforce US leadership and help ensure the benefits of AI are broadly shared,” he said.
As part of the collaboration, the companies will work to expand domestic sourcing, optimise rack architectures for US manufacturing, and scale local testing and assembly. These efforts are aimed at strengthening the US AI supply chain, accelerating deployment, and enhancing system reliability.
Foxconn will also produce key components for US-based AI data centres such as cabling, networking, cooling, and power systems to support the buildout of high-performance compute infrastructure.
We at Foxconn are excited to partner with OpenAI. As the world’s largest manufacturer of AI data servers, we are uniquely positioned to support OpenAI’s mission with reliable, scalable infrastructure,” said Foxconn chairman Young Liu.
OpenAI previously announced a multi-year strategic partnership with Broadcom to co-develop and deploy 10 gigawatts of OpenAI-designed AI accelerators and networking systems, marking a significant expansion of the company’s infrastructure ambitions.
Under the partnership, OpenAI will design the accelerators and systems, while Broadcom will supply Ethernet and other connectivity technologies for large-scale deployment across OpenAI facilities and partner data centres. Rack deployment is slated to begin in the second half of 2026 and finish by the end of 2029.
