Samsung and TSMC both plan to mass-produce 2nm processes in 2025, and Samsung hopes to be the first to achieve mass production and overwhelm each other with speed to gain a competitive advantage in the next generation of process nodes. Previously, it was reported that Samsung considered offering discounts for 2nm orders in order to obtain order support from industry giants such as Nvidia to further challenge TSMC's leadership. According to Business Korea, an industry source revealed that Samsung has acquired a new lease from Japanese artificial intelligence (AI) startup Preferred Networks Inc(PFN) to seize the lead in the 2nm foundry business. 
preferred networks inc.Founded in 2014, it focuses on the development of artificial intelligence deep learning, and has attracted significant investment from major companies in various fields, including Toyota, NTT, and Fanuc. It is understood that Samsung was selected because it has both memory and foundry services, and has strong comprehensive capabilities and technology accumulation, which can provide high-bandwidth memory (HBM) design to production and advanced 2A complete solution for 5D packaging. Past Preferred Networks IncI also cooperated with TSMC, and this time I turned to Samsung, to a certain extent, from the perspective of the ** chain, and reduced my dependence on TSMC.
Samsung mass-produced SF3E (3nm GAA) in June 2022 and introduced a new GAA (Gate-All-Around) architecture transistor technology. According to the previously announced plan, Samsung will bring the second-generation 3nm process technology called SF3 (3gap) this year, using the "second-generation multi-bridge-channel field-effect transistor (MBCFET)" to further optimize on the basis of the original SF3E, and then there will be a performance-enhancing SF3P (3gap+), which is more suitable for manufacturing high-performance chips. By 2025, Samsung will begin mass production of the SF2 (2nm) process.