IBM Unveils Sub-1nm AI Chip Technology

IBM has officially entered the next-generation semiconductor race, unveiling its sub-1 nanometer (nm) AI chip technology on the 25th (local time). The advanced 0.7nm class ultra-fine chip technology revealed by IBM is a new technique designed to enhance the performance and energy efficiency of semiconductors crucial for artificial intelligence (AI) computation.

IBM's newly revealed technology is based on a design method called 'nanostack'. Unlike conventional planar architectures, this method involves stacking the chip's core component, the transistor, in three dimensions. IBM states this innovation can boost chip performance by up to 50% while reducing power consumption by as much as 70%. Furthermore, this 0.7nm chip can integrate approximately 100 billion transistors within a fingernail-sized area, representing a doubling of integration density compared to the 2nm chip unveiled in 2021.

With this technological announcement, IBM is intensifying its competition for technological leadership against rivals such as TSMC and Intel, aiming for mass production within five years.