Nvidia is considering outsourcing some AI GPU production to Samsung

As TSMC's production capacity supply becomes increasingly tight, Nvidia is considering outsourcing the production of some artificial intelligence (AI) GPUs to Samsung Electronics. It is reported that the popularity of generative AI such as the chat robot ChatGPT has increased the demand for high-performance GPUs such as NVIDIA H100, A100, H800 and A800, which has enabled the company to capture 90% of the global AI GPU market share.

Investment bank JPMorgan Chase believes that with hardware products such as GPUs and network products, Nvidia will occupy up to 60% of the artificial intelligence product market this year.

Currently, Nvidia is favored by investors and is considered a key supplier to meet the demand for AI computing power. Its high-end processors have been used to train and run various chatbots.

At the end of May this year, Nvidia CEO Jen-Hsun Huang said that the company's supply chain will strive to diversify. In addition to TSMC, the current highest-end H100 GPU will also be manufactured by Samsung and Intel.

However, Korean media recently reported that Nvidia's A100 and H100 GPUs are exclusively supplied by TSMC, and Samsung Electronics failed to grab any orders because TSMC's advanced packaging technology CoWoS is ahead of Samsung Electronics.

Industry observers pointed out that if Samsung’s 3nm trial products pass performance verification and its 2.5D advanced packaging technology meets Nvidia’s requirements, Samsung may obtain some orders from Nvidia.