OpenAI and Foxconn Partner for Next-Generation AI Hardware
OpenAI has formed a new partnership with Hon Hai Technology Group (Foxconn) to advance U.S.-based production readiness for next-generation artificial intelligence infrastructure hardware, the companies announced on November 20.
Although the agreement does not require either company to make purchases, OpenAI will be able to test new hardware early and choose to buy it if desired. The collaboration focuses on meeting the rapidly increasing need for physical systems capable of supporting more powerful AI models.
Through this effort, OpenAI and Foxconn will jointly design and develop several generations of AI data-center racks at the same time, aligning OpenAI’s long-term infrastructure goals with Foxconn’s engineering and manufacturing strengths.
OpenAI CEO Sam Altman said the initiative represents a rare opportunity to revitalize American manufacturing. He noted that creating essential technologies domestically will reinforce U.S. leadership and help ensure the benefits of AI are shared widely.
As part of the partnership, both companies will work to increase the use of U.S.-sourced components, refine rack designs for domestic production, and expand local testing and assembly. These steps aim to boost supply-chain resilience, accelerate rollout, and improve system reliability.
Building a Strong U.S. AI Infrastructure
Foxconn will also produce major components for AI data centers in the United States—including networking gear, cooling equipment, cabling, and power systems—to support the creation of high-performance computing infrastructure.
Foxconn chairman Young Liu said the company is excited to collaborate with OpenAI, emphasizing that Foxconn’s role as the world’s largest manufacturer of AI data servers positions it well to deliver the robust, scalable technology OpenAI requires.
The announcement follows OpenAI’s earlier multi-year collaboration with Broadcom to design and deploy 10 gigawatts of custom AI accelerators and networking hardware, signaling a major expansion of OpenAI’s infrastructure capabilities.
In that arrangement, OpenAI will design the accelerators and systems, while Broadcom will supply Ethernet and other connectivity technologies for large-scale rollout across OpenAI facilities and partner data centers. Deployment of the new racks is scheduled to begin in late 2026 and finish by the end of 2029.
Also Read:
Insurtech Startup Pibit.AI Secures $7 Million from Stellaris, YC & Arali
Google ramps up AI scam protection in India, but gaps remain
