Samsung Electro-Mechanics receives combined investment of KRW 3.7 trillion from Tesla and NVIDIA to expand AI chip substrate production lines

According to a South Korean media report on October 22, Tesla and NVIDIA will jointly invest approximately 3.7 trillion KRW (~$2.5 billion), with Tesla contributing 1.1 trillion KRW and NVIDIA 2.6 trillion KRW, to support Samsung Electro-Mechanics’ expansion of its next-generation FC-BGA (Flip-Chip Ball Grid Array) substrate production line. FC-BGA substrates serve as critical packaging carriers for high-performance chips—including AI accelerators, GPUs, and CPUs—and have become a key bottleneck in the supply chain amid surging AI chip computational demands.

Samsung Electro-Mechanics will build a dedicated new production line for Tesla at its Sejong facility; the FC-BGA substrates produced there will be exclusively supplied to AI chips used in Tesla’s autonomous vehicles and humanoid robot Optimus. The two parties have signed a five-year long-term supply agreement covering 2029 through 2033. Meanwhile, NVIDIA’s investment will focus on expanding Samsung Electro-Mechanics’ facilities in both Sejong and Busan, primarily to ramp up mass production of substrates for AI computing chips and data center networking hardware.

Per the plan, Samsung Electro-Mechanics aims to complete the relevant factory construction by the end of 2027, begin mass production in the second half of 2028, and commence formal deliveries to Tesla starting in 2029. This collaboration is expected not only to significantly boost Samsung Electro-Mechanics’ capacity for premium FC-BGA substrates but also to secure highly predictable, long-term downstream demand.