Reports indicate that OpenAI is in discussions with Broadcom and other chip design companies to jointly develop a new type of AI server chip. This initiative, led by OpenAI CEO Sam Altman, aims to enhance the company's computing power for AI development. The project is seen as a response by OpenAI to address the current shortage of graphics processing units (GPUs) required for AI model development, which are primarily supplied by Nvidia.
Image source note: The image was generated by AI, and the authorized service provider is Midjourney.
Additionally, OpenAI plans to recruit former Google employees who were involved in the development of tensor processing units (TPUs) to assist with chip design. Broadcom's previous collaboration with Google on TPU development gives it an edge in custom AI chip experience.
OpenAI has neither fully confirmed nor denied this news. A spokesperson stated, "OpenAI is engaged in ongoing dialogue with industry and government stakeholders to increase the accessibility of AI infrastructure and ensure the benefits of AI are widely accessible to society."
OpenAI previously revealed intentions to develop custom AI chips in September 2023, and in January 2024, Sam Altman was reported to be seeking funds from institutions such as SoftBank Group and Abu Dhabi G42 to support a new chip manufacturing project. Although the reported fundraising amount was as high as $7 trillion, a figure far exceeding the economic output of Australia and Japan, it is likely too large, possibly prompting OpenAI to choose to collaborate with existing companies like Broadcom to reduce costs and achieve its goals.
OpenAI's move also reflects its ambition in the AI chip sector. Currently, Nvidia holds a market share of approximately 70% to 95% in the AI chip market. OpenAI's strategy clearly aims to reduce reliance on Nvidia, thereby enhancing its market competitiveness.
**Key Points:**
- 🗣️ OpenAI CEO Sam Altman is spearheading this plan to improve the computing power required for the company's AI development.
- 🤝 OpenAI is also recruiting former Google employees who participated in the development of Google's tensor processing units (TPUs) to aid in chip design.
- 🚀 OpenAI's chip development plan is partly to reduce dependence on Nvidia's graphics processing units (GPUs) and increase its competitiveness in the AI chip market.