Samsung Electronics announced on Monday, at the Industrial R&D Artificial Intelligence Application Seminar hosted by the Korea Industrial Association Forum, that it plans to fully introduce artificial intelligence technology, drastically compressing the development cycle for new semiconductor materials from the traditional 5 years to 2 years, cutting the time by as much as 3 years.
Choi Young-sang, head of the Samsung Advanced Institute of Technology, said at the seminar that with the deep empowerment of AI technology, the number of promising candidate materials screened can be increased by more than 5 times, and the volume of experiments needed to find optimal process conditions can be cut by more than 80%.
The development of semiconductor materials has long faced high costs in time and trial and error. Taking multilayer thin-film structures as an example, a seven-layer structure alone involves more than 100 million material combinations, while the traditional model, limited by researchers' experience and intuition, allows only about 50 experiments per month, resulting in a large amount of failed experiment data not being fully utilized.
The introduction of AI technology has broken through this bottleneck, accelerating material screening and process optimization through deep learning on massive data and efficient simulation. Samsung also plans to extend this AI-driven R&D paradigm to other industrial material fields such as batteries, continuously reducing overall development costs. This strategic transformation marks the acceleration of semiconductor material development into a new stage of intelligence and high efficiency.