OpenAI and Broadcom Team Up to Deploy 10 Gigawatts of Custom AI Accelerators

October 13, 2025

In a significant move that promises to reshape the landscape of artificial intelligence (AI), OpenAI and Broadcom have announced a strategic collaboration aimed at deploying 10 gigawatts of custom-designed AI accelerators. This multi-year partnership will see both companies co-developing systems that integrate OpenAI’s cutting-edge accelerators with Broadcom’s robust Ethernet solutions, designed for scale-up and scale-out capabilities.

What Changed Technically?

• Custom Accelerator Design: OpenAI is taking the helm in designing its own AI accelerators, a departure from using off-the-shelf hardware. This custom design allows them to embed insights gained from developing frontier models directly into the hardware, optimizing performance for their specific needs.
• Ethernet Integration: The systems will be scaled entirely with Ethernet and other connectivity solutions from Broadcom. This ensures seamless integration and high-speed data transfer within AI clusters, crucial for handling large-scale machine learning workloads.

Why It Matters to Practitioners

• Performance Gains: Custom-designed accelerators can offer significant performance improvements over generic hardware, especially when optimized for specific tasks like training large language models or running complex simulations.
• Scalability and Efficiency: By leveraging Broadcom’s Ethernet solutions, the systems will be highly scalable and power-efficient. This is essential for managing the growing demand for AI compute resources without breaking the bank on energy costs.
• Ecosystem Expansion: The collaboration not only benefits OpenAI but also contributes to a broader ecosystem of partners working together to push the boundaries of AI technology.

Deployment Timeline

• Start Date: The first racks of AI accelerator and network systems are targeted to start deployment in the second half of 2026.
• Completion Date: All deployments are expected to be completed by the end of 2029, spanning OpenAI’s facilities and partner data centers.

Key Quotes

• Sam Altman, Co-founder and CEO of OpenAI: "Partnering with Broadcom is a critical step in building the infrastructure needed to unlock AI’s potential and deliver real benefits for people and businesses. Developing our own accelerators adds to the broader ecosystem of partners all building the capacity required to push the frontier of AI to provide benefits to all humanity."

• Hock Tan, President and CEO of Broadcom: "Broadcom’s collaboration with OpenAI signifies a pivotal moment in the pursuit of artificial general intelligence. OpenAI has been at the forefront of the AI revolution since the ChatGPT moment, and we are thrilled to co-develop and deploy 10 gigawatts of next-generation accelerators and network systems to pave the way for the future of AI."

• Greg Brockman, Co-founder and President of OpenAI: "Our collaboration with Broadcom will power breakthroughs in AI and bring the technology’s full potential closer to reality. By building our own chip, we can embed what we’ve learned from creating frontier models and products directly into the hardware, unlocking new levels of capability and intelligence."

• Charlie Kawwas, Ph.D., President of the Semiconductor Solutions Group at Broadcom: "Our partnership with OpenAI continues to set new industry benchmarks for the design and deployment of open, scalable, and power-efficient AI clusters."

Technical Details

• Accelerator Design: The custom accelerators will be optimized for high-performance computing tasks, such as training deep neural networks. They are expected to offer significant improvements in throughput, latency, and energy efficiency compared to existing solutions.
• Ethernet Solutions: Broadcom’s Ethernet solutions will provide the necessary infrastructure for high-speed data transfer between nodes in the AI clusters. This is crucial for maintaining performance at scale and ensuring that data can be processed efficiently.

Conclusion

This collaboration between OpenAI and Broadcom represents a significant step forward in the development of advanced AI infrastructure. By combining custom-designed accelerators with robust networking solutions, the partnership aims to meet the growing global demand for AI compute resources while driving innovation and efficiency in the field.